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: Eric Dumazet <eric.dumazet@gmail.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al 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: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Instead of splitting refcount between (per-cpu) mnt_count
and (SMP-only) mnt_longrefs, make all references contribute
to mnt_count again and keep track of how many are longterm
ones.
Accounting rules for longterm count:
* 1 for each fs_struct.root.mnt
* 1 for each fs_struct.pwd.mnt
* 1 for having non-NULL ->mnt_ns
* decrement to 0 happens only under vfsmount lock exclusive
That allows nice common case for mntput() - since we can't drop the
final reference until after mnt_longterm has reached 0 due to the rules
above, mntput() can grab vfsmount lock shared and check mnt_longterm.
If it turns out to be non-zero (which is the common case), we know
that this is not the final mntput() and can just blindly decrement
percpu mnt_count. Otherwise we grab vfsmount lock exclusive and
do usual decrement-and-check of percpu mnt_count.
For fs_struct.c we have mnt_make_longterm() and mnt_make_shortterm();
namespace.c uses the latter in places where we don't already hold
vfsmount lock exclusive and opencodes a few remaining spots where
we need to manipulate mnt_longterm.
Note that we mostly revert the code outside of fs/namespace.c back
to what we used to have; in particular, normal code doesn't need
to care about two kinds of references, etc. And we get to keep
the optimization Nick's variant had bought us...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6: (41 commits)
fs: add documentation on fallocate hole punching
Gfs2: fail if we try to use hole punch
Btrfs: fail if we try to use hole punch
Ext4: fail if we try to use hole punch
Ocfs2: handle hole punching via fallocate properly
XFS: handle hole punching via fallocate properly
fs: add hole punching to fallocate
vfs: pass struct file to do_truncate on O_TRUNC opens (try #2)
fix signedness mess in rw_verify_area() on 64bit architectures
fs: fix kernel-doc for dcache::prepend_path
fs: fix kernel-doc for dcache::d_validate
sanitize ecryptfs ->mount()
switch afs
move internal-only parts of ncpfs headers to fs/ncpfs
switch ncpfs
switch 9p
pass default dentry_operations to mount_pseudo()
switch hostfs
switch affs
switch configfs
...
The problem that this patch aims to fix is vfsmount refcounting scalability.
We need to take a reference on the vfsmount for every successful path lookup,
which often go to the same mount point.
The fundamental difficulty is that a "simple" reference count can never be made
scalable, because any time a reference is dropped, we must check whether that
was the last reference. To do that requires communication with all other CPUs
that may have taken a reference count.
We can make refcounts more scalable in a couple of ways, involving keeping
distributed counters, and checking for the global-zero condition less
frequently.
- check the global sum once every interval (this will delay zero detection
for some interval, so it's probably a showstopper for vfsmounts).
- keep a local count and only taking the global sum when local reaches 0 (this
is difficult for vfsmounts, because we can't hold preempt off for the life of
a reference, so a counter would need to be per-thread or tied strongly to a
particular CPU which requires more locking).
- keep a local difference of increments and decrements, which allows us to sum
the total difference and hence find the refcount when summing all CPUs. Then,
keep a single integer "long" refcount for slow and long lasting references,
and only take the global sum of local counters when the long refcount is 0.
This last scheme is what I implemented here. Attached mounts and process root
and working directory references are "long" references, and everything else is
a short reference.
This allows scalable vfsmount references during path walking over mounted
subtrees and unattached (lazy umounted) mounts with processes still running
in them.
This results in one fewer atomic op in the fastpath: mntget is now just a
per-CPU inc, rather than an atomic inc; and mntput just requires a spinlock
and non-atomic decrement in the common case. However code is otherwise bigger
and heavier, so single threaded performance is basically a wash.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
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>
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>
Instead of always assigning an increasing inode number in new_inode
move the call to assign it into those callers that actually need it.
For now callers that need it is estimated conservatively, that is
the call is added to all filesystems that do not assign an i_ino
by themselves. For a few more filesystems we can avoid assigning
any inode number given that they aren't user visible, and for others
it could be done lazily when an inode number is actually needed,
but that's left for later patches.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
anon_inode_mkinode() sets inode->i_mode = S_IRUSR | S_IWUSR; This means
that (inode->i_mode & S_IFMT) == 0. This trips up some SELinux code that
needs to determine if a given inode is a regular file, a directory, etc.
The easiest solution is to just make sure that the anon_inode also sets
S_IFREG.
Signed-off-by: Eric Paris <eparis@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files. percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.
percpu.h -> slab.h dependency is about to be removed. Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability. As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.
http://userweb.kernel.org/~tj/misc/slabh-sweep.py
The script does the followings.
* Scan files for gfp and slab usages and update includes such that
only the necessary includes are there. ie. if only gfp is used,
gfp.h, if slab is used, slab.h.
* When the script inserts a new include, it looks at the include
blocks and try to put the new include such that its order conforms
to its surrounding. It's put in the include block which contains
core kernel includes, in the same order that the rest are ordered -
alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
doesn't seem to be any matching order.
* If the script can't find a place to put a new include (mostly
because the file doesn't have fitting include block), it prints out
an error message indicating which .h file needs to be added to the
file.
The conversion was done in the following steps.
1. The initial automatic conversion of all .c files updated slightly
over 4000 files, deleting around 700 includes and adding ~480 gfp.h
and ~3000 slab.h inclusions. The script emitted errors for ~400
files.
2. Each error was manually checked. Some didn't need the inclusion,
some needed manual addition while adding it to implementation .h or
embedding .c file was more appropriate for others. This step added
inclusions to around 150 files.
3. The script was run again and the output was compared to the edits
from #2 to make sure no file was left behind.
4. Several build tests were done and a couple of problems were fixed.
e.g. lib/decompress_*.c used malloc/free() wrappers around slab
APIs requiring slab.h to be added manually.
5. The script was run on all .h files but without automatically
editing them as sprinkling gfp.h and slab.h inclusions around .h
files could easily lead to inclusion dependency hell. Most gfp.h
inclusion directives were ignored as stuff from gfp.h was usually
wildly available and often used in preprocessor macros. Each
slab.h inclusion directive was examined and added manually as
necessary.
6. percpu.h was updated not to include slab.h.
7. Build test were done on the following configurations and failures
were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my
distributed build env didn't work with gcov compiles) and a few
more options had to be turned off depending on archs to make things
build (like ipr on powerpc/64 which failed due to missing writeq).
* x86 and x86_64 UP and SMP allmodconfig and a custom test config.
* powerpc and powerpc64 SMP allmodconfig
* sparc and sparc64 SMP allmodconfig
* ia64 SMP allmodconfig
* s390 SMP allmodconfig
* alpha SMP allmodconfig
* um on x86_64 SMP allmodconfig
8. percpu.h modifications were reverted so that it could be applied as
a separate patch and serve as bisection point.
Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.
Signed-off-by: Tejun Heo <tj@kernel.org>
Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
Inotify was switched to use anon_inode instead of its own private filesystem
which only had one inode in commit c44dcc56d2 "switch inotify_user to
anon_inode"
The problem with this is that now the inotify inode is not a distinct inode
which can be managed by LSMs. userspace tools which use inotify were allowed
to use the inotify inode but may not have had permission to do read/write type
operations on the anon_inode. After looking at the anon_inode and its users
it looks like the best solution is to just mark the anon_inode as S_PRIVATE
so the security system will ignore it.
Signed-off-by: Eric Paris <eparis@redhat.com>
Acked-by: James Morris <jmorris@namei.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* pull ACC_MODE to fs.h; we have several copies all over the place
* nightmarish expression calculating f_mode by f_flags deserves a helper
too (OPEN_FMODE(flags))
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
It seems a couple places such as arch/ia64/kernel/perfmon.c and
drivers/infiniband/core/uverbs_main.c could use anon_inode_getfile()
instead of a private pseudo-fs + alloc_file(), if only there were a way
to get a read-only file. So provide this by having anon_inode_getfile()
create a read-only file if we pass O_RDONLY in flags.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Filesystems outside the regular namespace do not have to clear DCACHE_UNHASHED
in order to have a working /proc/$pid/fd/XXX. Nothing in proc prevents the
fd link from being used if its dentry is not in the hash.
Also, it does not get put into the dcache hash if DCACHE_UNHASHED is clear;
that depends on the filesystem calling d_add or d_rehash.
So delete the misleading comments and needless code.
Acked-by: Miklos Szeredi <mszeredi@suse.cz>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Add a d_dname method for anon_inodes filesystem, the same way pipefs and
sockfs pseudo filesystems. This allows us to remove the DCACHE_UNHASHED
hack from anon_inodes.c (see next patch).
[AV: inumber is useless here, dropped from anon_inodefs_dname()]
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Miklos Szeredi <mszeredi@suse.cz>
Cc: Davide Libenzi <davidel@xmailserver.org>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Jens Axboe <jens.axboe@oracle.com>
Cc: Christoph Hellwig <hch@lst.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>
Split the anonfd interface into a bare file pointer creation one, and a
file pointer creation plus install one.
There are cases, like the usage of eventfds inside other kernel
interfaces, where the file pointer created by anonfd needs to be used
inside the initialization of other structures.
As it is right now, as soon as anon_inode_getfd() returns, the kenrle can
race with userspace closing the newly installed file descriptor.
This patch, while keeping the old anon_inode_getfd(), introduces a new
anon_inode_getfile() (whose services are reused in anon_inode_getfd())
that allows to split the file creation phase and the fd install one.
Once all the kernel structures are initialized, the code can call the
proper fd_install().
Gregory manifested the need for something like this inside KVM.
Signed-off-by: Davide Libenzi <davidel@xmailserver.org>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: James Morris <jmorris@namei.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Gregory Haskins <ghaskins@novell.com>
Acked-by: Serge Hallyn <serue@us.ibm.com>
Acked-by: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
.set_page_dirty() is one of those a_ops that defaults to the
buffer implementation when not set. Therefore provide a dummy
function to make it do nothing.
(Uncovered by perfcounters fd's which can now be writable-mmap-ed.)
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Davide Libenzi <davidel@xmailserver.org>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
There is an imbalance for anonymous inodes. If the fops->owner field is set,
the module reference count of owner is decreases on release.
("filp_close" --> "__fput" ---> "fops_put")
On the other hand, anon_inode_getfd does not increase the module reference
count of owner. This causes two problems:
- if owner is set, the module refcount goes negative
- if owner is not set, the module can be unloaded while code is running
This patch changes anon_inode_getfd to be symmetric regarding fops->owner
handling.
I have checked all existing users of anon_inode_getfd. Noone sets fops->owner,
thats why nobody has seen the module refcount negative. The refcounting was
tested with a patched and unpatched KVM module.(see patch 2/2) I also did an
epoll_open/close test.
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Reviewed-by: Davide Libenzi <davidel@xmailserver.org>
Signed-off-by: Avi Kivity <avi@redhat.com>
Wrap access to task credentials so that they can be separated more easily from
the task_struct during the introduction of COW creds.
Change most current->(|e|s|fs)[ug]id to current_(|e|s|fs)[ug]id().
Change some task->e?[ug]id to task_e?[ug]id(). In some places it makes more
sense to use RCU directly rather than a convenient wrapper; these will be
addressed by later patches.
Signed-off-by: David Howells <dhowells@redhat.com>
Reviewed-by: James Morris <jmorris@namei.org>
Acked-by: Serge Hallyn <serue@us.ibm.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: James Morris <jmorris@namei.org>
Building on the previous change to anon_inode_getfd, this patch introduces
support for handling of O_NONBLOCK in addition to the already supported
O_CLOEXEC. Following patches will take advantage of this support. As can be
seen, the additional support for supporting this functionality is minimal.
Signed-off-by: Ulrich Drepper <drepper@redhat.com>
Acked-by: Davide Libenzi <davidel@xmailserver.org>
Cc: Michael Kerrisk <mtk.manpages@googlemail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This patch just extends the anon_inode_getfd interface to take an additional
parameter with a flag value. The flag value is passed on to
get_unused_fd_flags in anticipation for a use with the O_CLOEXEC flag.
No actual semantic changes here, the changed callers all pass 0 for now.
[akpm@linux-foundation.org: KVM fix]
Signed-off-by: Ulrich Drepper <drepper@redhat.com>
Acked-by: Davide Libenzi <davidel@xmailserver.org>
Cc: Michael Kerrisk <mtk.manpages@googlemail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
a) none of the callers even looks at inode or file returned by anon_inode_getfd()
b) any caller that would try to look at those would be racy, since by the time
it returns we might have raced with close() from another thread and that
file would be pining for fjords.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Some new uses of get_empty_filp() have crept in; switched
to alloc_file() to make sure that pieces of initialization
won't be missing.
We really need to kill get_empty_filp().
[AV] fixed dentry leak on failure exit in anon_inode_getfd()
Cc: Erez Zadok <ezk@cs.sunysb.edu>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Cc: "J Bruce Fields" <bfields@fieldses.org>
Acked-by: Al Viro <viro@ZenIV.linux.org.uk>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Since we know the shared inode count is always >0, we can avoid igrab()
and use an open coded atomic_inc().
This also fixes a bug noticed by Yan Zheng <yanzheng@21cn.com>: were checking
for an IS_ERR() return from igrab(), but it actually returns NULL on error.
Signed-off-by: Davide Libenzi <davidel@xmailserver.org>
Cc: Yan Zheng <yanzheng@21cn.com>
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/avi/kvm: (80 commits)
KVM: Use CPU_DYING for disabling virtualization
KVM: Tune hotplug/suspend IPIs
KVM: Keep track of which cpus have virtualization enabled
SMP: Allow smp_call_function_single() to current cpu
i386: Allow smp_call_function_single() to current cpu
x86_64: Allow smp_call_function_single() to current cpu
HOTPLUG: Adapt thermal throttle to CPU_DYING
HOTPLUG: Adapt cpuset hotplug callback to CPU_DYING
HOTPLUG: Add CPU_DYING notifier
KVM: Clean up #includes
KVM: Remove kvmfs in favor of the anonymous inodes source
KVM: SVM: Reliably detect if SVM was disabled by BIOS
KVM: VMX: Remove unnecessary code in vmx_tlb_flush()
KVM: MMU: Fix Wrong tlb flush order
KVM: VMX: Reinitialize the real-mode tss when entering real mode
KVM: Avoid useless memory write when possible
KVM: Fix x86 emulator writeback
KVM: Add support for in-kernel pio handlers
KVM: VMX: Fix interrupt checking on lightweight exit
KVM: Adds support for in-kernel mmio handlers
...
kvm uses a pseudo filesystem, kvmfs, to generate inodes, a job that the
new anonymous inodes source does much better.
Cc: Davide Libenzi <davidel@xmailserver.org>
Signed-off-by: Avi Kivity <avi@qumranet.com>
This patch add an anonymous inode source, to be used for files that need
and inode only in order to create a file*. We do not care of having an
inode for each file, and we do not even care of having different names in
the associated dentries (dentry names will be same for classes of file*).
This allow code reuse, and will be used by epoll, signalfd and timerfd
(and whatever else there'll be).
Signed-off-by: Davide Libenzi <davidel@xmailserver.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
