I have benchmarked this on an x86_64 NUMA system and see no significant
performance difference on kernbench. Tested on both x86_64 and powerpc.
The way we do file struct accounting is not very suitable for batched
freeing. For scalability reasons, file accounting was
constructor/destructor based. This meant that nr_files was decremented
only when the object was removed from the slab cache. This is susceptible
to slab fragmentation. With RCU based file structure, consequent batched
freeing and a test program like Serge's, we just speed this up and end up
with a very fragmented slab -
llm22:~ # cat /proc/sys/fs/file-nr
587730 0 758844
At the same time, I see only a 2000+ objects in filp cache. The following
patch I fixes this problem.
This patch changes the file counting by removing the filp_count_lock.
Instead we use a separate percpu counter, nr_files, for now and all
accesses to it are through get_nr_files() api. In the sysctl handler for
nr_files, we populate files_stat.nr_files before returning to user.
Counting files as an when they are created and destroyed (as opposed to
inside slab) allows us to correctly count open files with RCU.
Signed-off-by: Dipankar Sarma <dipankar@in.ibm.com>
Cc: "Paul E. McKenney" <paulmck@us.ibm.com>
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Migrate a page with buffers without requiring writeback
This introduces a new address space operation migratepage() that may be used
by a filesystem to implement its own version of page migration.
A version is provided that migrates buffers attached to pages. Some
filesystems (ext2, ext3, xfs) are modified to utilize this feature.
The swapper address space operation are modified so that a regular
migrate_page() will occur for anonymous pages without writeback (migrate_pages
forces every anonymous page to have a swap entry).
Signed-off-by: Mike Kravetz <kravetz@us.ibm.com>
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Here is a series of patches which introduce in total 13 new system calls
which take a file descriptor/filename pair instead of a single file
name. These functions, openat etc, have been discussed on numerous
occasions. They are needed to implement race-free filesystem traversal,
they are necessary to implement a virtual per-thread current working
directory (think multi-threaded backup software), etc.
We have in glibc today implementations of the interfaces which use the
/proc/self/fd magic. But this code is rather expensive. Here are some
results (similar to what Jim Meyering posted before).
The test creates a deep directory hierarchy on a tmpfs filesystem. Then
rm -fr is used to remove all directories. Without syscall support I get
this:
real 0m31.921s
user 0m0.688s
sys 0m31.234s
With syscall support the results are much better:
real 0m20.699s
user 0m0.536s
sys 0m20.149s
The interfaces are for obvious reasons currently not much used. But they'll
be used. coreutils (and Jeff's posixutils) are already using them.
Furthermore, code like ftw/fts in libc (maybe even glob) will also start using
them. I expect a patch to make follow soon. Every program which is walking
the filesystem tree will benefit.
Signed-off-by: Ulrich Drepper <drepper@redhat.com>
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Al Viro <viro@ftp.linux.org.uk>
Acked-by: Ingo Molnar <mingo@elte.hu>
Cc: Michael Kerrisk <mtk-manpages@gmx.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
A Christoph suggested that the /proc/devices file be converted to use the
seq_file interface. This patch does that.
I've obxerved one or two installation that had sufficiently large sans that
they overran the 4k limit on /proc/devices.
Signed-off-by: Neil Horman <nhorman@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Turn noatime and nodiratime into per-mount instead of per-sb flags.
After all the preparations this is a rather trivial patch. The mount code
needs to treat the two options as per-mount instead of per-superblock, and
touch_atime needs to be changed to check the new MNT_ flags in addition to
the MS_ flags that are kept for filesystems that are always
noatime/nodiratime but not user settable anymore. Besides that core code
only nfs needed an update because it's leaving atime updates to the server
and thus sets the S_NOATIME flag on every inode, but needs to know whether
it's a real noatime mount for an getattr optimization.
While we're at it I've killed the IS_NOATIME/IS_NODIRATIME macros that were
only used by touch_atime.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
All callers use touch_atime now which takes a vfsmount and allows us to
implement per-mount noatime.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
To allow various options to work per-mount instead of per-sb we need a
struct vfsmount when updating ctime and mtime. This preparation patch
replaces the inode_update_time routine with a file_update_atime routine so
we can easily get at the vfsmount. (and the file makes more sense in this
context anyway). Also get rid of the unused second argument - we always
want to update the ctime when calling this routine.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Cc: Al Viro <viro@ftp.linux.org.uk>
Cc: Anton Altaparmakov <aia21@cantab.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch converts the superblock-lock semaphore to a mutex, affecting
lock_super()/unlock_super(). Tested on ext3 and XFS.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This patch converts the inode semaphore to a mutex. I have tested it on
XFS and compiled as much as one can consider on an ia64. Anyway your
luck with it might be different.
Modified-by: Ingo Molnar <mingo@elte.hu>
(finished the conversion)
Signed-off-by: Jes Sorensen <jes@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This patch inlines the single user of struct super_block field
s_old_blocksize and removes the field.
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
HDIO_GETGEO is implemented in most block drivers, and all of them have to
duplicate the code to copy the structure to userspace, as well as getting
the start sector. This patch moves that to common code [1] and adds a
->getgeo method to fill out the raw kernel hd_geometry structure. For many
drivers this means ->ioctl can go away now.
[1] the s390 block drivers are odd in this respect. xpram sets ->start
to 4 always which seems more than odd, and the dasd driver shifts
the start offset around, probably because of it's non-standard
sector size.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Cc: Jens Axboe <axboe@suse.de>
Cc: <mike.miller@hp.com>
Cc: Jeff Dike <jdike@addtoit.com>
Cc: Paolo Giarrusso <blaisorblade@yahoo.it>
Cc: Bartlomiej Zolnierkiewicz <B.Zolnierkiewicz@elka.pw.edu.pl>
Cc: Neil Brown <neilb@cse.unsw.edu.au>
Cc: Markus Lidel <Markus.Lidel@shadowconnect.com>
Cc: Russell King <rmk@arm.linux.org.uk>
Cc: David Woodhouse <dwmw2@infradead.org>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: James Bottomley <James.Bottomley@steeleye.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
SUS requires that when truncating a file to the size that it currently
is:
truncate and ftruncate should NOT modify ctime or mtime
O_TRUNC SHOULD modify ctime and mtime.
Currently mtime and ctime are always modified on most local
filesystems (side effect of ->truncate) or never modified (on NFS).
With this patch:
ATTR_CTIME|ATTR_MTIME are sent with ATTR_SIZE precisely when
an update of these times is required whether size changes or not
(via a new argument to do_truncate). This allows NFS to do
the right thing for O_TRUNC.
inode_setattr nolonger forces ATTR_MTIME|ATTR_CTIME when the ATTR_SIZE
sets the size to it's current value. This allows local filesystems
to do the right thing for f?truncate.
Also, the logic in inode_setattr is changed a bit so there are two return
points. One returns the error from vmtruncate if it failed, the other
returns 0 (there can be no other failure).
Finally, if vmtruncate succeeds, and ATTR_SIZE is the only change
requested, we now fall-through and mark_inode_dirty. If a filesystem did
not have a ->truncate function, then vmtruncate will have changed i_size,
without marking the inode as 'dirty', and I think this is wrong.
Signed-off-by: Neil Brown <neilb@suse.de>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Use atomic_inc_not_zero for rcu files instead of special case rcuref.
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: "Paul E. McKenney" <paulmck@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
If the server receives an NLM cancel call and finds no waiting lock to
cancel, then chances are the lock has already been applied, and the client
just hadn't yet processed the NLM granted callback before it sent the
cancel.
The Open Group text, for example, perimts a server to return either success
(LCK_GRANTED) or failure (LCK_DENIED) in this case. But returning an error
seems more helpful; the client may be able to use it to recognize that a
race has occurred and to recover from the race.
So, modify the relevant functions to return an error in this case.
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Here is the patch to implement madvise(MADV_REMOVE) - which frees up a
given range of pages & its associated backing store. Current
implementation supports only shmfs/tmpfs and other filesystems return
-ENOSYS.
"Some app allocates large tmpfs files, then when some task quits and some
client disconnect, some memory can be released. However the only way to
release tmpfs-swap is to MADV_REMOVE". - Andrea Arcangeli
Databases want to use this feature to drop a section of their bufferpool
(shared memory segments) - without writing back to disk/swap space.
This feature is also useful for supporting hot-plug memory on UML.
Concerns raised by Andrew Morton:
- "We have no plan for holepunching! If we _do_ have such a plan (or
might in the future) then what would the API look like? I think
sys_holepunch(fd, start, len), so we should start out with that."
- Using madvise is very weird, because people will ask "why do I need to
mmap my file before I can stick a hole in it?"
- None of the other madvise operations call into the filesystem in this
manner. A broad question is: is this capability an MM operation or a
filesytem operation? truncate, for example, is a filesystem operation
which sometimes has MM side-effects. madvise is an mm operation and with
this patch, it gains FS side-effects, only they're really, really
significant ones."
Comments:
- Andrea suggested the fs operation too but then it's more efficient to
have it as a mm operation with fs side effects, because they don't
immediatly know fd and physical offset of the range. It's possible to
fixup in userland and to use the fs operation but it's more expensive,
the vmas are already in the kernel and we can use them.
Short term plan & Future Direction:
- We seem to need this interface only for shmfs/tmpfs files in the short
term. We have to add hooks into the filesystem for correctness and
completeness. This is what this patch does.
- In the future, plan is to support both fs and mmap apis also. This
also involves (other) filesystem specific functions to be implemented.
- Current patch doesn't support VM_NONLINEAR - which can be addressed in
the future.
Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: Andrea Arcangeli <andrea@suse.de>
Cc: Michael Kerrisk <mtk-manpages@gmx.net>
Cc: Ulrich Drepper <drepper@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
readpage(), prepare_write(), and commit_write() callers are updated to
understand the special return code AOP_TRUNCATED_PAGE in the style of
writepage() and WRITEPAGE_ACTIVATE. AOP_TRUNCATED_PAGE tells the caller that
the callee has unlocked the page and that the operation should be tried again
with a new page. OCFS2 uses this to detect and work around a lock inversion in
its aop methods. There should be no change in behaviour for methods that don't
return AOP_TRUNCATED_PAGE.
WRITEPAGE_ACTIVATE is also prepended with AOP_ for consistency and they are
made enums so that kerneldoc can be used to document their semantics.
Signed-off-by: Zach Brown <zach.brown@oracle.com>
A few more callers of permission() just want to check for a different access
pattern on an already open file. This patch adds a wrapper for permission()
that takes a file in preparation of per-mount read-only support and to clean
up the callers a little. The helper is not intended for new code, everything
without the interface set in stone should use vfs_permission()
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Most permission() calls have a struct nameidata * available. This helper
takes that as an argument and thus makes sure we pass it down for lookup
intents and prepares for per-mount read-only support where we need a struct
vfsmount for checking whether a file is writeable.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
An unbindable mount does not forward or receive propagation. Also
unbindable mount disallows bind mounts. The semantics is as follows.
Bind semantics:
It is invalid to bind mount an unbindable mount.
Move semantics:
It is invalid to move an unbindable mount under shared mount.
Clone-namespace semantics:
If a mount is unbindable in the parent namespace, the corresponding
cloned mount in the child namespace becomes unbindable too. Note:
there is subtle difference, unbindable mounts cannot be bind mounted
but can be cloned during clone-namespace.
Signed-off-by: Ram Pai <linuxram@us.ibm.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
A slave mount always has a master mount from which it receives
mount/umount events. Unlike shared mount the event propagation does not
flow from the slave mount to the master.
Signed-off-by: Ram Pai <linuxram@us.ibm.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
An unmount of a mount creates a umount event on the parent. If the
parent is a shared mount, it gets propagated to all mounts in the peer
group.
Signed-off-by: Ram Pai <linuxram@us.ibm.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Implement handling of MS_BIND in presense of shared mounts (see
Documentation/sharedsubtree.txt in the end of patch series for detailed
description).
Signed-off-by: Ram Pai <linuxram@us.ibm.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This creates shared mounts. A shared mount when bind-mounted to some
mountpoint, propagates mount/umount events to each other. All the
shared mounts that propagate events to each other belong to the same
peer-group.
Signed-off-by: Ram Pai <linuxram@us.ibm.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
A private mount does not forward or receive propagation. This patch
provides user the ability to convert any mount to private.
Signed-off-by: Ram Pai <linuxram@us.ibm.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Fix various warnings in kernel-doc:
Warning(linux-2614-rc4//include/linux/net.h:89): Enum value 'SOCK_DCCP' not described in enum 'sock_type'
usercopy.c: should use !E instead of !I for exported symbols:
Warning(linux-2614-rc4//arch/i386/lib/usercopy.c): no structured comments found
fs.h does not need to use !E since it has no exported symbols:
Warning(linux-2614-rc4//include/linux/fs.h:1182): No description found for parameter 'find_exported_dentry'
Warning(linux-2614-rc4//include/linux/fs.h): no structured comments found
irq/manage.c should use !E for its exported symbols:
Warning(linux-2614-rc4//kernel/irq/manage.c): no structured comments found
macmodes.c should use !E for its exported symbols:
Warning(linux-2614-rc4//drivers/video/macmodes.c): no structured comments found
Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch extends the iattr structure with a file pointer memeber, and adds
an ATTR_FILE validity flag for this member.
This is set if do_truncate() is invoked from ftruncate() or from
do_coredump().
The change is source and binary compatible.
Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Now that RCU applied on 'struct file' seems stable, we can place f_rcuhead
in a memory location that is not anymore used at call_rcu(&f->f_rcuhead,
file_free_rcu) time, to reduce the size of this critical kernel object.
The trick I used is to move f_rcuhead and f_list in an union called f_u
The callers are changed so that f_rcuhead becomes f_u.fu_rcuhead and f_list
becomes f_u.f_list
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
- ->releasepage() annotated (s/int/gfp_t), instances updated
- missing gfp_t in fs/* added
- fixed misannotation from the original sweep caught by bitwise checks:
XFS used __nocast both for gfp_t and for flags used by XFS allocator.
The latter left with unsigned int __nocast; we might want to add a
different type for those but for now let's leave them alone. That,
BTW, is a case when __nocast use had been actively confusing - it had
been used in the same code for two different and similar types, with
no way to catch misuses. Switch of gfp_t to bitwise had caught that
immediately...
One tricky bit is left alone to be dealt with later - mapping->flags is
a mix of gfp_t and error indications. Left alone for now.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
With Nick Piggin <npiggin@suse.de>
Give some things static scope.
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Patch to eliminate struct files_struct.file_lock spinlock on the reader side
and use rcu refcounting rcuref_xxx api for the f_count refcounter. The
updates to the fdtable are done by allocating a new fdtable structure and
setting files->fdt to point to the new structure. The fdtable structure is
protected by RCU thereby allowing lock-free lookup. For fd arrays/sets that
are vmalloced, we use keventd to free them since RCU callbacks can't sleep. A
global list of fdtable to be freed is not scalable, so we use a per-cpu list.
If keventd is already handling the current cpu's work, we use a timer to defer
queueing of that work.
Since the last publication, this patch has been re-written to avoid using
explicit memory barriers and use rcu_assign_pointer(), rcu_dereference()
premitives instead. This required that the fd information is kept in a
separate structure (fdtable) and updated atomically.
Signed-off-by: Dipankar Sarma <dipankar@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Introduce new ll_rw_block() operation SWRITE meaning that block layer should
wait for the buffer lock and write-out afterwards. Hence data in buffers at
the time of call are guaranteed to be submitted to the disk.
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
64 bit architectures all implement their own compatibility sys_open(),
when in fact the difference is simply not forcing the O_LARGEFILE
flag. So use the a common function instead.
Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Cc: <viro@parcelfarce.linux.theplanet.co.uk>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
struct file cleanup: f_maxcount has an unique value (INT_MAX). Just use
the hard-wired value.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This bug could cause oopses and page state corruption, because ncpfs
used the generic page-cache symlink handlign functions. But those
functions only work if the page cache is guaranteed to be "stable", ie a
page that was installed when the symlink walk was started has to still
be installed in the page cache at the end of the walk.
We could have fixed ncpfs to not use the generic helper routines, but it
is in many ways much cleaner to instead improve on the symlink walking
helper routines so that they don't require that absolute stability.
We do this by allowing "follow_link()" to return a error-pointer as a
cookie, which is fed back to the cleanup "put_link()" routine. This
also simplifies NFS symlink handling.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
I believe that there is a problem with the handling of POSIX locks, which
the attached patch should address.
The problem appears to be a race between fcntl(2) and close(2). A
multithreaded application could close a file descriptor at the same time as
it is trying to acquire a lock using the same file descriptor. I would
suggest that that multithreaded application is not providing the proper
synchronization for itself, but the OS should still behave correctly.
SUS3 (Single UNIX Specification Version 3, read: POSIX) indicates that when
a file descriptor is closed, that all POSIX locks on the file, owned by the
process which closed the file descriptor, should be released.
The trick here is when those locks are released. The current code releases
all locks which exist when close is processing, but any locks in progress
are handled when the last reference to the open file is released.
There are three cases to consider.
One is the simple case, a multithreaded (mt) process has a file open and
races to close it and acquire a lock on it. In this case, the close will
release one reference to the open file and when the fcntl is done, it will
release the other reference. For this situation, no locks should exist on
the file when both the close and fcntl operations are done. The current
system will handle this case because the last reference to the open file is
being released.
The second case is when the mt process has dup(2)'d the file descriptor.
The close will release one reference to the file and the fcntl, when done,
will release another, but there will still be at least one more reference
to the open file. One could argue that the existence of a lock on the file
after the close has completed is okay, because it was acquired after the
close operation and there is still a way for the application to release the
lock on the file, using an existing file descriptor.
The third case is when the mt process has forked, after opening the file
and either before or after becoming an mt process. In this case, each
process would hold a reference to the open file. For each process, this
degenerates to first case above. However, the lock continues to exist
until both processes have released their references to the open file. This
lock could block other lock requests.
The changes to release the lock when the last reference to the open file
aren't quite right because they would allow the lock to exist as long as
there was a reference to the open file. This is too long.
The new proposed solution is to add support in the fcntl code path to
detect a race with close and then to release the lock which was just
acquired when such as race is detected. This causes locks to be released
in a timely fashion and for the system to conform to the POSIX semantic
specification.
This was tested by instrumenting a kernel to detect the handling locks and
then running a program which generates case #3 above. A dangling lock
could be reliably generated. When the changes to detect the close/fcntl
race were added, a dangling lock could no longer be generated.
Cc: Matthew Wilcox <willy@debian.org>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Something has changed in the core kernel such that we now get concurrent
inode write outs, one e.g via pdflush and one via sys_sync or whatever.
This causes a nasty deadlock in ntfs. The only clean solution
unfortunately requires a minor vfs api extension.
First the deadlock analysis:
Prerequisive knowledge: NTFS has a file $MFT (inode 0) loaded at mount
time. The NTFS driver uses the page cache for storing the file contents as
usual. More interestingly this file contains the table of on-disk inodes
as a sequence of MFT_RECORDs. Thus NTFS driver accesses the on-disk inodes
by accessing the MFT_RECORDs in the page cache pages of the loaded inode
$MFT.
The situation: VFS inode X on a mounted ntfs volume is dirty. For same
inode X, the ntfs_inode is dirty and thus corresponding on-disk inode,
which is as explained above in a dirty PAGE_CACHE_PAGE belonging to the
table of inodes ($MFT, inode 0).
What happens:
Process 1: sys_sync()/umount()/whatever... calls __sync_single_inode() for
$MFT -> do_writepages() -> write_page for the dirty page containing the
on-disk inode X, the page is now locked -> ntfs_write_mst_block() which
clears PageUptodate() on the page to prevent anyone else getting hold of it
whilst it does the write out (this is necessary as the on-disk inode needs
"fixups" applied before the write to disk which are removed again after the
write and PageUptodate is then set again). It then analyses the page
looking for dirty on-disk inodes and when it finds one it calls
ntfs_may_write_mft_record() to see if it is safe to write this on-disk
inode. This then calls ilookup5() to check if the corresponding VFS inode
is in icache(). This in turn calls ifind() which waits on the inode lock
via wait_on_inode whilst holding the global inode_lock.
Process 2: pdflush results in a call to __sync_single_inode for the same
VFS inode X on the ntfs volume. This locks the inode (I_LOCK) then calls
write-inode -> ntfs_write_inode -> map_mft_record() -> read_cache_page() of
the page (in page cache of table of inodes $MFT, inode 0) containing the
on-disk inode. This page has PageUptodate() clear because of Process 1
(see above) so read_cache_page() blocks when tries to take the page lock
for the page so it can call ntfs_read_page().
Thus Process 1 is holding the page lock on the page containing the on-disk
inode X and it is waiting on the inode X to be unlocked in ifind() so it
can write the page out and then unlock the page.
And Process 2 is holding the inode lock on inode X and is waiting for the
page to be unlocked so it can call ntfs_readpage() or discover that
Process 1 set PageUptodate() again and use the page.
Thus we have a deadlock due to ifind() waiting on the inode lock.
The only sensible solution: NTFS does not care whether the VFS inode is
locked or not when it calls ilookup5() (it doesn't use the VFS inode at
all, it just uses it to find the corresponding ntfs_inode which is of
course attached to the VFS inode (both are one single struct); and it uses
the ntfs_inode which is subject to its own locking so I_LOCK is irrelevant)
hence we want a modified ilookup5_nowait() which is the same as ilookup5()
but it does not wait on the inode lock.
Without such functionality I would have to keep my own ntfs_inode cache in
the NTFS driver just so I can find ntfs_inodes independent of their VFS
inodes which would be slow, memory and cpu cycle wasting, and incredibly
stupid given the icache already exists in the VFS.
Below is a patch that does the ilookup5_nowait() implementation in
fs/inode.c and exports it.
ilookup5_nowait.diff:
Introduce ilookup5_nowait() which is basically the same as ilookup5() but
it does not wait on the inode's lock (i.e. it omits the wait_on_inode()
done in ifind()).
This is needed to avoid a nasty deadlock in NTFS.
Signed-off-by: Anton Altaparmakov <aia21@cantab.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
inotify is intended to correct the deficiencies of dnotify, particularly
its inability to scale and its terrible user interface:
* dnotify requires the opening of one fd per each directory
that you intend to watch. This quickly results in too many
open files and pins removable media, preventing unmount.
* dnotify is directory-based. You only learn about changes to
directories. Sure, a change to a file in a directory affects
the directory, but you are then forced to keep a cache of
stat structures.
* dnotify's interface to user-space is awful. Signals?
inotify provides a more usable, simple, powerful solution to file change
notification:
* inotify's interface is a system call that returns a fd, not SIGIO.
You get a single fd, which is select()-able.
* inotify has an event that says "the filesystem that the item
you were watching is on was unmounted."
* inotify can watch directories or files.
Inotify is currently used by Beagle (a desktop search infrastructure),
Gamin (a FAM replacement), and other projects.
See Documentation/filesystems/inotify.txt.
Signed-off-by: Robert Love <rml@novell.com>
Cc: John McCutchan <ttb@tentacle.dhs.org>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
OCFS2 wants to mark an inode which has been orphaned by another node so
that during final iput it takes the correct path through the VFS and can
pass through the OCFS2 delete_inode callback. Since i_nlink can get out of
date with other nodes, the best way I see to accomplish this is by clearing
i_nlink on those inodes at drop_inode time. Other than this small amount
of work, nothing different needs to happen, so I think it would be cleanest
to be able to just call generic_drop_inode at the end of the OCFS2
drop_inode callback.
Signed-off-by: Mark Fasheh <mark.fasheh@oracle.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This updates the CFQ io scheduler to the new time sliced design (cfq
v3). It provides full process fairness, while giving excellent
aggregate system throughput even for many competing processes. It
supports io priorities, either inherited from the cpu nice value or set
directly with the ioprio_get/set syscalls. The latter closely mimic
set/getpriority.
This import is based on my latest from -mm.
Signed-off-by: Jens Axboe <axboe@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch reworks filemap_xip.c with the goal to reduce code duplication
from mm/filemap.c. It applies agains 2.6.12-rc6-mm1. Instead of
implementing the aio functions, this one implements the synchronous
read/write functions only. For readv and writev, the generic fallback is
used. For aio, we rely on the application doing the fallback. Since our
"synchronous" function does memcpy immediately anyway, there is no
performance difference between using the fallbacks or implementing each
operation.
Signed-off-by: Carsten Otte <cotte@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
These are the ext2 related parts. Ext2 now uses the xip_* file operations
along with the get_xip_page aop when mounted with -o xip.
Signed-off-by: Carsten Otte <cotte@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
- generic_file* file operations do no longer have a xip/non-xip split
- filemap_xip.c implements a new set of fops that require get_xip_page
aop to work proper. all new fops are exported GPL-only (don't like to
see whatever code use those except GPL modules)
- __xip_unmap now uses page_check_address, which is no longer static
in rmap.c, and defined in linux/rmap.h
- mm/filemap.h is now much more clean, plainly having just Linus'
inline funcs moved here from filemap.c
- fix includes in filemap_xip to make it build cleanly on i386
Signed-off-by: Carsten Otte <cotte@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This is the block device related part. The block device operation
direct_access now has a struct block_device as first parameter.
Signed-off-by: Carsten Otte <cotte@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
XFS will have to look at iocb->private to fix aio+dio. No other filesystem
is using the blockdev_direct_IO* end_io callback.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The following patch removes the f_error field and all checks of f_error.
Trond said:
f_error was introduced for NFS, and made sense when we were guaranteed
always to have a file pointer around when write errors occurred. Since
then, we have (for various reasons) had to introduce the nfs_open_context in
order to track the file read/write state, and it made sense to move our
f_error tracking there too.
Signed-off-by: Christoph Lameter <christoph@lameter.com>
Acked-by: Trond Myklebust <trond.myklebust@fys.uio.no>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch allows block device drivers to convert their ioctl functions to
unlocked_ioctl() like character devices and other subsystems. All
functions that were called with the BKL held before are still used that
way, but I would not be surprised if it could be removed from the ioctl
functions in drivers/block/ioctl.c themselves.
As a side note, I found that compat_blkdev_ioctl() acquires the BKL as
well, which looks like a bug. I have checked that every user of
disk->fops->compat_ioctl() in the current git tree gets the BKL itself, so
it could easily be removed from compat_blkdev_ioctl().
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
