Pull vfs updates from Al Viro:
"The first vfs pile, with deep apologies for being very late in this
window.
Assorted cleanups and fixes, plus a large preparatory part of iov_iter
work. There's a lot more of that, but it'll probably go into the next
merge window - it *does* shape up nicely, removes a lot of
boilerplate, gets rid of locking inconsistencie between aio_write and
splice_write and I hope to get Kent's direct-io rewrite merged into
the same queue, but some of the stuff after this point is having
(mostly trivial) conflicts with the things already merged into
mainline and with some I want more testing.
This one passes LTP and xfstests without regressions, in addition to
usual beating. BTW, readahead02 in ltp syscalls testsuite has started
giving failures since "mm/readahead.c: fix readahead failure for
memoryless NUMA nodes and limit readahead pages" - might be a false
positive, might be a real regression..."
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (63 commits)
missing bits of "splice: fix racy pipe->buffers uses"
cifs: fix the race in cifs_writev()
ceph_sync_{,direct_}write: fix an oops on ceph_osdc_new_request() failure
kill generic_file_buffered_write()
ocfs2_file_aio_write(): switch to generic_perform_write()
ceph_aio_write(): switch to generic_perform_write()
xfs_file_buffered_aio_write(): switch to generic_perform_write()
export generic_perform_write(), start getting rid of generic_file_buffer_write()
generic_file_direct_write(): get rid of ppos argument
btrfs_file_aio_write(): get rid of ppos
kill the 5th argument of generic_file_buffered_write()
kill the 4th argument of __generic_file_aio_write()
lustre: don't open-code kernel_recvmsg()
ocfs2: don't open-code kernel_recvmsg()
drbd: don't open-code kernel_recvmsg()
constify blk_rq_map_user_iov() and friends
lustre: switch to kernel_sendmsg()
ocfs2: don't open-code kernel_sendmsg()
take iov_iter stuff to mm/iov_iter.c
process_vm_access: tidy up a bit
...
Pull file locking updates from Jeff Layton:
"Highlights:
- maintainership change for fs/locks.c. Willy's not interested in
maintaining it these days, and is OK with Bruce and I taking it.
- fix for open vs setlease race that Al ID'ed
- cleanup and consolidation of file locking code
- eliminate unneeded BUG() call
- merge of file-private lock implementation"
* 'locks-3.15' of git://git.samba.org/jlayton/linux:
locks: make locks_mandatory_area check for file-private locks
locks: fix locks_mandatory_locked to respect file-private locks
locks: require that flock->l_pid be set to 0 for file-private locks
locks: add new fcntl cmd values for handling file private locks
locks: skip deadlock detection on FL_FILE_PVT locks
locks: pass the cmd value to fcntl_getlk/getlk64
locks: report l_pid as -1 for FL_FILE_PVT locks
locks: make /proc/locks show IS_FILE_PVT locks as type "FLPVT"
locks: rename locks_remove_flock to locks_remove_file
locks: consolidate checks for compatible filp->f_mode values in setlk handlers
locks: fix posix lock range overflow handling
locks: eliminate BUG() call when there's an unexpected lock on file close
locks: add __acquires and __releases annotations to locks_start and locks_stop
locks: remove "inline" qualifier from fl_link manipulation functions
locks: clean up comment typo
locks: close potential race between setlease and open
MAINTAINERS: update entry for fs/locks.c
the only thing it's doing these days is calculation of
upper limit for fs.nr_open sysctl and that can be done
statically
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
new flag in ->f_mode - FMODE_WRITER. Set by do_dentry_open() in case
when it has grabbed write access, checked by __fput() to decide whether
it wants to drop the sucker. Allows to stop bothering with mnt_clone_write()
in alloc_file(), along with fewer special_file() checks.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
it's pointless and actually leads to wrong behaviour in at least one
moderately convoluted case (pipe(), close one end, try to get to
another via /proc/*/fd and run into ETXTBUSY).
Cc: stable@vger.kernel.org
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
This function currently removes leases in addition to flock locks and in
a later patch we'll have it deal with file-private locks too. Rename it
to locks_remove_file to indicate that it removes locks that are
associated with a particular struct file, and not just flock locks.
Acked-by: J. Bruce Fields <bfields@fieldses.org>
Signed-off-by: Jeff Layton <jlayton@redhat.com>
Our write() system call has always been atomic in the sense that you get
the expected thread-safe contiguous write, but we haven't actually
guaranteed that concurrent writes are serialized wrt f_pos accesses, so
threads (or processes) that share a file descriptor and use "write()"
concurrently would quite likely overwrite each others data.
This violates POSIX.1-2008/SUSv4 Section XSI 2.9.7 that says:
"2.9.7 Thread Interactions with Regular File Operations
All of the following functions shall be atomic with respect to each
other in the effects specified in POSIX.1-2008 when they operate on
regular files or symbolic links: [...]"
and one of the effects is the file position update.
This unprotected file position behavior is not new behavior, and nobody
has ever cared. Until now. Yongzhi Pan reported unexpected behavior to
Michael Kerrisk that was due to this.
This resolves the issue with a f_pos-specific lock that is taken by
read/write/lseek on file descriptors that may be shared across threads
or processes.
Reported-by: Yongzhi Pan <panyongzhi@gmail.com>
Reported-by: Michael Kerrisk <mtk.manpages@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Pull vfs updates from Al Viro:
"All kinds of stuff this time around; some more notable parts:
- RCU'd vfsmounts handling
- new primitives for coredump handling
- files_lock is gone
- Bruce's delegations handling series
- exportfs fixes
plus misc stuff all over the place"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (101 commits)
ecryptfs: ->f_op is never NULL
locks: break delegations on any attribute modification
locks: break delegations on link
locks: break delegations on rename
locks: helper functions for delegation breaking
locks: break delegations on unlink
namei: minor vfs_unlink cleanup
locks: implement delegations
locks: introduce new FL_DELEG lock flag
vfs: take i_mutex on renamed file
vfs: rename I_MUTEX_QUOTA now that it's not used for quotas
vfs: don't use PARENT/CHILD lock classes for non-directories
vfs: pull ext4's double-i_mutex-locking into common code
exportfs: fix quadratic behavior in filehandle lookup
exportfs: better variable name
exportfs: move most of reconnect_path to helper function
exportfs: eliminate unused "noprogress" counter
exportfs: stop retrying once we race with rename/remove
exportfs: clear DISCONNECTED on all parents sooner
exportfs: more detailed comment for path_reconnect
...
The only thing we need it for is alt-sysrq-r (emergency remount r/o)
and these days we can do just as well without going through the
list of files.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Background: nfsd v[23] had throughput regression since delayed fput
went in; every read or write ends up doing fput() and we get a pair
of extra context switches out of that (plus quite a bit of work
in queue_work itselfi, apparently). Use of schedule_delayed_work()
gives it a chance to accumulate a bit before we do __fput() on all
of them. I'm not too happy about that solution, but... on at least
one real-world setup it reverts about 10% throughput loss we got from
switch to delayed fput.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
fput() and delayed_fput() can use llist and avoid the locking.
This is unlikely path, it is not that this change can improve
the performance, but this way the code looks simpler.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Suggested-by: Andrew Morton <akpm@linux-foundation.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Andrey Vagin <avagin@openvz.org>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Huang Ying <ying.huang@intel.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
A missed update to "fput: task_work_add() can fail if the caller has
passed exit_task_work()".
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Andrey Vagin <avagin@openvz.org>
Cc: David Howells <dhowells@redhat.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Replace a bunch of file->dentry->d_inode refs with file_inode().
In __fput(), use file->f_inode instead so as not to be affected by any tricks
that file_inode() might grow.
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
fput() assumes that it can't be called after exit_task_work() but
this is not true, for example free_ipc_ns()->shm_destroy() can do
this. In this case fput() silently leaks the file.
Change it to fallback to delayed_fput_work if task_work_add() fails.
The patch looks complicated but it is not, it changes the code from
if (PF_KTHREAD) {
schedule_work(...);
return;
}
task_work_add(...)
to
if (!PF_KTHREAD) {
if (!task_work_add(...))
return;
/* fallback */
}
schedule_work(...);
As for shm_destroy() in particular, we could make another fix but I
think this change makes sense anyway. There could be another similar
user, it is not safe to assume that task_work_add() can't fail.
Reported-by: Andrey Vagin <avagin@openvz.org>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Allocating a file structure in function get_empty_filp() might fail because
of several reasons:
- not enough memory for file structures
- operation is not allowed
- user is over its limit
Currently the function returns NULL in all cases and we loose the exact
reason of the error. All callers of get_empty_filp() assume that the function
can fail with ENFILE only.
Return error through pointer. Change all callers to preserve this error code.
[AV: cleaned up a bit, carved the get_empty_filp() part out into a separate commit
(things remaining here deal with alloc_file()), removed pipe(2) behaviour change]
Signed-off-by: Anatol Pomozov <anatol.pomozov@gmail.com>
Reviewed-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
File descriptors (even those for writing) do not hold freeze protection.
Thus mark_files_ro() must call __mnt_drop_write() to only drop protection
against remount read-only. Calling mnt_drop_write_file() as we do now
results in:
[ BUG: bad unlock balance detected! ]
3.7.0-rc6-00028-g88e75b6 #101 Not tainted
-------------------------------------
kworker/1:2/79 is trying to release lock (sb_writers) at:
[<ffffffff811b33b4>] mnt_drop_write+0x24/0x30
but there are no more locks to release!
Reported-by: Zdenek Kabelac <zkabelac@redhat.com>
CC: stable@vger.kernel.org
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
When the lglock doesn't need to be exported we can use
DEFINE_STATIC_LGLOCK().
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Pull security subsystem updates from James Morris:
"Highlights:
- Integrity: add local fs integrity verification to detect offline
attacks
- Integrity: add digital signature verification
- Simple stacking of Yama with other LSMs (per LSS discussions)
- IBM vTPM support on ppc64
- Add new driver for Infineon I2C TIS TPM
- Smack: add rule revocation for subject labels"
Fixed conflicts with the user namespace support in kernel/auditsc.c and
security/integrity/ima/ima_policy.c.
* 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security: (39 commits)
Documentation: Update git repository URL for Smack userland tools
ima: change flags container data type
Smack: setprocattr memory leak fix
Smack: implement revoking all rules for a subject label
Smack: remove task_wait() hook.
ima: audit log hashes
ima: generic IMA action flag handling
ima: rename ima_must_appraise_or_measure
audit: export audit_log_task_info
tpm: fix tpm_acpi sparse warning on different address spaces
samples/seccomp: fix 31 bit build on s390
ima: digital signature verification support
ima: add support for different security.ima data types
ima: add ima_inode_setxattr/removexattr function and calls
ima: add inode_post_setattr call
ima: replace iint spinblock with rwlock/read_lock
ima: allocating iint improvements
ima: add appraise action keywords and default rules
ima: integrity appraisal extension
vfs: move ima_file_free before releasing the file
...
ima_file_free(), called on __fput(), currently flags files that have
changed, so that the file is re-measured. For appraising a files's
integrity, the file's hash must be re-calculated and stored in the
'security.ima' xattr to reflect any changes.
This patch moves the ima_file_free() call to before releasing the file
in preparation of ima-appraisal measuring the file and updating the
'security.ima' xattr.
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Acked-by: Serge Hallyn <serge.hallyn@ubuntu.com>
Acked-by: Dmitry Kasatkin <dmitry.kasatkin@intel.com>
Most of places where we want freeze protection coincides with the places where
we also have remount-ro protection. So make mnt_want_write() and
mnt_drop_write() (and their _file alternative) prevent freezing as well.
For the few cases that are really interested only in remount-ro protection
provide new function variants.
BugLink: https://bugs.launchpad.net/bugs/897421
Tested-by: Kamal Mostafa <kamal@canonical.com>
Tested-by: Peter M. Petrakis <peter.petrakis@canonical.com>
Tested-by: Dann Frazier <dann.frazier@canonical.com>
Tested-by: Massimo Morana <massimo.morana@canonical.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
... and schedule_work() for interrupt/kernel_thread callers
(and yes, now it *is* OK to call from interrupt).
We are guaranteed that __fput() will be done before we return
to userland (or exit). Note that for fput() from a kernel
thread we get an async behaviour; it's almost always OK, but
sometimes you might need to have __fput() completed before
you do anything else. There are two mechanisms for that -
a general barrier (flush_delayed_fput()) and explicit
__fput_sync(). Both should be used with care (as was the
case for fput() from kernel threads all along). See comments
in fs/file_table.c for details.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
lglocks and brlocks are currently generated with some complicated macros
in lglock.h. But there's no reason to not just use common utility
functions and put all the data into a common data structure.
In preparation, this patch changes the API to look more like normal
function calls with pointers, not magic macros.
The patch is rather large because I move over all users in one go to keep
it bisectable. This impacts the VFS somewhat in terms of lines changed.
But no actual behaviour change.
[akpm@linux-foundation.org: checkpatch fixes]
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
lglocks and brlocks are currently generated with some complicated macros
in lglock.h. But there's no reason to not just use common utility
functions and put all the data into a common data structure.
Since there are at least two users it makes sense to share this code in a
library. This is also easier maintainable than a macro forest.
This will also make it later possible to dynamically allocate lglocks and
also use them in modules (this would both still need some additional, but
now straightforward, code)
[akpm@linux-foundation.org: checkpatch fixes]
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
If there are any inodes on the super block that have been unlinked
(i_nlink == 0) but have not yet been deleted then prevent the
remounting the super block read-only.
Reported-by: Toshiyuki Okajima <toshi.okajima@jp.fujitsu.com>
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Tested-by: Toshiyuki Okajima <toshi.okajima@jp.fujitsu.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
This allows us to move duplicated code in <asm/atomic.h>
(atomic_inc_not_zero() for now) to <linux/atomic.h>
Signed-off-by: Arun Sharma <asharma@fb.com>
Reviewed-by: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: David Miller <davem@davemloft.net>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Acked-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/security-testing-2.6: (33 commits)
AppArmor: kill unused macros in lsm.c
AppArmor: cleanup generated files correctly
KEYS: Add an iovec version of KEYCTL_INSTANTIATE
KEYS: Add a new keyctl op to reject a key with a specified error code
KEYS: Add a key type op to permit the key description to be vetted
KEYS: Add an RCU payload dereference macro
AppArmor: Cleanup make file to remove cruft and make it easier to read
SELinux: implement the new sb_remount LSM hook
LSM: Pass -o remount options to the LSM
SELinux: Compute SID for the newly created socket
SELinux: Socket retains creator role and MLS attribute
SELinux: Auto-generate security_is_socket_class
TOMOYO: Fix memory leak upon file open.
Revert "selinux: simplify ioctl checking"
selinux: drop unused packet flow permissions
selinux: Fix packet forwarding checks on postrouting
selinux: Fix wrong checks for selinux_policycap_netpeer
selinux: Fix check for xfrm selinux context algorithm
ima: remove unnecessary call to ima_must_measure
IMA: remove IMA imbalance checking
...
Just need to make sure that AF_UNIX garbage collector won't
confuse O_PATHed socket on filesystem for real AF_UNIX opened
socket.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
New flag for open(2) - O_PATH. Semantics:
* pathname is resolved, but the file itself is _NOT_ opened
as far as filesystem is concerned.
* almost all operations on the resulting descriptors shall
fail with -EBADF. Exceptions are:
1) operations on descriptors themselves (i.e.
close(), dup(), dup2(), dup3(), fcntl(fd, F_DUPFD),
fcntl(fd, F_DUPFD_CLOEXEC, ...), fcntl(fd, F_GETFD),
fcntl(fd, F_SETFD, ...))
2) fcntl(fd, F_GETFL), for a common non-destructive way to
check if descriptor is open
3) "dfd" arguments of ...at(2) syscalls, i.e. the starting
points of pathname resolution
* closing such descriptor does *NOT* affect dnotify or
posix locks.
* permissions are checked as usual along the way to file;
no permission checks are applied to the file itself. Of course,
giving such thing to syscall will result in permission checks (at
the moment it means checking that starting point of ....at() is
a directory and caller has exec permissions on it).
fget() and fget_light() return NULL on such descriptors; use of
fget_raw() and fget_raw_light() is needed to get them. That protects
existing code from dealing with those things.
There are two things still missing (they come in the next commits):
one is handling of symlinks (right now we refuse to open them that
way; see the next commit for semantics related to those) and another
is descriptor passing via SCM_RIGHTS datagrams.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
ima_counts_get() updated the readcount and invalidated the PCR,
as necessary. Only update the i_readcount in the VFS layer.
Move the PCR invalidation checks to ima_file_check(), where it
belongs.
Maintaining the i_readcount in the VFS layer, will allow other
subsystems to use i_readcount.
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Acked-by: Eric Paris <eparis@redhat.com>
In get_empty_filp() since 2.6.29, file_free(f) is called with f->f_cred == NULL
when security_file_alloc() returned an error. As a result, kernel will panic()
due to put_cred(NULL) call within RCU callback.
Fix this bug by assigning f->f_cred before calling security_file_alloc().
Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
There's an unlikely() in fget_light() that assumes the file ref count
will be 1. Running the annotate branch profiler on a desktop that is
performing daily tasks (running firefox, evolution, xchat and is also part
of a distcc farm), it shows that the ref count is not 1 that often.
correct incorrect % Function File Line
------- --------- - -------- ---- ----
1035099358 6209599193 85 fget_light file_table.c 315
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Robin Holt tried to boot a 16TB system and found af_unix was overflowing
a 32bit value :
<quote>
We were seeing a failure which prevented boot. The kernel was incapable
of creating either a named pipe or unix domain socket. This comes down
to a common kernel function called unix_create1() which does:
atomic_inc(&unix_nr_socks);
if (atomic_read(&unix_nr_socks) > 2 * get_max_files())
goto out;
The function get_max_files() is a simple return of files_stat.max_files.
files_stat.max_files is a signed integer and is computed in
fs/file_table.c's files_init().
n = (mempages * (PAGE_SIZE / 1024)) / 10;
files_stat.max_files = n;
In our case, mempages (total_ram_pages) is approx 3,758,096,384
(0xe0000000). That leaves max_files at approximately 1,503,238,553.
This causes 2 * get_max_files() to integer overflow.
</quote>
Fix is to let /proc/sys/fs/file-nr & /proc/sys/fs/file-max use long
integers, and change af_unix to use an atomic_long_t instead of atomic_t.
get_max_files() is changed to return an unsigned long. get_nr_files() is
changed to return a long.
unix_nr_socks is changed from atomic_t to atomic_long_t, while not
strictly needed to address Robin problem.
Before patch (on a 64bit kernel) :
# echo 2147483648 >/proc/sys/fs/file-max
# cat /proc/sys/fs/file-max
-18446744071562067968
After patch:
# echo 2147483648 >/proc/sys/fs/file-max
# cat /proc/sys/fs/file-max
2147483648
# cat /proc/sys/fs/file-nr
704 0 2147483648
Reported-by: Robin Holt <holt@sgi.com>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Acked-by: David Miller <davem@davemloft.net>
Reviewed-by: Robin Holt <holt@sgi.com>
Tested-by: Robin Holt <holt@sgi.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: Linus Torvalds <torvalds@linux-foundation.org>
fs: scale files_lock
Improve scalability of files_lock by adding per-cpu, per-sb files lists,
protected with an lglock. The lglock provides fast access to the per-cpu lists
to add and remove files. It also provides a snapshot of all the per-cpu lists
(although this is very slow).
One difficulty with this approach is that a file can be removed from the list
by another CPU. We must track which per-cpu list the file is on with a new
variale in the file struct (packed into a hole on 64-bit archs). Scalability
could suffer if files are frequently removed from different cpu's list.
However loads with frequent removal of files imply short interval between
adding and removing the files, and the scheduler attempts to avoid moving
processes too far away. Also, even in the case of cross-CPU removal, the
hardware has much more opportunity to parallelise cacheline transfers with N
cachelines than with 1.
A worst-case test of 1 CPU allocating files subsequently being freed by N CPUs
degenerates to contending on a single lock, which is no worse than before. When
more than one CPU are allocating files, even if they are always freed by
different CPUs, there will be more parallelism than the single-lock case.
Testing results:
On a 2 socket, 8 core opteron, I measure the number of times the lock is taken
to remove the file, the number of times it is removed by the same CPU that
added it, and the number of times it is removed by the same node that added it.
Booting: locks= 25049 cpu-hits= 23174 (92.5%) node-hits= 23945 (95.6%)
kbuild -j16 locks=2281913 cpu-hits=2208126 (96.8%) node-hits=2252674 (98.7%)
dbench 64 locks=4306582 cpu-hits=4287247 (99.6%) node-hits=4299527 (99.8%)
So a file is removed from the same CPU it was added by over 90% of the time.
It remains within the same node 95% of the time.
Tim Chen ran some numbers for a 64 thread Nehalem system performing a compile.
throughput
2.6.34-rc2 24.5
+patch 24.9
us sys idle IO wait (in %)
2.6.34-rc2 51.25 28.25 17.25 3.25
+patch 53.75 18.5 19 8.75
So significantly less CPU time spent in kernel code, higher idle time and
slightly higher throughput.
Single threaded performance difference was within the noise of microbenchmarks.
That is not to say penalty does not exist, the code is larger and more memory
accesses required so it will be slightly slower.
Cc: linux-kernel@vger.kernel.org
Cc: Tim Chen <tim.c.chen@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
fs: cleanup files_lock locking
Lock tty_files with a new spinlock, tty_files_lock; provide helpers to
manipulate the per-sb files list; unexport the files_lock spinlock.
Cc: linux-kernel@vger.kernel.org
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Acked-by: Andi Kleen <ak@linux.intel.com>
Acked-by: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
This reverts commit 3bcf3860a4 (and the
accompanying commit c1e5c95402 "vfs/fsnotify: fsnotify_close can delay
the final work in fput" that was a horribly ugly hack to make it work at
all).
The 'struct file' approach not only causes that disgusting hack, it
somehow breaks pulseaudio, probably due to some other subtlety with
f_count handling.
Fix up various conflicts due to later fsnotify work.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Improve the description of fget_light(), which is currently incorrect
about needing a prior refcnt (judging by the way it is actually used).
Signed-off-by: Tony Battersby <tonyb@cybernetics.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>
