AuxXxilium/linux_dsm_epyc7002
1
0
Fork 0
mirror of https://github.com/AuxXxilium/linux_dsm_epyc7002.git synced 2024-11-24 02:00:54 +07:00
Code Issues Actions Packages Projects Releases Wiki Activity
4316c3c685
linux_dsm_epyc7002/fs/locks.c

2826 lines
74 KiB
C
Raw Normal View History

Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
/*
* linux/fs/locks.c
*
* Provide support for fcntl()'s F_GETLK, F_SETLK, and F_SETLKW calls.
* Doug Evans (dje@spiff.uucp), August 07, 1992
*
* Deadlock detection added.
* FIXME: one thing isn't handled yet:
* - mandatory locks (requires lots of changes elsewhere)
* Kelly Carmichael (kelly@[142.24.8.65]), September 17, 1994.
*
* Miscellaneous edits, and a total rewrite of posix_lock_file() code.
* Kai Petzke (wpp@marie.physik.tu-berlin.de), 1994
*
* Converted file_lock_table to a linked list from an array, which eliminates
* the limits on how many active file locks are open.
* Chad Page (pageone@netcom.com), November 27, 1994
*
* Removed dependency on file descriptors. dup()'ed file descriptors now
* get the same locks as the original file descriptors, and a close() on
* any file descriptor removes ALL the locks on the file for the current
* process. Since locks still depend on the process id, locks are inherited
* after an exec() but not after a fork(). This agrees with POSIX, and both
* BSD and SVR4 practice.
* Andy Walker (andy@lysaker.kvaerner.no), February 14, 1995
*
* Scrapped free list which is redundant now that we allocate locks
* dynamically with kmalloc()/kfree().
* Andy Walker (andy@lysaker.kvaerner.no), February 21, 1995
*
* Implemented two lock personalities - FL_FLOCK and FL_POSIX.
*
* FL_POSIX locks are created with calls to fcntl() and lockf() through the
* fcntl() system call. They have the semantics described above.
*
* FL_FLOCK locks are created with calls to flock(), through the flock()
* system call, which is new. Old C libraries implement flock() via fcntl()
* and will continue to use the old, broken implementation.
*
* FL_FLOCK locks follow the 4.4 BSD flock() semantics. They are associated
* with a file pointer (filp). As a result they can be shared by a parent
* process and its children after a fork(). They are removed when the last
* file descriptor referring to the file pointer is closed (unless explicitly
* unlocked).
*
* FL_FLOCK locks never deadlock, an existing lock is always removed before
* upgrading from shared to exclusive (or vice versa). When this happens
* any processes blocked by the current lock are woken up and allowed to
* run before the new lock is applied.
* Andy Walker (andy@lysaker.kvaerner.no), June 09, 1995
*
* Removed some race conditions in flock_lock_file(), marked other possible
* races. Just grep for FIXME to see them.
* Dmitry Gorodchanin (pgmdsg@ibi.com), February 09, 1996.
*
* Addressed Dmitry's concerns. Deadlock checking no longer recursive.
* Lock allocation changed to GFP_ATOMIC as we can't afford to sleep
* once we've checked for blocking and deadlocking.
* Andy Walker (andy@lysaker.kvaerner.no), April 03, 1996.
*
* Initial implementation of mandatory locks. SunOS turned out to be
* a rotten model, so I implemented the "obvious" semantics.
doc: fix broken references There are numerous broken references to Documentation files (in other Documentation files, in comments, etc.). These broken references are caused by typo's in the references, and by renames or removals of the Documentation files. Some broken references are simply odd. Fix these broken references, sometimes by dropping the irrelevant text they were part of. Signed-off-by: Paul Bolle <pebolle@tiscali.nl> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2011-08-15 07:02:26 +07:00
* See 'Documentation/filesystems/mandatory-locking.txt' for details.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
* Andy Walker (andy@lysaker.kvaerner.no), April 06, 1996.
*
* Don't allow mandatory locks on mmap()'ed files. Added simple functions to
* check if a file has mandatory locks, used by mmap(), open() and creat() to
* see if system call should be rejected. Ref. HP-UX/SunOS/Solaris Reference
* Manual, Section 2.
* Andy Walker (andy@lysaker.kvaerner.no), April 09, 1996.
*
* Tidied up block list handling. Added '/proc/locks' interface.
* Andy Walker (andy@lysaker.kvaerner.no), April 24, 1996.
*
* Fixed deadlock condition for pathological code that mixes calls to
* flock() and fcntl().
* Andy Walker (andy@lysaker.kvaerner.no), April 29, 1996.
*
* Allow only one type of locking scheme (FL_POSIX or FL_FLOCK) to be in use
* for a given file at a time. Changed the CONFIG_LOCK_MANDATORY scheme to
* guarantee sensible behaviour in the case where file system modules might
* be compiled with different options than the kernel itself.
* Andy Walker (andy@lysaker.kvaerner.no), May 15, 1996.
*
* Added a couple of missing wake_up() calls. Thanks to Thomas Meckel
* (Thomas.Meckel@mni.fh-giessen.de) for spotting this.
* Andy Walker (andy@lysaker.kvaerner.no), May 15, 1996.
*
* Changed FL_POSIX locks to use the block list in the same way as FL_FLOCK
* locks. Changed process synchronisation to avoid dereferencing locks that
* have already been freed.
* Andy Walker (andy@lysaker.kvaerner.no), Sep 21, 1996.
*
* Made the block list a circular list to minimise searching in the list.
* Andy Walker (andy@lysaker.kvaerner.no), Sep 25, 1996.
*
* Made mandatory locking a mount option. Default is not to allow mandatory
* locking.
* Andy Walker (andy@lysaker.kvaerner.no), Oct 04, 1996.
*
* Some adaptations for NFS support.
* Olaf Kirch (okir@monad.swb.de), Dec 1996,
*
* Fixed /proc/locks interface so that we can't overrun the buffer we are handed.
* Andy Walker (andy@lysaker.kvaerner.no), May 12, 1997.
*
* Use slab allocator instead of kmalloc/kfree.
* Use generic list implementation from <linux/list.h>.
* Sped up posix_locks_deadlock by only considering blocked locks.
* Matthew Wilcox <willy@debian.org>, March, 2000.
*
* Leases and LOCK_MAND
* Matthew Wilcox <willy@debian.org>, June, 2000.
* Stephen Rothwell <sfr@canb.auug.org.au>, June, 2000.
*/
#include <linux/capability.h>
#include <linux/file.h>
[PATCH] split linux/file.h Initial splitoff of the low-level stuff; taken to fdtable.h Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2008-04-24 18:44:08 +07:00
#include <linux/fdtable.h>
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
#include <linux/fs.h>
#include <linux/init.h>
#include <linux/security.h>
#include <linux/slab.h>
#include <linux/syscalls.h>
#include <linux/time.h>
[PATCH] files: fix preemption issues With the new fdtable locking rules, you have to protect fdtable with either ->file_lock or rcu_read_lock/unlock(). There are some places where we aren't doing either. This patch fixes those places. 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>
2005-09-17 09:28:13 +07:00
#include <linux/rcupdate.h>
pid-namespaces-vs-locks-interaction fcntl(F_GETLK,..) can return pid of process for not current pid namespace (if process is belonged to the several namespaces). It is true also for pids in /proc/locks. So correct behavior is saving pointer to the struct pid of the process lock owner. Signed-off-by: Vitaliy Gusev <vgusev@openvz.org> Acked-by: Serge Hallyn <serue@us.ibm.com> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-01-17 07:07:08 +07:00
#include <linux/pid_namespace.h>
locks: turn the blocked_list into a hashtable Break up the blocked_list into a hashtable, using the fl_owner as a key. This speeds up searching the hash chains, which is especially significant for deadlock detection. Note that the initial implementation assumes that hashing on fl_owner is sufficient. In most cases it should be, with the notable exception being server-side lockd, which compares ownership using a tuple of the nlm_host and the pid sent in the lock request. So, this may degrade to a single hash bucket when you only have a single NFS client. That will be addressed in a later patch. The careful observer may note that this patch leaves the file_lock_list alone. There's much less of a case for turning the file_lock_list into a hashtable. The only user of that list is the code that generates /proc/locks, and it always walks the entire list. Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:18 +07:00
#include <linux/hashtable.h>
locks: move file_lock_list to a set of percpu hlist_heads and convert file_lock_lock to an lglock The file_lock_list is only used for /proc/locks. The vastly common case is for locks to be put onto the list and come off again, without ever being traversed. Help optimize for this use-case by moving to percpu hlist_head-s. At the same time, we can make the locking less contentious by moving to an lglock. When iterating over the lists for /proc/locks, we must take the global lock and then iterate over each CPU's list in turn. This change necessitates a new fl_link_cpu field to keep track of which CPU the entry is on. On x86_64 at least, this field is placed within an existing hole in the struct to avoid growing the size. Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:22 +07:00
#include <linux/percpu.h>
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
locks: add some tracepoints in the lease handling code v2: add a __break_lease tracepoint for non-blocking case Recently, I needed these to help track down a softlockup when recalling a delegation, but they might be helpful in other situations as well. Cc: "J. Bruce Fields" <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@poochiereds.net>
2014-05-10 01:13:05 +07:00
#define CREATE_TRACE_POINTS
#include <trace/events/filelock.h>
Replace <asm/uaccess.h> with <linux/uaccess.h> globally This was entirely automated, using the script by Al: PATT='^[[:blank:]]*#[[:blank:]]*include[[:blank:]]*<asm/uaccess.h>' sed -i -e "s!$PATT!#include <linux/uaccess.h>!" \ $(git grep -l "$PATT"|grep -v ^include/linux/uaccess.h) to do the replacement at the end of the merge window. Requested-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-12-25 02:46:01 +07:00
#include <linux/uaccess.h>
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
#define IS_POSIX(fl) (fl->fl_flags & FL_POSIX)
#define IS_FLOCK(fl) (fl->fl_flags & FL_FLOCK)
fs: add FL_LAYOUT lease type This (ab-)uses the file locking code to allow filesystems to recall outstanding pNFS layouts on a file. This new lease type is similar but not quite the same as FL_DELEG. A FL_LAYOUT lease can always be granted, an a per-filesystem lock (XFS iolock for the initial implementation) ensures not FL_LAYOUT leases granted when we would need to recall them. Also included are changes that allow multiple outstanding read leases of different types on the same file as long as they have a differnt owner. This wasn't a problem until now as nfsd never set FL_LEASE leases, and no one else used FL_DELEG leases, but given that nfsd will also issues FL_LAYOUT leases we will have to handle it now. Signed-off-by: Christoph Hellwig <hch@lst.de>
2015-01-22 01:17:03 +07:00
#define IS_LEASE(fl) (fl->fl_flags & (FL_LEASE|FL_DELEG|FL_LAYOUT))
locks: rename FL_FILE_PVT and IS_FILE_PVT to use "*_OFDLCK" instead File-private locks have been re-christened as "open file description" locks. Finish the symbol name cleanup in the internal implementation. Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-04-22 19:24:32 +07:00
#define IS_OFDLCK(fl) (fl->fl_flags & FL_OFDLCK)
fs/locks: Remove fl_nspid and use fs-specific l_pid for remote locks Since commit c69899a17ca4 "NFSv4: Update of VFS byte range lock must be atomic with the stateid update", NFSv4 has been inserting locks in rpciod worker context. The result is that the file_lock's fl_nspid is the kworker's pid instead of the original userspace pid. The fl_nspid is only used to represent the namespaced virtual pid number when displaying locks or returning from F_GETLK. There's no reason to set it for every inserted lock, since we can usually just look it up from fl_pid. So, instead of looking up and holding struct pid for every lock, let's just look up the virtual pid number from fl_pid when it is needed. That means we can remove fl_nspid entirely. The translaton and presentation of fl_pid should handle the following four cases: 1 - F_GETLK on a remote file with a remote lock: In this case, the filesystem should determine the l_pid to return here. Filesystems should indicate that the fl_pid represents a non-local pid value that should not be translated by returning an fl_pid <= 0. 2 - F_GETLK on a local file with a remote lock: This should be the l_pid of the lock manager process, and translated. 3 - F_GETLK on a remote file with a local lock, and 4 - F_GETLK on a local file with a local lock: These should be the translated l_pid of the local locking process. Fuse was already doing the correct thing by translating the pid into the caller's namespace. With this change we must update fuse to translate to init's pid namespace, so that the locks API can then translate from init's pid namespace into the pid namespace of the caller. With this change, the locks API will expect that if a filesystem returns a remote pid as opposed to a local pid for F_GETLK, that remote pid will be <= 0. This signifies that the pid is remote, and the locks API will forego translating that pid into the pid namespace of the local calling process. Finally, we convert remote filesystems to present remote pids using negative numbers. Have lustre, 9p, ceph, cifs, and dlm negate the remote pid returned for F_GETLK lock requests. Since local pids will never be larger than PID_MAX_LIMIT (which is currently defined as <= 4 million), but pid_t is an unsigned int, we should have plenty of room to represent remote pids with negative numbers if we assume that remote pid numbers are similarly limited. If this is not the case, then we run the risk of having a remote pid returned for which there is also a corresponding local pid. This is a problem we have now, but this patch should reduce the chances of that occurring, while also returning those remote pid numbers, for whatever that may be worth. Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-07-16 21:28:22 +07:00
#define IS_REMOTELCK(fl) (fl->fl_pid <= 0)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
locks: minor lease cleanup Use a helper function, to simplify upcoming changes. Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-27 07:10:51 +07:00
static bool lease_breaking(struct file_lock *fl)
{
locks: fix tracking of inprogress lease breaks We currently use a bit in fl_flags to record whether a lease is being broken, and set fl_type to the type (RDLCK or UNLCK) that it will eventually have. This means that once the lease break starts, we forget what the lease's type *used* to be. Breaking a read lease will then result in blocking read opens, even though there's no conflict--because the lease type is now F_UNLCK and we can no longer tell whether it was previously a read or write lease. So, instead keep fl_type as the original type (the type which we enforce), and keep track of whether we're unlocking or merely downgrading by replacing the single FL_INPROGRESS flag by FL_UNLOCK_PENDING and FL_DOWNGRADE_PENDING flags. To get this right we also need to track separate downgrade and break times, to handle the case where a write-leased file gets conflicting opens first for read, then later for write. (I first considered just eliminating the downgrade behavior completely--nfsv4 doesn't need it, and nobody as far as I can tell actually uses it currently--but Jeremy Allison tells me that Windows oplocks do behave this way, so Samba will probably use this some day.) Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-27 05:25:49 +07:00
return fl->fl_flags & (FL_UNLOCK_PENDING | FL_DOWNGRADE_PENDING);
}
static int target_leasetype(struct file_lock *fl)
{
if (fl->fl_flags & FL_UNLOCK_PENDING)
return F_UNLCK;
if (fl->fl_flags & FL_DOWNGRADE_PENDING)
return F_RDLCK;
return fl->fl_type;
locks: minor lease cleanup Use a helper function, to simplify upcoming changes. Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-27 07:10:51 +07:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
int leases_enable = 1;
int lease_break_time = 45;
locks: protect most of the file_lock handling with i_lock Having a global lock that protects all of this code is a clear scalability problem. Instead of doing that, move most of the code to be protected by the i_lock instead. The exceptions are the global lists that the ->fl_link sits on, and the ->fl_block list. ->fl_link is what connects these structures to the global lists, so we must ensure that we hold those locks when iterating over or updating these lists. Furthermore, sound deadlock detection requires that we hold the blocked_list state steady while checking for loops. We also must ensure that the search and update to the list are atomic. For the checking and insertion side of the blocked_list, push the acquisition of the global lock into __posix_lock_file and ensure that checking and update of the blocked_list is done without dropping the lock in between. On the removal side, when waking up blocked lock waiters, take the global lock before walking the blocked list and dequeue the waiters from the global list prior to removal from the fl_block list. With this, deadlock detection should be race free while we minimize excessive file_lock_lock thrashing. Finally, in order to avoid a lock inversion problem when handling /proc/locks output we must ensure that manipulations of the fl_block list are also protected by the file_lock_lock. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:15 +07:00
/*
locks: move file_lock_list to a set of percpu hlist_heads and convert file_lock_lock to an lglock The file_lock_list is only used for /proc/locks. The vastly common case is for locks to be put onto the list and come off again, without ever being traversed. Help optimize for this use-case by moving to percpu hlist_head-s. At the same time, we can make the locking less contentious by moving to an lglock. When iterating over the lists for /proc/locks, we must take the global lock and then iterate over each CPU's list in turn. This change necessitates a new fl_link_cpu field to keep track of which CPU the entry is on. On x86_64 at least, this field is placed within an existing hole in the struct to avoid growing the size. Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:22 +07:00
* The global file_lock_list is only used for displaying /proc/locks, so we
fs/locks: Replace lg_local with a per-cpu spinlock As Oleg suggested, replace file_lock_list with a structure containing the hlist head and a spinlock. This completely removes the lglock from fs/locks. Suggested-by: Oleg Nesterov <oleg@redhat.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Al Viro <viro@ZenIV.linux.org.uk> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: dave@stgolabs.net Cc: der.herr@hofr.at Cc: paulmck@linux.vnet.ibm.com Cc: riel@redhat.com Cc: tj@kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-06-22 19:16:34 +07:00
* keep a list on each CPU, with each list protected by its own spinlock.
* Global serialization is done using file_rwsem.
*
* Note that alterations to the list also require that the relevant flc_lock is
* held.
locks: protect most of the file_lock handling with i_lock Having a global lock that protects all of this code is a clear scalability problem. Instead of doing that, move most of the code to be protected by the i_lock instead. The exceptions are the global lists that the ->fl_link sits on, and the ->fl_block list. ->fl_link is what connects these structures to the global lists, so we must ensure that we hold those locks when iterating over or updating these lists. Furthermore, sound deadlock detection requires that we hold the blocked_list state steady while checking for loops. We also must ensure that the search and update to the list are atomic. For the checking and insertion side of the blocked_list, push the acquisition of the global lock into __posix_lock_file and ensure that checking and update of the blocked_list is done without dropping the lock in between. On the removal side, when waking up blocked lock waiters, take the global lock before walking the blocked list and dequeue the waiters from the global list prior to removal from the fl_block list. With this, deadlock detection should be race free while we minimize excessive file_lock_lock thrashing. Finally, in order to avoid a lock inversion problem when handling /proc/locks output we must ensure that manipulations of the fl_block list are also protected by the file_lock_lock. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:15 +07:00
*/
fs/locks: Replace lg_local with a per-cpu spinlock As Oleg suggested, replace file_lock_list with a structure containing the hlist head and a spinlock. This completely removes the lglock from fs/locks. Suggested-by: Oleg Nesterov <oleg@redhat.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Al Viro <viro@ZenIV.linux.org.uk> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: dave@stgolabs.net Cc: der.herr@hofr.at Cc: paulmck@linux.vnet.ibm.com Cc: riel@redhat.com Cc: tj@kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-06-22 19:16:34 +07:00
struct file_lock_list_struct {
spinlock_t lock;
struct hlist_head hlist;
};
static DEFINE_PER_CPU(struct file_lock_list_struct, file_lock_list);
fs/locks: Replace lg_global with a percpu-rwsem Replace the global part of the lglock with a percpu-rwsem. Since fcl_lock is a spinlock and itself nests under i_lock, which too is a spinlock we cannot acquire sleeping locks at locks_{insert,remove}_global_locks(). We can however wrap all fcl_lock acquisitions with percpu_down_read such that all invocations of locks_{insert,remove}_global_locks() have that read lock held. This allows us to replace the lg_global part of the lglock with the write side of the rwsem. In the absense of writers, percpu_{down,up}_read() are free of atomic instructions. This further avoids the very long preempt-disable regions caused by lglock on larger machines. Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Al Viro <viro@ZenIV.linux.org.uk> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: dave@stgolabs.net Cc: der.herr@hofr.at Cc: paulmck@linux.vnet.ibm.com Cc: riel@redhat.com Cc: tj@kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-06-22 19:16:33 +07:00
DEFINE_STATIC_PERCPU_RWSEM(file_rwsem);
locks: encapsulate the fl_link list handling Move the fl_link list handling routines into a separate set of helpers. Also ensure that locks and requests are always put on global lists last (after fully initializing them) and are taken off before unintializing them. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:14 +07:00
locks: protect most of the file_lock handling with i_lock Having a global lock that protects all of this code is a clear scalability problem. Instead of doing that, move most of the code to be protected by the i_lock instead. The exceptions are the global lists that the ->fl_link sits on, and the ->fl_block list. ->fl_link is what connects these structures to the global lists, so we must ensure that we hold those locks when iterating over or updating these lists. Furthermore, sound deadlock detection requires that we hold the blocked_list state steady while checking for loops. We also must ensure that the search and update to the list are atomic. For the checking and insertion side of the blocked_list, push the acquisition of the global lock into __posix_lock_file and ensure that checking and update of the blocked_list is done without dropping the lock in between. On the removal side, when waking up blocked lock waiters, take the global lock before walking the blocked list and dequeue the waiters from the global list prior to removal from the fl_block list. With this, deadlock detection should be race free while we minimize excessive file_lock_lock thrashing. Finally, in order to avoid a lock inversion problem when handling /proc/locks output we must ensure that manipulations of the fl_block list are also protected by the file_lock_lock. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:15 +07:00
/*
locks: turn the blocked_list into a hashtable Break up the blocked_list into a hashtable, using the fl_owner as a key. This speeds up searching the hash chains, which is especially significant for deadlock detection. Note that the initial implementation assumes that hashing on fl_owner is sufficient. In most cases it should be, with the notable exception being server-side lockd, which compares ownership using a tuple of the nlm_host and the pid sent in the lock request. So, this may degrade to a single hash bucket when you only have a single NFS client. That will be addressed in a later patch. The careful observer may note that this patch leaves the file_lock_list alone. There's much less of a case for turning the file_lock_list into a hashtable. The only user of that list is the code that generates /proc/locks, and it always walks the entire list. Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:18 +07:00
* The blocked_hash is used to find POSIX lock loops for deadlock detection.
locks: give the blocked_hash its own spinlock There's no reason we have to protect the blocked_hash and file_lock_list with the same spinlock. With the tests I have, breaking it in two gives a barely measurable performance benefit, but it seems reasonable to make this locking as granular as possible. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:20 +07:00
* It is protected by blocked_lock_lock.
locks: turn the blocked_list into a hashtable Break up the blocked_list into a hashtable, using the fl_owner as a key. This speeds up searching the hash chains, which is especially significant for deadlock detection. Note that the initial implementation assumes that hashing on fl_owner is sufficient. In most cases it should be, with the notable exception being server-side lockd, which compares ownership using a tuple of the nlm_host and the pid sent in the lock request. So, this may degrade to a single hash bucket when you only have a single NFS client. That will be addressed in a later patch. The careful observer may note that this patch leaves the file_lock_list alone. There's much less of a case for turning the file_lock_list into a hashtable. The only user of that list is the code that generates /proc/locks, and it always walks the entire list. Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:18 +07:00
*
* We hash locks by lockowner in order to optimize searching for the lock a
* particular lockowner is waiting on.
*
* FIXME: make this value scale via some heuristic? We generally will want more
* buckets when we have more lockowners holding locks, but that's a little
* difficult to determine without knowing what the workload will look like.
locks: protect most of the file_lock handling with i_lock Having a global lock that protects all of this code is a clear scalability problem. Instead of doing that, move most of the code to be protected by the i_lock instead. The exceptions are the global lists that the ->fl_link sits on, and the ->fl_block list. ->fl_link is what connects these structures to the global lists, so we must ensure that we hold those locks when iterating over or updating these lists. Furthermore, sound deadlock detection requires that we hold the blocked_list state steady while checking for loops. We also must ensure that the search and update to the list are atomic. For the checking and insertion side of the blocked_list, push the acquisition of the global lock into __posix_lock_file and ensure that checking and update of the blocked_list is done without dropping the lock in between. On the removal side, when waking up blocked lock waiters, take the global lock before walking the blocked list and dequeue the waiters from the global list prior to removal from the fl_block list. With this, deadlock detection should be race free while we minimize excessive file_lock_lock thrashing. Finally, in order to avoid a lock inversion problem when handling /proc/locks output we must ensure that manipulations of the fl_block list are also protected by the file_lock_lock. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:15 +07:00
*/
locks: turn the blocked_list into a hashtable Break up the blocked_list into a hashtable, using the fl_owner as a key. This speeds up searching the hash chains, which is especially significant for deadlock detection. Note that the initial implementation assumes that hashing on fl_owner is sufficient. In most cases it should be, with the notable exception being server-side lockd, which compares ownership using a tuple of the nlm_host and the pid sent in the lock request. So, this may degrade to a single hash bucket when you only have a single NFS client. That will be addressed in a later patch. The careful observer may note that this patch leaves the file_lock_list alone. There's much less of a case for turning the file_lock_list into a hashtable. The only user of that list is the code that generates /proc/locks, and it always walks the entire list. Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:18 +07:00
#define BLOCKED_HASH_BITS 7
static DEFINE_HASHTABLE(blocked_hash, BLOCKED_HASH_BITS);
locks: encapsulate the fl_link list handling Move the fl_link list handling routines into a separate set of helpers. Also ensure that locks and requests are always put on global lists last (after fully initializing them) and are taken off before unintializing them. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:14 +07:00
locks: protect most of the file_lock handling with i_lock Having a global lock that protects all of this code is a clear scalability problem. Instead of doing that, move most of the code to be protected by the i_lock instead. The exceptions are the global lists that the ->fl_link sits on, and the ->fl_block list. ->fl_link is what connects these structures to the global lists, so we must ensure that we hold those locks when iterating over or updating these lists. Furthermore, sound deadlock detection requires that we hold the blocked_list state steady while checking for loops. We also must ensure that the search and update to the list are atomic. For the checking and insertion side of the blocked_list, push the acquisition of the global lock into __posix_lock_file and ensure that checking and update of the blocked_list is done without dropping the lock in between. On the removal side, when waking up blocked lock waiters, take the global lock before walking the blocked list and dequeue the waiters from the global list prior to removal from the fl_block list. With this, deadlock detection should be race free while we minimize excessive file_lock_lock thrashing. Finally, in order to avoid a lock inversion problem when handling /proc/locks output we must ensure that manipulations of the fl_block list are also protected by the file_lock_lock. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:15 +07:00
/*
locks: give the blocked_hash its own spinlock There's no reason we have to protect the blocked_hash and file_lock_list with the same spinlock. With the tests I have, breaking it in two gives a barely measurable performance benefit, but it seems reasonable to make this locking as granular as possible. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:20 +07:00
* This lock protects the blocked_hash. Generally, if you're accessing it, you
* want to be holding this lock.
locks: protect most of the file_lock handling with i_lock Having a global lock that protects all of this code is a clear scalability problem. Instead of doing that, move most of the code to be protected by the i_lock instead. The exceptions are the global lists that the ->fl_link sits on, and the ->fl_block list. ->fl_link is what connects these structures to the global lists, so we must ensure that we hold those locks when iterating over or updating these lists. Furthermore, sound deadlock detection requires that we hold the blocked_list state steady while checking for loops. We also must ensure that the search and update to the list are atomic. For the checking and insertion side of the blocked_list, push the acquisition of the global lock into __posix_lock_file and ensure that checking and update of the blocked_list is done without dropping the lock in between. On the removal side, when waking up blocked lock waiters, take the global lock before walking the blocked list and dequeue the waiters from the global list prior to removal from the fl_block list. With this, deadlock detection should be race free while we minimize excessive file_lock_lock thrashing. Finally, in order to avoid a lock inversion problem when handling /proc/locks output we must ensure that manipulations of the fl_block list are also protected by the file_lock_lock. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:15 +07:00
*
fs/locks: rename some lists and pointers. struct file lock contains an 'fl_next' pointer which is used to point to the lock that this request is blocked waiting for. So rename it to fl_blocker. The fl_blocked list_head in an active lock is the head of a list of blocked requests. In a request it is a node in that list. These are two distinct uses, so replace with two list_heads with different names. fl_blocked_requests is the head of a list of blocked requests fl_blocked_member is a node in a member of that list. The two different list_heads are never used at the same time, but that will change in a future patch. Note that a tracepoint is changed to report fl_blocker instead of fl_next. Signed-off-by: NeilBrown <neilb@suse.com> Reviewed-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Jeff Layton <jlayton@kernel.org>
2018-11-30 06:04:08 +07:00
* In addition, it also protects the fl->fl_blocked_requests list, and the
* fl->fl_blocker pointer for file_lock structures that are acting as lock
* requests (in contrast to those that are acting as records of acquired locks).
locks: protect most of the file_lock handling with i_lock Having a global lock that protects all of this code is a clear scalability problem. Instead of doing that, move most of the code to be protected by the i_lock instead. The exceptions are the global lists that the ->fl_link sits on, and the ->fl_block list. ->fl_link is what connects these structures to the global lists, so we must ensure that we hold those locks when iterating over or updating these lists. Furthermore, sound deadlock detection requires that we hold the blocked_list state steady while checking for loops. We also must ensure that the search and update to the list are atomic. For the checking and insertion side of the blocked_list, push the acquisition of the global lock into __posix_lock_file and ensure that checking and update of the blocked_list is done without dropping the lock in between. On the removal side, when waking up blocked lock waiters, take the global lock before walking the blocked list and dequeue the waiters from the global list prior to removal from the fl_block list. With this, deadlock detection should be race free while we minimize excessive file_lock_lock thrashing. Finally, in order to avoid a lock inversion problem when handling /proc/locks output we must ensure that manipulations of the fl_block list are also protected by the file_lock_lock. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:15 +07:00
*
* Note that when we acquire this lock in order to change the above fields,
locks: add a dedicated spinlock to protect i_flctx lists We can now add a dedicated spinlock without expanding struct inode. Change to using that to protect the various i_flctx lists. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
* we often hold the flc_lock as well. In certain cases, when reading the fields
locks: protect most of the file_lock handling with i_lock Having a global lock that protects all of this code is a clear scalability problem. Instead of doing that, move most of the code to be protected by the i_lock instead. The exceptions are the global lists that the ->fl_link sits on, and the ->fl_block list. ->fl_link is what connects these structures to the global lists, so we must ensure that we hold those locks when iterating over or updating these lists. Furthermore, sound deadlock detection requires that we hold the blocked_list state steady while checking for loops. We also must ensure that the search and update to the list are atomic. For the checking and insertion side of the blocked_list, push the acquisition of the global lock into __posix_lock_file and ensure that checking and update of the blocked_list is done without dropping the lock in between. On the removal side, when waking up blocked lock waiters, take the global lock before walking the blocked list and dequeue the waiters from the global list prior to removal from the fl_block list. With this, deadlock detection should be race free while we minimize excessive file_lock_lock thrashing. Finally, in order to avoid a lock inversion problem when handling /proc/locks output we must ensure that manipulations of the fl_block list are also protected by the file_lock_lock. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:15 +07:00
* protected by this lock, we can skip acquiring it iff we already hold the
locks: add a dedicated spinlock to protect i_flctx lists We can now add a dedicated spinlock without expanding struct inode. Change to using that to protect the various i_flctx lists. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
* flc_lock.
locks: protect most of the file_lock handling with i_lock Having a global lock that protects all of this code is a clear scalability problem. Instead of doing that, move most of the code to be protected by the i_lock instead. The exceptions are the global lists that the ->fl_link sits on, and the ->fl_block list. ->fl_link is what connects these structures to the global lists, so we must ensure that we hold those locks when iterating over or updating these lists. Furthermore, sound deadlock detection requires that we hold the blocked_list state steady while checking for loops. We also must ensure that the search and update to the list are atomic. For the checking and insertion side of the blocked_list, push the acquisition of the global lock into __posix_lock_file and ensure that checking and update of the blocked_list is done without dropping the lock in between. On the removal side, when waking up blocked lock waiters, take the global lock before walking the blocked list and dequeue the waiters from the global list prior to removal from the fl_block list. With this, deadlock detection should be race free while we minimize excessive file_lock_lock thrashing. Finally, in order to avoid a lock inversion problem when handling /proc/locks output we must ensure that manipulations of the fl_block list are also protected by the file_lock_lock. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:15 +07:00
*/
locks: give the blocked_hash its own spinlock There's no reason we have to protect the blocked_hash and file_lock_list with the same spinlock. With the tests I have, breaking it in two gives a barely measurable performance benefit, but it seems reasonable to make this locking as granular as possible. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:20 +07:00
static DEFINE_SPINLOCK(blocked_lock_lock);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
locks: add a new struct file_locking_context pointer to struct inode The current scheme of using the i_flock list is really difficult to manage. There is also a legitimate desire for a per-inode spinlock to manage these lists that isn't the i_lock. Start conversion to a new scheme to eventually replace the old i_flock list with a new "file_lock_context" object. We start by adding a new i_flctx to struct inode. For now, it lives in parallel with i_flock list, but will eventually replace it. The idea is to allocate a structure to sit in that pointer and act as a locus for all things file locking. We allocate a file_lock_context for an inode when the first lock is added to it, and it's only freed when the inode is freed. We use the i_lock to protect the assignment, but afterward it should mostly be accessed locklessly. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:54 +07:00
static struct kmem_cache *flctx_cache __read_mostly;
[PATCH] slab: remove kmem_cache_t Replace all uses of kmem_cache_t with struct kmem_cache. The patch was generated using the following script: #!/bin/sh # # Replace one string by another in all the kernel sources. # set -e for file in `find * -name "*.c" -o -name "*.h"|xargs grep -l $1`; do quilt add $file sed -e "1,\$s/$1/$2/g" $file >/tmp/$$ mv /tmp/$$ $file quilt refresh done The script was run like this sh replace kmem_cache_t "struct kmem_cache" Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 11:33:20 +07:00
static struct kmem_cache *filelock_cache __read_mostly;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
locks: add a new struct file_locking_context pointer to struct inode The current scheme of using the i_flock list is really difficult to manage. There is also a legitimate desire for a per-inode spinlock to manage these lists that isn't the i_lock. Start conversion to a new scheme to eventually replace the old i_flock list with a new "file_lock_context" object. We start by adding a new i_flctx to struct inode. For now, it lives in parallel with i_flock list, but will eventually replace it. The idea is to allocate a structure to sit in that pointer and act as a locus for all things file locking. We allocate a file_lock_context for an inode when the first lock is added to it, and it's only freed when the inode is freed. We use the i_lock to protect the assignment, but afterward it should mostly be accessed locklessly. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:54 +07:00
static struct file_lock_context *
locks: don't allocate a lock context for an F_UNLCK request In the event that we get an F_UNLCK request on an inode that has no lock context, there is no reason to allocate one. Change locks_get_lock_context to take a "type" pointer and avoid allocating a new context if it's F_UNLCK. Then, fix the callers to return appropriately if that function returns NULL. Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2015-04-03 20:04:03 +07:00
locks_get_lock_context(struct inode *inode, int type)
locks: add a new struct file_locking_context pointer to struct inode The current scheme of using the i_flock list is really difficult to manage. There is also a legitimate desire for a per-inode spinlock to manage these lists that isn't the i_lock. Start conversion to a new scheme to eventually replace the old i_flock list with a new "file_lock_context" object. We start by adding a new i_flctx to struct inode. For now, it lives in parallel with i_flock list, but will eventually replace it. The idea is to allocate a structure to sit in that pointer and act as a locus for all things file locking. We allocate a file_lock_context for an inode when the first lock is added to it, and it's only freed when the inode is freed. We use the i_lock to protect the assignment, but afterward it should mostly be accessed locklessly. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:54 +07:00
{
fs: fix data races on inode->i_flctx locks_get_lock_context() uses cmpxchg() to install i_flctx. cmpxchg() is a release operation which is correct. But it uses a plain load to load i_flctx. This is incorrect. Subsequent loads from i_flctx can hoist above the load of i_flctx pointer itself and observe uninitialized garbage there. This in turn can lead to corruption of ctx->flc_lock and other members. Documentation/memory-barriers.txt explicitly requires to use a barrier in such context: "A load-load control dependency requires a full read memory barrier". Use smp_load_acquire() in locks_get_lock_context() and in bunch of other functions that can proceed concurrently with locks_get_lock_context(). The data race was found with KernelThreadSanitizer (KTSAN). Signed-off-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-09-21 14:43:06 +07:00
struct file_lock_context *ctx;
locks: add a new struct file_locking_context pointer to struct inode The current scheme of using the i_flock list is really difficult to manage. There is also a legitimate desire for a per-inode spinlock to manage these lists that isn't the i_lock. Start conversion to a new scheme to eventually replace the old i_flock list with a new "file_lock_context" object. We start by adding a new i_flctx to struct inode. For now, it lives in parallel with i_flock list, but will eventually replace it. The idea is to allocate a structure to sit in that pointer and act as a locus for all things file locking. We allocate a file_lock_context for an inode when the first lock is added to it, and it's only freed when the inode is freed. We use the i_lock to protect the assignment, but afterward it should mostly be accessed locklessly. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:54 +07:00
fs: fix data races on inode->i_flctx locks_get_lock_context() uses cmpxchg() to install i_flctx. cmpxchg() is a release operation which is correct. But it uses a plain load to load i_flctx. This is incorrect. Subsequent loads from i_flctx can hoist above the load of i_flctx pointer itself and observe uninitialized garbage there. This in turn can lead to corruption of ctx->flc_lock and other members. Documentation/memory-barriers.txt explicitly requires to use a barrier in such context: "A load-load control dependency requires a full read memory barrier". Use smp_load_acquire() in locks_get_lock_context() and in bunch of other functions that can proceed concurrently with locks_get_lock_context(). The data race was found with KernelThreadSanitizer (KTSAN). Signed-off-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-09-21 14:43:06 +07:00
/* paired with cmpxchg() below */
ctx = smp_load_acquire(&inode->i_flctx);
if (likely(ctx) || type == F_UNLCK)
locks: add a new struct file_locking_context pointer to struct inode The current scheme of using the i_flock list is really difficult to manage. There is also a legitimate desire for a per-inode spinlock to manage these lists that isn't the i_lock. Start conversion to a new scheme to eventually replace the old i_flock list with a new "file_lock_context" object. We start by adding a new i_flctx to struct inode. For now, it lives in parallel with i_flock list, but will eventually replace it. The idea is to allocate a structure to sit in that pointer and act as a locus for all things file locking. We allocate a file_lock_context for an inode when the first lock is added to it, and it's only freed when the inode is freed. We use the i_lock to protect the assignment, but afterward it should mostly be accessed locklessly. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:54 +07:00
goto out;
fs: fix data races on inode->i_flctx locks_get_lock_context() uses cmpxchg() to install i_flctx. cmpxchg() is a release operation which is correct. But it uses a plain load to load i_flctx. This is incorrect. Subsequent loads from i_flctx can hoist above the load of i_flctx pointer itself and observe uninitialized garbage there. This in turn can lead to corruption of ctx->flc_lock and other members. Documentation/memory-barriers.txt explicitly requires to use a barrier in such context: "A load-load control dependency requires a full read memory barrier". Use smp_load_acquire() in locks_get_lock_context() and in bunch of other functions that can proceed concurrently with locks_get_lock_context(). The data race was found with KernelThreadSanitizer (KTSAN). Signed-off-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-09-21 14:43:06 +07:00
ctx = kmem_cache_alloc(flctx_cache, GFP_KERNEL);
if (!ctx)
locks: add a new struct file_locking_context pointer to struct inode The current scheme of using the i_flock list is really difficult to manage. There is also a legitimate desire for a per-inode spinlock to manage these lists that isn't the i_lock. Start conversion to a new scheme to eventually replace the old i_flock list with a new "file_lock_context" object. We start by adding a new i_flctx to struct inode. For now, it lives in parallel with i_flock list, but will eventually replace it. The idea is to allocate a structure to sit in that pointer and act as a locus for all things file locking. We allocate a file_lock_context for an inode when the first lock is added to it, and it's only freed when the inode is freed. We use the i_lock to protect the assignment, but afterward it should mostly be accessed locklessly. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:54 +07:00
goto out;
fs: fix data races on inode->i_flctx locks_get_lock_context() uses cmpxchg() to install i_flctx. cmpxchg() is a release operation which is correct. But it uses a plain load to load i_flctx. This is incorrect. Subsequent loads from i_flctx can hoist above the load of i_flctx pointer itself and observe uninitialized garbage there. This in turn can lead to corruption of ctx->flc_lock and other members. Documentation/memory-barriers.txt explicitly requires to use a barrier in such context: "A load-load control dependency requires a full read memory barrier". Use smp_load_acquire() in locks_get_lock_context() and in bunch of other functions that can proceed concurrently with locks_get_lock_context(). The data race was found with KernelThreadSanitizer (KTSAN). Signed-off-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-09-21 14:43:06 +07:00
spin_lock_init(&ctx->flc_lock);
INIT_LIST_HEAD(&ctx->flc_flock);
INIT_LIST_HEAD(&ctx->flc_posix);
INIT_LIST_HEAD(&ctx->flc_lease);
locks: add a new struct file_locking_context pointer to struct inode The current scheme of using the i_flock list is really difficult to manage. There is also a legitimate desire for a per-inode spinlock to manage these lists that isn't the i_lock. Start conversion to a new scheme to eventually replace the old i_flock list with a new "file_lock_context" object. We start by adding a new i_flctx to struct inode. For now, it lives in parallel with i_flock list, but will eventually replace it. The idea is to allocate a structure to sit in that pointer and act as a locus for all things file locking. We allocate a file_lock_context for an inode when the first lock is added to it, and it's only freed when the inode is freed. We use the i_lock to protect the assignment, but afterward it should mostly be accessed locklessly. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:54 +07:00
/*
* Assign the pointer if it's not already assigned. If it is, then
* free the context we just allocated.
*/
fs: fix data races on inode->i_flctx locks_get_lock_context() uses cmpxchg() to install i_flctx. cmpxchg() is a release operation which is correct. But it uses a plain load to load i_flctx. This is incorrect. Subsequent loads from i_flctx can hoist above the load of i_flctx pointer itself and observe uninitialized garbage there. This in turn can lead to corruption of ctx->flc_lock and other members. Documentation/memory-barriers.txt explicitly requires to use a barrier in such context: "A load-load control dependency requires a full read memory barrier". Use smp_load_acquire() in locks_get_lock_context() and in bunch of other functions that can proceed concurrently with locks_get_lock_context(). The data race was found with KernelThreadSanitizer (KTSAN). Signed-off-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-09-21 14:43:06 +07:00
if (cmpxchg(&inode->i_flctx, NULL, ctx)) {
kmem_cache_free(flctx_cache, ctx);
ctx = smp_load_acquire(&inode->i_flctx);
}
locks: add a new struct file_locking_context pointer to struct inode The current scheme of using the i_flock list is really difficult to manage. There is also a legitimate desire for a per-inode spinlock to manage these lists that isn't the i_lock. Start conversion to a new scheme to eventually replace the old i_flock list with a new "file_lock_context" object. We start by adding a new i_flctx to struct inode. For now, it lives in parallel with i_flock list, but will eventually replace it. The idea is to allocate a structure to sit in that pointer and act as a locus for all things file locking. We allocate a file_lock_context for an inode when the first lock is added to it, and it's only freed when the inode is freed. We use the i_lock to protect the assignment, but afterward it should mostly be accessed locklessly. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:54 +07:00
out:
locks: sprinkle some tracepoints around the file locking code Add some tracepoints around the POSIX locking code. These were useful when tracking down problems when handling the race between setlk and close. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Acked-by: "J. Bruce Fields" <bfields@fieldses.org>
2016-01-07 09:26:10 +07:00
trace_locks_get_lock_context(inode, type, ctx);
fs: fix data races on inode->i_flctx locks_get_lock_context() uses cmpxchg() to install i_flctx. cmpxchg() is a release operation which is correct. But it uses a plain load to load i_flctx. This is incorrect. Subsequent loads from i_flctx can hoist above the load of i_flctx pointer itself and observe uninitialized garbage there. This in turn can lead to corruption of ctx->flc_lock and other members. Documentation/memory-barriers.txt explicitly requires to use a barrier in such context: "A load-load control dependency requires a full read memory barrier". Use smp_load_acquire() in locks_get_lock_context() and in bunch of other functions that can proceed concurrently with locks_get_lock_context(). The data race was found with KernelThreadSanitizer (KTSAN). Signed-off-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-09-21 14:43:06 +07:00
return ctx;
locks: add a new struct file_locking_context pointer to struct inode The current scheme of using the i_flock list is really difficult to manage. There is also a legitimate desire for a per-inode spinlock to manage these lists that isn't the i_lock. Start conversion to a new scheme to eventually replace the old i_flock list with a new "file_lock_context" object. We start by adding a new i_flctx to struct inode. For now, it lives in parallel with i_flock list, but will eventually replace it. The idea is to allocate a structure to sit in that pointer and act as a locus for all things file locking. We allocate a file_lock_context for an inode when the first lock is added to it, and it's only freed when the inode is freed. We use the i_lock to protect the assignment, but afterward it should mostly be accessed locklessly. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:54 +07:00
}
locks: prink more detail when there are leaked locks Right now, we just get WARN_ON_ONCE, which is not particularly helpful. Have it dump some info about the locks and the inode to make it easier to track down leaked locks in the future. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Acked-by: "J. Bruce Fields" <bfields@fieldses.org>
2016-01-07 09:28:41 +07:00
static void
locks_dump_ctx_list(struct list_head *list, char *list_type)
{
struct file_lock *fl;
list_for_each_entry(fl, list, fl_list) {
pr_warn("%s: fl_owner=%p fl_flags=0x%x fl_type=0x%x fl_pid=%u\n", list_type, fl->fl_owner, fl->fl_flags, fl->fl_type, fl->fl_pid);
}
}
static void
locks_check_ctx_lists(struct inode *inode)
{
struct file_lock_context *ctx = inode->i_flctx;
if (unlikely(!list_empty(&ctx->flc_flock) ||
!list_empty(&ctx->flc_posix) ||
!list_empty(&ctx->flc_lease))) {
pr_warn("Leaked locks on dev=0x%x:0x%x ino=0x%lx:\n",
MAJOR(inode->i_sb->s_dev), MINOR(inode->i_sb->s_dev),
inode->i_ino);
locks_dump_ctx_list(&ctx->flc_flock, "FLOCK");
locks_dump_ctx_list(&ctx->flc_posix, "POSIX");
locks_dump_ctx_list(&ctx->flc_lease, "LEASE");
}
}
locks: restore a warn for leaked locks on close When locks.c moved to using file_lock_context, the check for any locks that were not released was moved from the __fput() to destroy_inode() path in commit 8634b51f6ca2 ("locks: convert lease handling to file_lock_context"). This warning has been quite useful for catching bugs, particularly in NFS where lock handling still sees some churn. Let's bring back the warning for leaked locks on __fput, as this warning is much more likely to be seen and reported by users. Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-07-22 00:36:25 +07:00
static void
locks_check_ctx_file_list(struct file *filp, struct list_head *list,
char *list_type)
{
struct file_lock *fl;
struct inode *inode = locks_inode(filp);
list_for_each_entry(fl, list, fl_list)
if (fl->fl_file == filp)
pr_warn("Leaked %s lock on dev=0x%x:0x%x ino=0x%lx "
" fl_owner=%p fl_flags=0x%x fl_type=0x%x fl_pid=%u\n",
list_type, MAJOR(inode->i_sb->s_dev),
MINOR(inode->i_sb->s_dev), inode->i_ino,
fl->fl_owner, fl->fl_flags, fl->fl_type, fl->fl_pid);
}
locks: add a new struct file_locking_context pointer to struct inode The current scheme of using the i_flock list is really difficult to manage. There is also a legitimate desire for a per-inode spinlock to manage these lists that isn't the i_lock. Start conversion to a new scheme to eventually replace the old i_flock list with a new "file_lock_context" object. We start by adding a new i_flctx to struct inode. For now, it lives in parallel with i_flock list, but will eventually replace it. The idea is to allocate a structure to sit in that pointer and act as a locus for all things file locking. We allocate a file_lock_context for an inode when the first lock is added to it, and it's only freed when the inode is freed. We use the i_lock to protect the assignment, but afterward it should mostly be accessed locklessly. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:54 +07:00
void
locks: pass inode pointer to locks_free_lock_context ...so we can print information about it if there are leaked locks. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Acked-by: "J. Bruce Fields" <bfields@fieldses.org>
2016-01-08 03:08:51 +07:00
locks_free_lock_context(struct inode *inode)
locks: add a new struct file_locking_context pointer to struct inode The current scheme of using the i_flock list is really difficult to manage. There is also a legitimate desire for a per-inode spinlock to manage these lists that isn't the i_lock. Start conversion to a new scheme to eventually replace the old i_flock list with a new "file_lock_context" object. We start by adding a new i_flctx to struct inode. For now, it lives in parallel with i_flock list, but will eventually replace it. The idea is to allocate a structure to sit in that pointer and act as a locus for all things file locking. We allocate a file_lock_context for an inode when the first lock is added to it, and it's only freed when the inode is freed. We use the i_lock to protect the assignment, but afterward it should mostly be accessed locklessly. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:54 +07:00
{
locks: pass inode pointer to locks_free_lock_context ...so we can print information about it if there are leaked locks. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Acked-by: "J. Bruce Fields" <bfields@fieldses.org>
2016-01-08 03:08:51 +07:00
struct file_lock_context *ctx = inode->i_flctx;
locks: prink more detail when there are leaked locks Right now, we just get WARN_ON_ONCE, which is not particularly helpful. Have it dump some info about the locks and the inode to make it easier to track down leaked locks in the future. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Acked-by: "J. Bruce Fields" <bfields@fieldses.org>
2016-01-07 09:28:41 +07:00
if (unlikely(ctx)) {
locks_check_ctx_lists(inode);
locks: add a new struct file_locking_context pointer to struct inode The current scheme of using the i_flock list is really difficult to manage. There is also a legitimate desire for a per-inode spinlock to manage these lists that isn't the i_lock. Start conversion to a new scheme to eventually replace the old i_flock list with a new "file_lock_context" object. We start by adding a new i_flctx to struct inode. For now, it lives in parallel with i_flock list, but will eventually replace it. The idea is to allocate a structure to sit in that pointer and act as a locus for all things file locking. We allocate a file_lock_context for an inode when the first lock is added to it, and it's only freed when the inode is freed. We use the i_lock to protect the assignment, but afterward it should mostly be accessed locklessly. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:54 +07:00
kmem_cache_free(flctx_cache, ctx);
}
}
fs: locks: remove init_once From: Miklos Szeredi <mszeredi@suse.cz> Remove SLAB initialization entirely, as suggested by Bruce and Linus. Allocate with __GFP_ZERO instead and only initialize list heads. Signed-off-by: Miklos Szeredi <mszeredi@suse.cz> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-07 18:06:09 +07:00
static void locks_init_lock_heads(struct file_lock *fl)
fs: fix lock initialization locks_alloc_lock() assumed that the allocated struct file_lock is already initialized to zero members. This is only true for the first allocation of the structure, after reuse some of the members will have random values. This will for example result in passing random fl_start values to userspace in fuse for FL_FLOCK locks, which is an information leak at best. Fix by reinitializing those members which may be non-zero after freeing. Signed-off-by: Miklos Szeredi <mszeredi@suse.cz> CC: stable@kernel.org Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-07-06 17:33:55 +07:00
{
locks: convert fl_link to a hlist_node Testing has shown that iterating over the blocked_list for deadlock detection turns out to be a bottleneck. In order to alleviate that, begin the process of turning it into a hashtable. We start by turning the fl_link into a hlist_node and the global lists into hlists. A later patch will do the conversion of the blocked_list to a hashtable. Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:17 +07:00
INIT_HLIST_NODE(&fl->fl_link);
locks: add new struct list_head to struct file_lock ...that we can use to queue file_locks to per-ctx list_heads. Go ahead and convert locks_delete_lock and locks_dispose_list to use it instead of the fl_block list. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:54 +07:00
INIT_LIST_HEAD(&fl->fl_list);
fs/locks: rename some lists and pointers. struct file lock contains an 'fl_next' pointer which is used to point to the lock that this request is blocked waiting for. So rename it to fl_blocker. The fl_blocked list_head in an active lock is the head of a list of blocked requests. In a request it is a node in that list. These are two distinct uses, so replace with two list_heads with different names. fl_blocked_requests is the head of a list of blocked requests fl_blocked_member is a node in a member of that list. The two different list_heads are never used at the same time, but that will change in a future patch. Note that a tracepoint is changed to report fl_blocker instead of fl_next. Signed-off-by: NeilBrown <neilb@suse.com> Reviewed-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Jeff Layton <jlayton@kernel.org>
2018-11-30 06:04:08 +07:00
INIT_LIST_HEAD(&fl->fl_blocked_requests);
INIT_LIST_HEAD(&fl->fl_blocked_member);
fs: locks: remove init_once From: Miklos Szeredi <mszeredi@suse.cz> Remove SLAB initialization entirely, as suggested by Bruce and Linus. Allocate with __GFP_ZERO instead and only initialize list heads. Signed-off-by: Miklos Szeredi <mszeredi@suse.cz> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-07 18:06:09 +07:00
init_waitqueue_head(&fl->fl_wait);
fs: fix lock initialization locks_alloc_lock() assumed that the allocated struct file_lock is already initialized to zero members. This is only true for the first allocation of the structure, after reuse some of the members will have random values. This will for example result in passing random fl_start values to userspace in fuse for FL_FLOCK locks, which is an information leak at best. Fix by reinitializing those members which may be non-zero after freeing. Signed-off-by: Miklos Szeredi <mszeredi@suse.cz> CC: stable@kernel.org Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-07-06 17:33:55 +07:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
/* Allocate an empty lock structure. */
locks/nfsd: allocate file lock outside of spinlock As suggested by Christoph Hellwig, this moves allocation of new file locks out of generic_setlease into the callers, nfs4_open_delegation and fcntl_setlease in order to allow GFP_KERNEL allocations when lock_flocks has become a spinlock. Signed-off-by: Arnd Bergmann <arnd@arndb.de> Acked-by: J. Bruce Fields <bfields@redhat.com>
2010-10-27 20:46:08 +07:00
struct file_lock *locks_alloc_lock(void)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
fs: locks: remove init_once From: Miklos Szeredi <mszeredi@suse.cz> Remove SLAB initialization entirely, as suggested by Bruce and Linus. Allocate with __GFP_ZERO instead and only initialize list heads. Signed-off-by: Miklos Szeredi <mszeredi@suse.cz> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-07 18:06:09 +07:00
struct file_lock *fl = kmem_cache_zalloc(filelock_cache, GFP_KERNEL);
fs: fix lock initialization locks_alloc_lock() assumed that the allocated struct file_lock is already initialized to zero members. This is only true for the first allocation of the structure, after reuse some of the members will have random values. This will for example result in passing random fl_start values to userspace in fuse for FL_FLOCK locks, which is an information leak at best. Fix by reinitializing those members which may be non-zero after freeing. Signed-off-by: Miklos Szeredi <mszeredi@suse.cz> CC: stable@kernel.org Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-07-06 17:33:55 +07:00
if (fl)
fs: locks: remove init_once From: Miklos Szeredi <mszeredi@suse.cz> Remove SLAB initialization entirely, as suggested by Bruce and Linus. Allocate with __GFP_ZERO instead and only initialize list heads. Signed-off-by: Miklos Szeredi <mszeredi@suse.cz> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-07 18:06:09 +07:00
locks_init_lock_heads(fl);
fs: fix lock initialization locks_alloc_lock() assumed that the allocated struct file_lock is already initialized to zero members. This is only true for the first allocation of the structure, after reuse some of the members will have random values. This will for example result in passing random fl_start values to userspace in fuse for FL_FLOCK locks, which is an information leak at best. Fix by reinitializing those members which may be non-zero after freeing. Signed-off-by: Miklos Szeredi <mszeredi@suse.cz> CC: stable@kernel.org Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-07-06 17:33:55 +07:00
return fl;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
locks/nfsd: allocate file lock outside of spinlock As suggested by Christoph Hellwig, this moves allocation of new file locks out of generic_setlease into the callers, nfs4_open_delegation and fcntl_setlease in order to allow GFP_KERNEL allocations when lock_flocks has become a spinlock. Signed-off-by: Arnd Bergmann <arnd@arndb.de> Acked-by: J. Bruce Fields <bfields@redhat.com>
2010-10-27 20:46:08 +07:00
EXPORT_SYMBOL_GPL(locks_alloc_lock);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
lockd: call locks_release_private to cleanup per-filesystem state For every lock request lockd creates a new file_lock object in nlmsvc_setgrantargs() by copying the passed in file_lock with locks_copy_lock(). A filesystem can attach it's own lock_operations vector to the file_lock. It has to be cleaned up at the end of the file_lock's life. However, lockd doesn't do it today, yet it asserts in nlmclnt_release_lockargs() that the per-filesystem state is clean. This patch fixes it by exporting locks_release_private() and adding it to nlmsvc_freegrantargs(), to be symmetrical to creating a file_lock in nlmsvc_setgrantargs(). Signed-off-by: Felix Blyakher <felixb@sgi.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-04-01 03:12:56 +07:00
void locks_release_private(struct file_lock *fl)
VFS: Fix __posix_lock_file() copy of private lock area The struct file_lock->fl_u area must be copied using the fl_copy_lock() operation. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-03-21 01:44:05 +07:00
{
if (fl->fl_ops) {
if (fl->fl_ops->fl_release_private)
fl->fl_ops->fl_release_private(fl);
fl->fl_ops = NULL;
}
locks: New ops in lock_manager_operations for get/put owner NFSD or other lockmanager may increase the owner's reference, so adds two new options for copying and releasing owner. v5: change order from 2/6 to 3/6 v4: rename lm_copy_owner/lm_release_owner to lm_get_owner/lm_put_owner Reviewed-by: Jeff Layton <jlayton@primarydata.com> Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-08-22 21:18:43 +07:00
if (fl->fl_lmops) {
locks: change lm_get_owner and lm_put_owner prototypes The current prototypes for these operations are somewhat awkward as they deal with fl_owners but take struct file_lock arguments. In the future, we'll want to be able to take references without necessarily dealing with a struct file_lock. Change them to take fl_owner_t arguments instead and have the callers deal with assigning the values to the file_lock structs. Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2015-04-03 20:04:04 +07:00
if (fl->fl_lmops->lm_put_owner) {
fl->fl_lmops->lm_put_owner(fl->fl_owner);
fl->fl_owner = NULL;
}
locks: New ops in lock_manager_operations for get/put owner NFSD or other lockmanager may increase the owner's reference, so adds two new options for copying and releasing owner. v5: change order from 2/6 to 3/6 v4: rename lm_copy_owner/lm_release_owner to lm_get_owner/lm_put_owner Reviewed-by: Jeff Layton <jlayton@primarydata.com> Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-08-22 21:18:43 +07:00
fl->fl_lmops = NULL;
}
VFS: Fix __posix_lock_file() copy of private lock area The struct file_lock->fl_u area must be copied using the fl_copy_lock() operation. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-03-21 01:44:05 +07:00
}
lockd: call locks_release_private to cleanup per-filesystem state For every lock request lockd creates a new file_lock object in nlmsvc_setgrantargs() by copying the passed in file_lock with locks_copy_lock(). A filesystem can attach it's own lock_operations vector to the file_lock. It has to be cleaned up at the end of the file_lock's life. However, lockd doesn't do it today, yet it asserts in nlmclnt_release_lockargs() that the per-filesystem state is clean. This patch fixes it by exporting locks_release_private() and adding it to nlmsvc_freegrantargs(), to be symmetrical to creating a file_lock in nlmsvc_setgrantargs(). Signed-off-by: Felix Blyakher <felixb@sgi.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2009-04-01 03:12:56 +07:00
EXPORT_SYMBOL_GPL(locks_release_private);
VFS: Fix __posix_lock_file() copy of private lock area The struct file_lock->fl_u area must be copied using the fl_copy_lock() operation. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-03-21 01:44:05 +07:00
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
/* Free a lock which is not in use. */
locks: fix setlease methods to free passed-in lock We modified setlease to require the caller to allocate the new lease in the case of creating a new lease, but forgot to fix up the filesystem methods. Cc: Steven Whitehouse <swhiteho@redhat.com> Cc: Steve French <sfrench@samba.org> Cc: Trond Myklebust <Trond.Myklebust@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com> Acked-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-10-31 04:31:15 +07:00
void locks_free_lock(struct file_lock *fl)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
[PATCH] locks: don't panic Don't panic! Just BUG_ON(). Signed-off-by: Miklos Szeredi <miklos@szeredi.hu> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-03-31 17:30:29 +07:00
BUG_ON(waitqueue_active(&fl->fl_wait));
locks: add new struct list_head to struct file_lock ...that we can use to queue file_locks to per-ctx list_heads. Go ahead and convert locks_delete_lock and locks_dispose_list to use it instead of the fl_block list. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:54 +07:00
BUG_ON(!list_empty(&fl->fl_list));
fs/locks: rename some lists and pointers. struct file lock contains an 'fl_next' pointer which is used to point to the lock that this request is blocked waiting for. So rename it to fl_blocker. The fl_blocked list_head in an active lock is the head of a list of blocked requests. In a request it is a node in that list. These are two distinct uses, so replace with two list_heads with different names. fl_blocked_requests is the head of a list of blocked requests fl_blocked_member is a node in a member of that list. The two different list_heads are never used at the same time, but that will change in a future patch. Note that a tracepoint is changed to report fl_blocker instead of fl_next. Signed-off-by: NeilBrown <neilb@suse.com> Reviewed-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Jeff Layton <jlayton@kernel.org>
2018-11-30 06:04:08 +07:00
BUG_ON(!list_empty(&fl->fl_blocked_requests));
BUG_ON(!list_empty(&fl->fl_blocked_member));
locks: convert fl_link to a hlist_node Testing has shown that iterating over the blocked_list for deadlock detection turns out to be a bottleneck. In order to alleviate that, begin the process of turning it into a hashtable. We start by turning the fl_link into a hlist_node and the global lists into hlists. A later patch will do the conversion of the blocked_list to a hashtable. Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:17 +07:00
BUG_ON(!hlist_unhashed(&fl->fl_link));
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
VFS: Fix __posix_lock_file() copy of private lock area The struct file_lock->fl_u area must be copied using the fl_copy_lock() operation. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-03-21 01:44:05 +07:00
locks_release_private(fl);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
kmem_cache_free(filelock_cache, fl);
}
locks: fix setlease methods to free passed-in lock We modified setlease to require the caller to allocate the new lease in the case of creating a new lease, but forgot to fix up the filesystem methods. Cc: Steven Whitehouse <swhiteho@redhat.com> Cc: Steve French <sfrench@samba.org> Cc: Trond Myklebust <Trond.Myklebust@netapp.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com> Acked-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-10-31 04:31:15 +07:00
EXPORT_SYMBOL(locks_free_lock);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
locks: defer freeing locks in locks_delete_lock until after i_lock has been dropped In commit 72f98e72551fa (locks: turn lock_flocks into a spinlock), we moved from using the BKL to a global spinlock. With this change, we lost the ability to block in the fl_release_private operation. This is problematic for NFS (and probably some other filesystems as well). Add a new list_head argument to locks_delete_lock. If that argument is non-NULL, then queue any locks that we want to free to the list instead of freeing them. Then, add a new locks_dispose_list function that will walk such a list and call locks_free_lock on them after the i_lock has been dropped. Finally, change all of the callers of locks_delete_lock to pass in a list_head, except for lease_modify. That function can be called long after the i_lock has been acquired. Deferring the freeing of a lease after unlocking it in that function is non-trivial until we overhaul some of the spinlocking in the lease code. Currently though, no filesystem that sets fl_release_private supports leases, so this is not currently a problem. We'll eventually want to make the same change in the lease code, but it needs a lot more work before we can reasonably do so. Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-08-12 01:20:31 +07:00
static void
locks_dispose_list(struct list_head *dispose)
{
struct file_lock *fl;
while (!list_empty(dispose)) {
locks: add new struct list_head to struct file_lock ...that we can use to queue file_locks to per-ctx list_heads. Go ahead and convert locks_delete_lock and locks_dispose_list to use it instead of the fl_block list. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:54 +07:00
fl = list_first_entry(dispose, struct file_lock, fl_list);
list_del_init(&fl->fl_list);
locks: defer freeing locks in locks_delete_lock until after i_lock has been dropped In commit 72f98e72551fa (locks: turn lock_flocks into a spinlock), we moved from using the BKL to a global spinlock. With this change, we lost the ability to block in the fl_release_private operation. This is problematic for NFS (and probably some other filesystems as well). Add a new list_head argument to locks_delete_lock. If that argument is non-NULL, then queue any locks that we want to free to the list instead of freeing them. Then, add a new locks_dispose_list function that will walk such a list and call locks_free_lock on them after the i_lock has been dropped. Finally, change all of the callers of locks_delete_lock to pass in a list_head, except for lease_modify. That function can be called long after the i_lock has been acquired. Deferring the freeing of a lease after unlocking it in that function is non-trivial until we overhaul some of the spinlocking in the lease code. Currently though, no filesystem that sets fl_release_private supports leases, so this is not currently a problem. We'll eventually want to make the same change in the lease code, but it needs a lot more work before we can reasonably do so. Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-08-12 01:20:31 +07:00
locks_free_lock(fl);
}
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
void locks_init_lock(struct file_lock *fl)
{
fs: locks: remove init_once From: Miklos Szeredi <mszeredi@suse.cz> Remove SLAB initialization entirely, as suggested by Bruce and Linus. Allocate with __GFP_ZERO instead and only initialize list heads. Signed-off-by: Miklos Szeredi <mszeredi@suse.cz> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-07 18:06:09 +07:00
memset(fl, 0, sizeof(struct file_lock));
locks_init_lock_heads(fl);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
EXPORT_SYMBOL(locks_init_lock);
/*
* Initialize a new lock from an existing file_lock structure.
*/
locks: Rename __locks_copy_lock() to locks_copy_conflock() Jeff advice, " Right now __locks_copy_lock is only used to copy conflocks. It would be good to rename that to something more distinct (i.e.locks_copy_conflock), to make it clear that we're generating a conflock there." v5: change order from 3/6 to 2/6 v4: new patch only renaming function name Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-08-22 21:18:42 +07:00
void locks_copy_conflock(struct file_lock *new, struct file_lock *fl)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
new->fl_owner = fl->fl_owner;
new->fl_pid = fl->fl_pid;
lockd: posix_test_lock() should not call locks_copy_lock() The caller of posix_test_lock() should never need to look at the lock private data, so do not copy that information. This also means that there is no need to call the fl_release_private methods. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-03-21 01:44:38 +07:00
new->fl_file = NULL;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
new->fl_flags = fl->fl_flags;
new->fl_type = fl->fl_type;
new->fl_start = fl->fl_start;
new->fl_end = fl->fl_end;
locks: Copy fl_lmops information for conflock in locks_copy_conflock() Commit d5b9026a67 ([PATCH] knfsd: locks: flag NFSv4-owned locks) using fl_lmops field in file_lock for checking nfsd4 lockowner. But, commit 1a747ee0cc (locks: don't call ->copy_lock methods on return of conflicting locks) causes the fl_lmops of conflock always be NULL. Also, commit 0996905f93 (lockd: posix_test_lock() should not call locks_copy_lock()) caused the fl_lmops of conflock always be NULL too. Make sure copy the private information by fl_copy_lock() in struct file_lock_operations, merge __locks_copy_lock() to fl_copy_lock(). Jeff advice, "Set fl_lmops on conflocks, but don't set fl_ops. fl_ops are superfluous, since they are callbacks into the filesystem. There should be no need to bother the filesystem at all with info in a conflock. But, lock _ownership_ matters for conflocks and that's indicated by the fl_lmops. So you really do want to copy the fl_lmops for conflocks I think." v5: add missing calling of locks_release_private() in nlmsvc_testlock() v4: only copy fl_lmops for conflock, don't copy fl_ops Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-08-22 21:18:43 +07:00
new->fl_lmops = fl->fl_lmops;
lockd: posix_test_lock() should not call locks_copy_lock() The caller of posix_test_lock() should never need to look at the lock private data, so do not copy that information. This also means that there is no need to call the fl_release_private methods. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-03-21 01:44:38 +07:00
new->fl_ops = NULL;
locks: Copy fl_lmops information for conflock in locks_copy_conflock() Commit d5b9026a67 ([PATCH] knfsd: locks: flag NFSv4-owned locks) using fl_lmops field in file_lock for checking nfsd4 lockowner. But, commit 1a747ee0cc (locks: don't call ->copy_lock methods on return of conflicting locks) causes the fl_lmops of conflock always be NULL. Also, commit 0996905f93 (lockd: posix_test_lock() should not call locks_copy_lock()) caused the fl_lmops of conflock always be NULL too. Make sure copy the private information by fl_copy_lock() in struct file_lock_operations, merge __locks_copy_lock() to fl_copy_lock(). Jeff advice, "Set fl_lmops on conflocks, but don't set fl_ops. fl_ops are superfluous, since they are callbacks into the filesystem. There should be no need to bother the filesystem at all with info in a conflock. But, lock _ownership_ matters for conflocks and that's indicated by the fl_lmops. So you really do want to copy the fl_lmops for conflocks I think." v5: add missing calling of locks_release_private() in nlmsvc_testlock() v4: only copy fl_lmops for conflock, don't copy fl_ops Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-08-22 21:18:43 +07:00
if (fl->fl_lmops) {
if (fl->fl_lmops->lm_get_owner)
locks: change lm_get_owner and lm_put_owner prototypes The current prototypes for these operations are somewhat awkward as they deal with fl_owners but take struct file_lock arguments. In the future, we'll want to be able to take references without necessarily dealing with a struct file_lock. Change them to take fl_owner_t arguments instead and have the callers deal with assigning the values to the file_lock structs. Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2015-04-03 20:04:04 +07:00
fl->fl_lmops->lm_get_owner(fl->fl_owner);
locks: Copy fl_lmops information for conflock in locks_copy_conflock() Commit d5b9026a67 ([PATCH] knfsd: locks: flag NFSv4-owned locks) using fl_lmops field in file_lock for checking nfsd4 lockowner. But, commit 1a747ee0cc (locks: don't call ->copy_lock methods on return of conflicting locks) causes the fl_lmops of conflock always be NULL. Also, commit 0996905f93 (lockd: posix_test_lock() should not call locks_copy_lock()) caused the fl_lmops of conflock always be NULL too. Make sure copy the private information by fl_copy_lock() in struct file_lock_operations, merge __locks_copy_lock() to fl_copy_lock(). Jeff advice, "Set fl_lmops on conflocks, but don't set fl_ops. fl_ops are superfluous, since they are callbacks into the filesystem. There should be no need to bother the filesystem at all with info in a conflock. But, lock _ownership_ matters for conflocks and that's indicated by the fl_lmops. So you really do want to copy the fl_lmops for conflocks I think." v5: add missing calling of locks_release_private() in nlmsvc_testlock() v4: only copy fl_lmops for conflock, don't copy fl_ops Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-08-22 21:18:43 +07:00
}
lockd: posix_test_lock() should not call locks_copy_lock() The caller of posix_test_lock() should never need to look at the lock private data, so do not copy that information. This also means that there is no need to call the fl_release_private methods. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-03-21 01:44:38 +07:00
}
locks: Rename __locks_copy_lock() to locks_copy_conflock() Jeff advice, " Right now __locks_copy_lock is only used to copy conflocks. It would be good to rename that to something more distinct (i.e.locks_copy_conflock), to make it clear that we're generating a conflock there." v5: change order from 3/6 to 2/6 v4: new patch only renaming function name Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-08-22 21:18:42 +07:00
EXPORT_SYMBOL(locks_copy_conflock);
lockd: posix_test_lock() should not call locks_copy_lock() The caller of posix_test_lock() should never need to look at the lock private data, so do not copy that information. This also means that there is no need to call the fl_release_private methods. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-03-21 01:44:38 +07:00
void locks_copy_lock(struct file_lock *new, struct file_lock *fl)
{
locks: don't call locks_release_private from locks_copy_lock All callers of locks_copy_lock pass in a brand new file_lock struct, so there's no need to call locks_release_private on it. Replace that with a warning that fires in the event that we receive a target lock that doesn't look like it's properly initialized. Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-08-12 01:09:35 +07:00
/* "new" must be a freshly-initialized lock */
WARN_ON_ONCE(new->fl_ops);
lockd: posix_test_lock() should not call locks_copy_lock() The caller of posix_test_lock() should never need to look at the lock private data, so do not copy that information. This also means that there is no need to call the fl_release_private methods. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-03-21 01:44:38 +07:00
locks: Rename __locks_copy_lock() to locks_copy_conflock() Jeff advice, " Right now __locks_copy_lock is only used to copy conflocks. It would be good to rename that to something more distinct (i.e.locks_copy_conflock), to make it clear that we're generating a conflock there." v5: change order from 3/6 to 2/6 v4: new patch only renaming function name Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-08-22 21:18:42 +07:00
locks_copy_conflock(new, fl);
locks: Copy fl_lmops information for conflock in locks_copy_conflock() Commit d5b9026a67 ([PATCH] knfsd: locks: flag NFSv4-owned locks) using fl_lmops field in file_lock for checking nfsd4 lockowner. But, commit 1a747ee0cc (locks: don't call ->copy_lock methods on return of conflicting locks) causes the fl_lmops of conflock always be NULL. Also, commit 0996905f93 (lockd: posix_test_lock() should not call locks_copy_lock()) caused the fl_lmops of conflock always be NULL too. Make sure copy the private information by fl_copy_lock() in struct file_lock_operations, merge __locks_copy_lock() to fl_copy_lock(). Jeff advice, "Set fl_lmops on conflocks, but don't set fl_ops. fl_ops are superfluous, since they are callbacks into the filesystem. There should be no need to bother the filesystem at all with info in a conflock. But, lock _ownership_ matters for conflocks and that's indicated by the fl_lmops. So you really do want to copy the fl_lmops for conflocks I think." v5: add missing calling of locks_release_private() in nlmsvc_testlock() v4: only copy fl_lmops for conflock, don't copy fl_ops Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-08-22 21:18:43 +07:00
lockd: posix_test_lock() should not call locks_copy_lock() The caller of posix_test_lock() should never need to look at the lock private data, so do not copy that information. This also means that there is no need to call the fl_release_private methods. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-03-21 01:44:38 +07:00
new->fl_file = fl->fl_file;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
new->fl_ops = fl->fl_ops;
VFS: Fix __posix_lock_file() copy of private lock area The struct file_lock->fl_u area must be copied using the fl_copy_lock() operation. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-03-21 01:44:05 +07:00
locks: Copy fl_lmops information for conflock in locks_copy_conflock() Commit d5b9026a67 ([PATCH] knfsd: locks: flag NFSv4-owned locks) using fl_lmops field in file_lock for checking nfsd4 lockowner. But, commit 1a747ee0cc (locks: don't call ->copy_lock methods on return of conflicting locks) causes the fl_lmops of conflock always be NULL. Also, commit 0996905f93 (lockd: posix_test_lock() should not call locks_copy_lock()) caused the fl_lmops of conflock always be NULL too. Make sure copy the private information by fl_copy_lock() in struct file_lock_operations, merge __locks_copy_lock() to fl_copy_lock(). Jeff advice, "Set fl_lmops on conflocks, but don't set fl_ops. fl_ops are superfluous, since they are callbacks into the filesystem. There should be no need to bother the filesystem at all with info in a conflock. But, lock _ownership_ matters for conflocks and that's indicated by the fl_lmops. So you really do want to copy the fl_lmops for conflocks I think." v5: add missing calling of locks_release_private() in nlmsvc_testlock() v4: only copy fl_lmops for conflock, don't copy fl_ops Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-08-22 21:18:43 +07:00
if (fl->fl_ops) {
if (fl->fl_ops->fl_copy_lock)
fl->fl_ops->fl_copy_lock(new, fl);
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
EXPORT_SYMBOL(locks_copy_lock);
static inline int flock_translate_cmd(int cmd) {
if (cmd & LOCK_MAND)
return cmd & (LOCK_MAND | LOCK_RW);
switch (cmd) {
case LOCK_SH:
return F_RDLCK;
case LOCK_EX:
return F_WRLCK;
case LOCK_UN:
return F_UNLCK;
}
return -EINVAL;
}
/* Fill in a file_lock structure with an appropriate FLOCK lock. */
locks: flock_make_lock should return a struct file_lock (or PTR_ERR) Eliminate the need for a return pointer. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-09-04 21:25:06 +07:00
static struct file_lock *
flock_make_lock(struct file *filp, unsigned int cmd)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
struct file_lock *fl;
int type = flock_translate_cmd(cmd);
locks: flock_make_lock should return a struct file_lock (or PTR_ERR) Eliminate the need for a return pointer. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-09-04 21:25:06 +07:00
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
if (type < 0)
locks: flock_make_lock should return a struct file_lock (or PTR_ERR) Eliminate the need for a return pointer. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-09-04 21:25:06 +07:00
return ERR_PTR(type);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
fl = locks_alloc_lock();
if (fl == NULL)
locks: flock_make_lock should return a struct file_lock (or PTR_ERR) Eliminate the need for a return pointer. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-09-04 21:25:06 +07:00
return ERR_PTR(-ENOMEM);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
fl->fl_file = filp;
locks: purge fl_owner_t from fs/locks.c Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-07-13 22:00:38 +07:00
fl->fl_owner = filp;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
fl->fl_pid = current->tgid;
fl->fl_flags = FL_FLOCK;
fl->fl_type = type;
fl->fl_end = OFFSET_MAX;
locks: flock_make_lock should return a struct file_lock (or PTR_ERR) Eliminate the need for a return pointer. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-09-04 21:25:06 +07:00
return fl;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
locks: fix checking of fcntl_setlease argument The only checks of the long argument passed to fcntl(fd,F_SETLEASE,.) are done after converting the long to an int. Thus some illegal values may be let through and cause problems in later code. [ They actually *don't* cause problems in mainline, as of Dave Jones's commit 8d657eb3b438 "Remove easily user-triggerable BUG from generic_setlease", but we should fix this anyway. And this patch will be necessary to fix real bugs on earlier kernels. ] Cc: stable@vger.kernel.org Signed-off-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-07-24 02:17:17 +07:00
static int assign_type(struct file_lock *fl, long type)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
switch (type) {
case F_RDLCK:
case F_WRLCK:
case F_UNLCK:
fl->fl_type = type;
break;
default:
return -EINVAL;
}
return 0;
}
locks: fix posix lock range overflow handling In the 32-bit case fcntl assigns the 64-bit f_pos and i_size to a 32-bit off_t. The existing range checks also seem to depend on signed arithmetic wrapping when it overflows. In practice maybe that works, but we can be more careful. That also allows us to make a more reliable distinction between -EINVAL and -EOVERFLOW. Note that in the 32-bit case SEEK_CUR or SEEK_END might allow the caller to set a lock with starting point no longer representable as a 32-bit value. We could return -EOVERFLOW in such cases, but the locks code is capable of handling such ranges, so we choose to be lenient here. The only problem is that subsequent GETLK calls on such a lock will fail with EOVERFLOW. While we're here, do some cleanup including consolidating code for the flock and flock64 cases. Signed-off-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:08 +07:00
static int flock64_to_posix_lock(struct file *filp, struct file_lock *fl,
struct flock64 *l)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
switch (l->l_whence) {
[PATCH] VFS: Use SEEK_{SET, CUR, END} instead of hardcoded values VFS: Use SEEK_{SET,CUR,END} instead of hardcoded values Signed-off-by: Josef 'Jeff' Sipek <jeffpc@josefsipek.net> 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>
2006-10-01 13:27:35 +07:00
case SEEK_SET:
locks: fix posix lock range overflow handling In the 32-bit case fcntl assigns the 64-bit f_pos and i_size to a 32-bit off_t. The existing range checks also seem to depend on signed arithmetic wrapping when it overflows. In practice maybe that works, but we can be more careful. That also allows us to make a more reliable distinction between -EINVAL and -EOVERFLOW. Note that in the 32-bit case SEEK_CUR or SEEK_END might allow the caller to set a lock with starting point no longer representable as a 32-bit value. We could return -EOVERFLOW in such cases, but the locks code is capable of handling such ranges, so we choose to be lenient here. The only problem is that subsequent GETLK calls on such a lock will fail with EOVERFLOW. While we're here, do some cleanup including consolidating code for the flock and flock64 cases. Signed-off-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:08 +07:00
fl->fl_start = 0;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
break;
[PATCH] VFS: Use SEEK_{SET, CUR, END} instead of hardcoded values VFS: Use SEEK_{SET,CUR,END} instead of hardcoded values Signed-off-by: Josef 'Jeff' Sipek <jeffpc@josefsipek.net> 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>
2006-10-01 13:27:35 +07:00
case SEEK_CUR:
locks: fix posix lock range overflow handling In the 32-bit case fcntl assigns the 64-bit f_pos and i_size to a 32-bit off_t. The existing range checks also seem to depend on signed arithmetic wrapping when it overflows. In practice maybe that works, but we can be more careful. That also allows us to make a more reliable distinction between -EINVAL and -EOVERFLOW. Note that in the 32-bit case SEEK_CUR or SEEK_END might allow the caller to set a lock with starting point no longer representable as a 32-bit value. We could return -EOVERFLOW in such cases, but the locks code is capable of handling such ranges, so we choose to be lenient here. The only problem is that subsequent GETLK calls on such a lock will fail with EOVERFLOW. While we're here, do some cleanup including consolidating code for the flock and flock64 cases. Signed-off-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:08 +07:00
fl->fl_start = filp->f_pos;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
break;
[PATCH] VFS: Use SEEK_{SET, CUR, END} instead of hardcoded values VFS: Use SEEK_{SET,CUR,END} instead of hardcoded values Signed-off-by: Josef 'Jeff' Sipek <jeffpc@josefsipek.net> 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>
2006-10-01 13:27:35 +07:00
case SEEK_END:
locks: fix posix lock range overflow handling In the 32-bit case fcntl assigns the 64-bit f_pos and i_size to a 32-bit off_t. The existing range checks also seem to depend on signed arithmetic wrapping when it overflows. In practice maybe that works, but we can be more careful. That also allows us to make a more reliable distinction between -EINVAL and -EOVERFLOW. Note that in the 32-bit case SEEK_CUR or SEEK_END might allow the caller to set a lock with starting point no longer representable as a 32-bit value. We could return -EOVERFLOW in such cases, but the locks code is capable of handling such ranges, so we choose to be lenient here. The only problem is that subsequent GETLK calls on such a lock will fail with EOVERFLOW. While we're here, do some cleanup including consolidating code for the flock and flock64 cases. Signed-off-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:08 +07:00
fl->fl_start = i_size_read(file_inode(filp));
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
break;
default:
return -EINVAL;
}
locks: fix posix lock range overflow handling In the 32-bit case fcntl assigns the 64-bit f_pos and i_size to a 32-bit off_t. The existing range checks also seem to depend on signed arithmetic wrapping when it overflows. In practice maybe that works, but we can be more careful. That also allows us to make a more reliable distinction between -EINVAL and -EOVERFLOW. Note that in the 32-bit case SEEK_CUR or SEEK_END might allow the caller to set a lock with starting point no longer representable as a 32-bit value. We could return -EOVERFLOW in such cases, but the locks code is capable of handling such ranges, so we choose to be lenient here. The only problem is that subsequent GETLK calls on such a lock will fail with EOVERFLOW. While we're here, do some cleanup including consolidating code for the flock and flock64 cases. Signed-off-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:08 +07:00
if (l->l_start > OFFSET_MAX - fl->fl_start)
return -EOVERFLOW;
fl->fl_start += l->l_start;
if (fl->fl_start < 0)
return -EINVAL;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
/* POSIX-1996 leaves the case l->l_len < 0 undefined;
POSIX-2001 defines it. */
Fix Connectathon locking test failure We currently fail Connectathon test 6.10 in the case of 32-bit locks due to incorrect error checking. Also add support for l->l_len < 0 to 64-bit locks. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2005-10-19 04:20:21 +07:00
if (l->l_len > 0) {
locks: fix posix lock range overflow handling In the 32-bit case fcntl assigns the 64-bit f_pos and i_size to a 32-bit off_t. The existing range checks also seem to depend on signed arithmetic wrapping when it overflows. In practice maybe that works, but we can be more careful. That also allows us to make a more reliable distinction between -EINVAL and -EOVERFLOW. Note that in the 32-bit case SEEK_CUR or SEEK_END might allow the caller to set a lock with starting point no longer representable as a 32-bit value. We could return -EOVERFLOW in such cases, but the locks code is capable of handling such ranges, so we choose to be lenient here. The only problem is that subsequent GETLK calls on such a lock will fail with EOVERFLOW. While we're here, do some cleanup including consolidating code for the flock and flock64 cases. Signed-off-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:08 +07:00
if (l->l_len - 1 > OFFSET_MAX - fl->fl_start)
return -EOVERFLOW;
fl->fl_end = fl->fl_start + l->l_len - 1;
Fix Connectathon locking test failure We currently fail Connectathon test 6.10 in the case of 32-bit locks due to incorrect error checking. Also add support for l->l_len < 0 to 64-bit locks. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2005-10-19 04:20:21 +07:00
} else if (l->l_len < 0) {
locks: fix posix lock range overflow handling In the 32-bit case fcntl assigns the 64-bit f_pos and i_size to a 32-bit off_t. The existing range checks also seem to depend on signed arithmetic wrapping when it overflows. In practice maybe that works, but we can be more careful. That also allows us to make a more reliable distinction between -EINVAL and -EOVERFLOW. Note that in the 32-bit case SEEK_CUR or SEEK_END might allow the caller to set a lock with starting point no longer representable as a 32-bit value. We could return -EOVERFLOW in such cases, but the locks code is capable of handling such ranges, so we choose to be lenient here. The only problem is that subsequent GETLK calls on such a lock will fail with EOVERFLOW. While we're here, do some cleanup including consolidating code for the flock and flock64 cases. Signed-off-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:08 +07:00
if (fl->fl_start + l->l_len < 0)
Fix Connectathon locking test failure We currently fail Connectathon test 6.10 in the case of 32-bit locks due to incorrect error checking. Also add support for l->l_len < 0 to 64-bit locks. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2005-10-19 04:20:21 +07:00
return -EINVAL;
locks: fix posix lock range overflow handling In the 32-bit case fcntl assigns the 64-bit f_pos and i_size to a 32-bit off_t. The existing range checks also seem to depend on signed arithmetic wrapping when it overflows. In practice maybe that works, but we can be more careful. That also allows us to make a more reliable distinction between -EINVAL and -EOVERFLOW. Note that in the 32-bit case SEEK_CUR or SEEK_END might allow the caller to set a lock with starting point no longer representable as a 32-bit value. We could return -EOVERFLOW in such cases, but the locks code is capable of handling such ranges, so we choose to be lenient here. The only problem is that subsequent GETLK calls on such a lock will fail with EOVERFLOW. While we're here, do some cleanup including consolidating code for the flock and flock64 cases. Signed-off-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:08 +07:00
fl->fl_end = fl->fl_start - 1;
fl->fl_start += l->l_len;
} else
fl->fl_end = OFFSET_MAX;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
fl->fl_owner = current->files;
fl->fl_pid = current->tgid;
fl->fl_file = filp;
fl->fl_flags = FL_POSIX;
fl->fl_ops = NULL;
fl->fl_lmops = NULL;
return assign_type(fl, l->l_type);
}
locks: fix posix lock range overflow handling In the 32-bit case fcntl assigns the 64-bit f_pos and i_size to a 32-bit off_t. The existing range checks also seem to depend on signed arithmetic wrapping when it overflows. In practice maybe that works, but we can be more careful. That also allows us to make a more reliable distinction between -EINVAL and -EOVERFLOW. Note that in the 32-bit case SEEK_CUR or SEEK_END might allow the caller to set a lock with starting point no longer representable as a 32-bit value. We could return -EOVERFLOW in such cases, but the locks code is capable of handling such ranges, so we choose to be lenient here. The only problem is that subsequent GETLK calls on such a lock will fail with EOVERFLOW. While we're here, do some cleanup including consolidating code for the flock and flock64 cases. Signed-off-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:08 +07:00
/* Verify a "struct flock" and copy it to a "struct file_lock" as a POSIX
* style lock.
*/
static int flock_to_posix_lock(struct file *filp, struct file_lock *fl,
struct flock *l)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
locks: fix posix lock range overflow handling In the 32-bit case fcntl assigns the 64-bit f_pos and i_size to a 32-bit off_t. The existing range checks also seem to depend on signed arithmetic wrapping when it overflows. In practice maybe that works, but we can be more careful. That also allows us to make a more reliable distinction between -EINVAL and -EOVERFLOW. Note that in the 32-bit case SEEK_CUR or SEEK_END might allow the caller to set a lock with starting point no longer representable as a 32-bit value. We could return -EOVERFLOW in such cases, but the locks code is capable of handling such ranges, so we choose to be lenient here. The only problem is that subsequent GETLK calls on such a lock will fail with EOVERFLOW. While we're here, do some cleanup including consolidating code for the flock and flock64 cases. Signed-off-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:08 +07:00
struct flock64 ll = {
.l_type = l->l_type,
.l_whence = l->l_whence,
.l_start = l->l_start,
.l_len = l->l_len,
};
return flock64_to_posix_lock(filp, fl, &ll);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
/* default lease lock manager operations */
locks: give lm_break a return value Christoph suggests: "Add a return value to lm_break so that the lock manager can tell the core code "you can delete this lease right now". That gets rid of the games with the timeout which require all kinds of race avoidance code in the users." Do that here and have the nfsd lease break routine use it when it detects that there was a race between setting up the lease and it being broken. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-09-02 02:06:54 +07:00
static bool
lease_break_callback(struct file_lock *fl)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
kill_fasync(&fl->fl_fasync, SIGIO, POLL_MSG);
locks: give lm_break a return value Christoph suggests: "Add a return value to lm_break so that the lock manager can tell the core code "you can delete this lease right now". That gets rid of the games with the timeout which require all kinds of race avoidance code in the users." Do that here and have the nfsd lease break routine use it when it detects that there was a race between setting up the lease and it being broken. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-09-02 02:06:54 +07:00
return false;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
locks: define a lm_setup handler for leases ...and move the fasync setup into it for fcntl lease calls. At the same time, change the semantics of how the file_lock double-pointer is handled. Up until now, on a successful lease return you got a pointer to the lock on the list. This is bad, since that pointer can no longer be relied on as valid once the inode->i_lock has been released. Change the code to instead just zero out the pointer if the lease we passed in ended up being used. Then the callers can just check to see if it's NULL after the call and free it if it isn't. The priv argument has the same semantics. The lm_setup function can zero the pointer out to signal to the caller that it should not be freed after the function returns. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-08-22 21:55:47 +07:00
static void
lease_setup(struct file_lock *fl, void **priv)
{
struct file *filp = fl->fl_file;
struct fasync_struct *fa = *priv;
/*
* fasync_insert_entry() returns the old entry if any. If there was no
* old entry, then it used "priv" and inserted it into the fasync list.
* Clear the pointer to indicate that it shouldn't be freed.
*/
if (!fasync_insert_entry(fa->fa_fd, filp, &fl->fl_fasync, fa))
*priv = NULL;
signal: Use PIDTYPE_TGID to clearly store where file signals will be sent When f_setown is called a pid and a pid type are stored. Replace the use of PIDTYPE_PID with PIDTYPE_TGID as PIDTYPE_TGID goes to the entire thread group. Replace the use of PIDTYPE_MAX with PIDTYPE_PID as PIDTYPE_PID now is only for a thread. Update the users of __f_setown to use PIDTYPE_TGID instead of PIDTYPE_PID. For now the code continues to capture task_pid (when task_tgid would really be appropriate), and iterate on PIDTYPE_PID (even when type == PIDTYPE_TGID) out of an abundance of caution to preserve existing behavior. Oleg Nesterov suggested using the test to ensure we use PIDTYPE_PID for tgid lookup also be used to avoid taking the tasklist lock. Suggested-by: Oleg Nesterov <oleg@redhat.com> Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
2017-07-17 10:05:57 +07:00
__f_setown(filp, task_pid(current), PIDTYPE_TGID, 0);
locks: define a lm_setup handler for leases ...and move the fasync setup into it for fcntl lease calls. At the same time, change the semantics of how the file_lock double-pointer is handled. Up until now, on a successful lease return you got a pointer to the lock on the list. This is bad, since that pointer can no longer be relied on as valid once the inode->i_lock has been released. Change the code to instead just zero out the pointer if the lease we passed in ended up being used. Then the callers can just check to see if it's NULL after the call and free it if it isn't. The priv argument has the same semantics. The lm_setup function can zero the pointer out to signal to the caller that it should not be freed after the function returns. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-08-22 21:55:47 +07:00
}
const: make lock_manager_operations const Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-09-22 07:01:12 +07:00
static const struct lock_manager_operations lease_manager_ops = {
locks: rename lock-manager ops 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: J. Bruce Fields <bfields@redhat.com>
2011-07-21 07:21:59 +07:00
.lm_break = lease_break_callback,
.lm_change = lease_modify,
locks: define a lm_setup handler for leases ...and move the fasync setup into it for fcntl lease calls. At the same time, change the semantics of how the file_lock double-pointer is handled. Up until now, on a successful lease return you got a pointer to the lock on the list. This is bad, since that pointer can no longer be relied on as valid once the inode->i_lock has been released. Change the code to instead just zero out the pointer if the lease we passed in ended up being used. Then the callers can just check to see if it's NULL after the call and free it if it isn't. The priv argument has the same semantics. The lm_setup function can zero the pointer out to signal to the caller that it should not be freed after the function returns. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-08-22 21:55:47 +07:00
.lm_setup = lease_setup,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
};
/*
* Initialize a lease, use the default lock manager operations
*/
locks: fix checking of fcntl_setlease argument The only checks of the long argument passed to fcntl(fd,F_SETLEASE,.) are done after converting the long to an int. Thus some illegal values may be let through and cause problems in later code. [ They actually *don't* cause problems in mainline, as of Dave Jones's commit 8d657eb3b438 "Remove easily user-triggerable BUG from generic_setlease", but we should fix this anyway. And this patch will be necessary to fix real bugs on earlier kernels. ] Cc: stable@vger.kernel.org Signed-off-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-07-24 02:17:17 +07:00
static int lease_init(struct file *filp, long type, struct file_lock *fl)
treewide: Align function definition open/close braces Some functions definitions have either the initial open brace and/or the closing brace outside of column 1. Move those braces to column 1. This allows various function analyzers like gnu complexity to work properly for these modified functions. Signed-off-by: Joe Perches <joe@perches.com> Acked-by: Andy Shevchenko <andy.shevchenko@gmail.com> Acked-by: Paul Moore <paul@paul-moore.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Acked-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com> Acked-by: Alexandre Belloni <alexandre.belloni@free-electrons.com> Acked-by: Martin K. Petersen <martin.petersen@oracle.com> Acked-by: Takashi Iwai <tiwai@suse.de> Acked-by: Mauro Carvalho Chehab <mchehab@s-opensource.com> Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Nicolin Chen <nicoleotsuka@gmail.com> Acked-by: Martin K. Petersen <martin.petersen@oracle.com> Acked-by: Steven Rostedt (VMware) <rostedt@goodmis.org> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2018-03-22 05:09:32 +07:00
{
[PATCH] fs/locks.c: Fix lease_init It is insane to be giving lease_init() the task of freeing the lock it is supposed to initialise, given that the lock is not guaranteed to be allocated on the stack. This causes lockups in fcntl_setlease(). Problem diagnosed by Daniel Hokka Zakrisson <daniel@hozac.com> Also fix a slab leak in __setlease() due to an uninitialised return value. Problem diagnosed by Björn Steinbrink. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com> Tested-by: Daniel Hokka Zakrisson <daniel@hozac.com> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-05-08 10:02:42 +07:00
if (assign_type(fl, type) != 0)
return -EINVAL;
locks: set fl_owner for leases to filp instead of current->files Like flock locks, leases are owned by the file description. Now that the i_have_this_lease check in __break_lease is gone, we don't actually use the fl_owner for leases for anything. So, it's now safe to set this more appropriately to the same value as the fl_file. While we're at it, fix up the comments over the fl_owner_t definition since they're rather out of date. Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-09-02 06:04:48 +07:00
fl->fl_owner = filp;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
fl->fl_pid = current->tgid;
fl->fl_file = filp;
fl->fl_flags = FL_LEASE;
fl->fl_start = 0;
fl->fl_end = OFFSET_MAX;
fl->fl_ops = NULL;
fl->fl_lmops = &lease_manager_ops;
return 0;
}
/* Allocate a file_lock initialised to this type of lease */
locks: fix checking of fcntl_setlease argument The only checks of the long argument passed to fcntl(fd,F_SETLEASE,.) are done after converting the long to an int. Thus some illegal values may be let through and cause problems in later code. [ They actually *don't* cause problems in mainline, as of Dave Jones's commit 8d657eb3b438 "Remove easily user-triggerable BUG from generic_setlease", but we should fix this anyway. And this patch will be necessary to fix real bugs on earlier kernels. ] Cc: stable@vger.kernel.org Signed-off-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-07-24 02:17:17 +07:00
static struct file_lock *lease_alloc(struct file *filp, long type)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
struct file_lock *fl = locks_alloc_lock();
[PATCH] fs/locks.c: Fix lease_init It is insane to be giving lease_init() the task of freeing the lock it is supposed to initialise, given that the lock is not guaranteed to be allocated on the stack. This causes lockups in fcntl_setlease(). Problem diagnosed by Daniel Hokka Zakrisson <daniel@hozac.com> Also fix a slab leak in __setlease() due to an uninitialised return value. Problem diagnosed by Björn Steinbrink. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com> Tested-by: Daniel Hokka Zakrisson <daniel@hozac.com> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-05-08 10:02:42 +07:00
int error = -ENOMEM;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
if (fl == NULL)
locks: clean up lease_alloc() Return the newly allocated structure as the return value instead of using a struct ** parameter. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2007-03-02 02:34:35 +07:00
return ERR_PTR(error);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
error = lease_init(filp, type, fl);
[PATCH] fs/locks.c: Fix lease_init It is insane to be giving lease_init() the task of freeing the lock it is supposed to initialise, given that the lock is not guaranteed to be allocated on the stack. This causes lockups in fcntl_setlease(). Problem diagnosed by Daniel Hokka Zakrisson <daniel@hozac.com> Also fix a slab leak in __setlease() due to an uninitialised return value. Problem diagnosed by Björn Steinbrink. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com> Tested-by: Daniel Hokka Zakrisson <daniel@hozac.com> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-05-08 10:02:42 +07:00
if (error) {
locks_free_lock(fl);
locks: clean up lease_alloc() Return the newly allocated structure as the return value instead of using a struct ** parameter. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2007-03-02 02:34:35 +07:00
return ERR_PTR(error);
[PATCH] fs/locks.c: Fix lease_init It is insane to be giving lease_init() the task of freeing the lock it is supposed to initialise, given that the lock is not guaranteed to be allocated on the stack. This causes lockups in fcntl_setlease(). Problem diagnosed by Daniel Hokka Zakrisson <daniel@hozac.com> Also fix a slab leak in __setlease() due to an uninitialised return value. Problem diagnosed by Björn Steinbrink. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com> Tested-by: Daniel Hokka Zakrisson <daniel@hozac.com> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-05-08 10:02:42 +07:00
}
locks: clean up lease_alloc() Return the newly allocated structure as the return value instead of using a struct ** parameter. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2007-03-02 02:34:35 +07:00
return fl;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
/* Check if two locks overlap each other.
*/
static inline int locks_overlap(struct file_lock *fl1, struct file_lock *fl2)
{
return ((fl1->fl_end >= fl2->fl_start) &&
(fl2->fl_end >= fl1->fl_start));
}
/*
* Check whether two locks have the same owner.
*/
[PATCH] tiny: Uninline some fslocks.c functions uninline some file locking functions add/remove: 3/0 grow/shrink: 0/15 up/down: 256/-1525 (-1269) function old new delta locks_free_lock - 134 +134 posix_same_owner - 69 +69 __locks_delete_block - 53 +53 posix_locks_conflict 126 108 -18 locks_remove_posix 266 237 -29 locks_wake_up_blocks 121 87 -34 locks_block_on_timeout 83 47 -36 locks_insert_block 157 120 -37 locks_delete_block 62 23 -39 posix_unblock_lock 104 59 -45 posix_locks_deadlock 162 100 -62 locks_delete_lock 228 119 -109 sys_flock 338 217 -121 __break_lease 600 474 -126 lease_init 252 122 -130 fcntl_setlk64 793 649 -144 fcntl_setlk 793 649 -144 __posix_lock_file 1477 1026 -451 Signed-off-by: Matt Mackall <mpm@selenic.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-08 16:05:22 +07:00
static int posix_same_owner(struct file_lock *fl1, struct file_lock *fl2)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
locks: rename lock-manager ops 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: J. Bruce Fields <bfields@redhat.com>
2011-07-21 07:21:59 +07:00
if (fl1->fl_lmops && fl1->fl_lmops->lm_compare_owner)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
return fl2->fl_lmops == fl1->fl_lmops &&
locks: rename lock-manager ops 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: J. Bruce Fields <bfields@redhat.com>
2011-07-21 07:21:59 +07:00
fl1->fl_lmops->lm_compare_owner(fl1, fl2);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
return fl1->fl_owner == fl2->fl_owner;
}
locks: add a dedicated spinlock to protect i_flctx lists We can now add a dedicated spinlock without expanding struct inode. Change to using that to protect the various i_flctx lists. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
/* Must be called with the flc_lock held! */
locks: remove "inline" qualifier from fl_link manipulation functions It's best to let the compiler decide that. Acked-by: J. Bruce Fields <bfields@fieldses.org> Reported-by: Stephen Rothwell <sfr@canb.auug.org.au> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:07 +07:00
static void locks_insert_global_locks(struct file_lock *fl)
locks: encapsulate the fl_link list handling Move the fl_link list handling routines into a separate set of helpers. Also ensure that locks and requests are always put on global lists last (after fully initializing them) and are taken off before unintializing them. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:14 +07:00
{
fs/locks: Replace lg_local with a per-cpu spinlock As Oleg suggested, replace file_lock_list with a structure containing the hlist head and a spinlock. This completely removes the lglock from fs/locks. Suggested-by: Oleg Nesterov <oleg@redhat.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Al Viro <viro@ZenIV.linux.org.uk> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: dave@stgolabs.net Cc: der.herr@hofr.at Cc: paulmck@linux.vnet.ibm.com Cc: riel@redhat.com Cc: tj@kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-06-22 19:16:34 +07:00
struct file_lock_list_struct *fll = this_cpu_ptr(&file_lock_list);
fs/locks: Replace lg_global with a percpu-rwsem Replace the global part of the lglock with a percpu-rwsem. Since fcl_lock is a spinlock and itself nests under i_lock, which too is a spinlock we cannot acquire sleeping locks at locks_{insert,remove}_global_locks(). We can however wrap all fcl_lock acquisitions with percpu_down_read such that all invocations of locks_{insert,remove}_global_locks() have that read lock held. This allows us to replace the lg_global part of the lglock with the write side of the rwsem. In the absense of writers, percpu_{down,up}_read() are free of atomic instructions. This further avoids the very long preempt-disable regions caused by lglock on larger machines. Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Al Viro <viro@ZenIV.linux.org.uk> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: dave@stgolabs.net Cc: der.herr@hofr.at Cc: paulmck@linux.vnet.ibm.com Cc: riel@redhat.com Cc: tj@kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-06-22 19:16:33 +07:00
percpu_rwsem_assert_held(&file_rwsem);
fs/locks: Replace lg_local with a per-cpu spinlock As Oleg suggested, replace file_lock_list with a structure containing the hlist head and a spinlock. This completely removes the lglock from fs/locks. Suggested-by: Oleg Nesterov <oleg@redhat.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Al Viro <viro@ZenIV.linux.org.uk> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: dave@stgolabs.net Cc: der.herr@hofr.at Cc: paulmck@linux.vnet.ibm.com Cc: riel@redhat.com Cc: tj@kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-06-22 19:16:34 +07:00
spin_lock(&fll->lock);
locks: move file_lock_list to a set of percpu hlist_heads and convert file_lock_lock to an lglock The file_lock_list is only used for /proc/locks. The vastly common case is for locks to be put onto the list and come off again, without ever being traversed. Help optimize for this use-case by moving to percpu hlist_head-s. At the same time, we can make the locking less contentious by moving to an lglock. When iterating over the lists for /proc/locks, we must take the global lock and then iterate over each CPU's list in turn. This change necessitates a new fl_link_cpu field to keep track of which CPU the entry is on. On x86_64 at least, this field is placed within an existing hole in the struct to avoid growing the size. Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:22 +07:00
fl->fl_link_cpu = smp_processor_id();
fs/locks: Replace lg_local with a per-cpu spinlock As Oleg suggested, replace file_lock_list with a structure containing the hlist head and a spinlock. This completely removes the lglock from fs/locks. Suggested-by: Oleg Nesterov <oleg@redhat.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Al Viro <viro@ZenIV.linux.org.uk> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: dave@stgolabs.net Cc: der.herr@hofr.at Cc: paulmck@linux.vnet.ibm.com Cc: riel@redhat.com Cc: tj@kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-06-22 19:16:34 +07:00
hlist_add_head(&fl->fl_link, &fll->hlist);
spin_unlock(&fll->lock);
locks: encapsulate the fl_link list handling Move the fl_link list handling routines into a separate set of helpers. Also ensure that locks and requests are always put on global lists last (after fully initializing them) and are taken off before unintializing them. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:14 +07:00
}
locks: add a dedicated spinlock to protect i_flctx lists We can now add a dedicated spinlock without expanding struct inode. Change to using that to protect the various i_flctx lists. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
/* Must be called with the flc_lock held! */
locks: remove "inline" qualifier from fl_link manipulation functions It's best to let the compiler decide that. Acked-by: J. Bruce Fields <bfields@fieldses.org> Reported-by: Stephen Rothwell <sfr@canb.auug.org.au> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:07 +07:00
static void locks_delete_global_locks(struct file_lock *fl)
locks: encapsulate the fl_link list handling Move the fl_link list handling routines into a separate set of helpers. Also ensure that locks and requests are always put on global lists last (after fully initializing them) and are taken off before unintializing them. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:14 +07:00
{
fs/locks: Replace lg_local with a per-cpu spinlock As Oleg suggested, replace file_lock_list with a structure containing the hlist head and a spinlock. This completely removes the lglock from fs/locks. Suggested-by: Oleg Nesterov <oleg@redhat.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Al Viro <viro@ZenIV.linux.org.uk> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: dave@stgolabs.net Cc: der.herr@hofr.at Cc: paulmck@linux.vnet.ibm.com Cc: riel@redhat.com Cc: tj@kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-06-22 19:16:34 +07:00
struct file_lock_list_struct *fll;
fs/locks: Replace lg_global with a percpu-rwsem Replace the global part of the lglock with a percpu-rwsem. Since fcl_lock is a spinlock and itself nests under i_lock, which too is a spinlock we cannot acquire sleeping locks at locks_{insert,remove}_global_locks(). We can however wrap all fcl_lock acquisitions with percpu_down_read such that all invocations of locks_{insert,remove}_global_locks() have that read lock held. This allows us to replace the lg_global part of the lglock with the write side of the rwsem. In the absense of writers, percpu_{down,up}_read() are free of atomic instructions. This further avoids the very long preempt-disable regions caused by lglock on larger machines. Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Al Viro <viro@ZenIV.linux.org.uk> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: dave@stgolabs.net Cc: der.herr@hofr.at Cc: paulmck@linux.vnet.ibm.com Cc: riel@redhat.com Cc: tj@kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-06-22 19:16:33 +07:00
percpu_rwsem_assert_held(&file_rwsem);
locks: move file_lock_list to a set of percpu hlist_heads and convert file_lock_lock to an lglock The file_lock_list is only used for /proc/locks. The vastly common case is for locks to be put onto the list and come off again, without ever being traversed. Help optimize for this use-case by moving to percpu hlist_head-s. At the same time, we can make the locking less contentious by moving to an lglock. When iterating over the lists for /proc/locks, we must take the global lock and then iterate over each CPU's list in turn. This change necessitates a new fl_link_cpu field to keep track of which CPU the entry is on. On x86_64 at least, this field is placed within an existing hole in the struct to avoid growing the size. Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:22 +07:00
/*
* Avoid taking lock if already unhashed. This is safe since this check
locks: add a dedicated spinlock to protect i_flctx lists We can now add a dedicated spinlock without expanding struct inode. Change to using that to protect the various i_flctx lists. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
* is done while holding the flc_lock, and new insertions into the list
locks: move file_lock_list to a set of percpu hlist_heads and convert file_lock_lock to an lglock The file_lock_list is only used for /proc/locks. The vastly common case is for locks to be put onto the list and come off again, without ever being traversed. Help optimize for this use-case by moving to percpu hlist_head-s. At the same time, we can make the locking less contentious by moving to an lglock. When iterating over the lists for /proc/locks, we must take the global lock and then iterate over each CPU's list in turn. This change necessitates a new fl_link_cpu field to keep track of which CPU the entry is on. On x86_64 at least, this field is placed within an existing hole in the struct to avoid growing the size. Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:22 +07:00
* also require that it be held.
*/
if (hlist_unhashed(&fl->fl_link))
return;
fs/locks: Replace lg_local with a per-cpu spinlock As Oleg suggested, replace file_lock_list with a structure containing the hlist head and a spinlock. This completely removes the lglock from fs/locks. Suggested-by: Oleg Nesterov <oleg@redhat.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Al Viro <viro@ZenIV.linux.org.uk> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: dave@stgolabs.net Cc: der.herr@hofr.at Cc: paulmck@linux.vnet.ibm.com Cc: riel@redhat.com Cc: tj@kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-06-22 19:16:34 +07:00
fll = per_cpu_ptr(&file_lock_list, fl->fl_link_cpu);
spin_lock(&fll->lock);
locks: convert fl_link to a hlist_node Testing has shown that iterating over the blocked_list for deadlock detection turns out to be a bottleneck. In order to alleviate that, begin the process of turning it into a hashtable. We start by turning the fl_link into a hlist_node and the global lists into hlists. A later patch will do the conversion of the blocked_list to a hashtable. Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:17 +07:00
hlist_del_init(&fl->fl_link);
fs/locks: Replace lg_local with a per-cpu spinlock As Oleg suggested, replace file_lock_list with a structure containing the hlist head and a spinlock. This completely removes the lglock from fs/locks. Suggested-by: Oleg Nesterov <oleg@redhat.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Al Viro <viro@ZenIV.linux.org.uk> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: dave@stgolabs.net Cc: der.herr@hofr.at Cc: paulmck@linux.vnet.ibm.com Cc: riel@redhat.com Cc: tj@kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-06-22 19:16:34 +07:00
spin_unlock(&fll->lock);
locks: encapsulate the fl_link list handling Move the fl_link list handling routines into a separate set of helpers. Also ensure that locks and requests are always put on global lists last (after fully initializing them) and are taken off before unintializing them. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:14 +07:00
}
locks: add a new "lm_owner_key" lock operation Currently, the hashing that the locking code uses to add these values to the blocked_hash is simply calculated using fl_owner field. That's valid in most cases except for server-side lockd, which validates the owner of a lock based on fl_owner and fl_pid. In the case where you have a small number of NFS clients doing a lot of locking between different processes, you could end up with all the blocked requests sitting in a very small number of hash buckets. Add a new lm_owner_key operation to the lock_manager_operations that will generate an unsigned long to use as the key in the hashtable. That function is only implemented for server-side lockd, and simply XORs the fl_owner and fl_pid. Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:19 +07:00
static unsigned long
posix_owner_key(struct file_lock *fl)
{
if (fl->fl_lmops && fl->fl_lmops->lm_owner_key)
return fl->fl_lmops->lm_owner_key(fl);
return (unsigned long)fl->fl_owner;
}
locks: remove "inline" qualifier from fl_link manipulation functions It's best to let the compiler decide that. Acked-by: J. Bruce Fields <bfields@fieldses.org> Reported-by: Stephen Rothwell <sfr@canb.auug.org.au> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:07 +07:00
static void locks_insert_global_blocked(struct file_lock *waiter)
locks: encapsulate the fl_link list handling Move the fl_link list handling routines into a separate set of helpers. Also ensure that locks and requests are always put on global lists last (after fully initializing them) and are taken off before unintializing them. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:14 +07:00
{
locks: Add lockdep assertion for blocked_lock_lock Annonate insert, remove and iterate function that we need blocked_lock_lock held. Signed-off-by: Daniel Wagner <daniel.wagner@bmw-carit.de> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-04-03 20:04:03 +07:00
lockdep_assert_held(&blocked_lock_lock);
locks: add a new "lm_owner_key" lock operation Currently, the hashing that the locking code uses to add these values to the blocked_hash is simply calculated using fl_owner field. That's valid in most cases except for server-side lockd, which validates the owner of a lock based on fl_owner and fl_pid. In the case where you have a small number of NFS clients doing a lot of locking between different processes, you could end up with all the blocked requests sitting in a very small number of hash buckets. Add a new lm_owner_key operation to the lock_manager_operations that will generate an unsigned long to use as the key in the hashtable. That function is only implemented for server-side lockd, and simply XORs the fl_owner and fl_pid. Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:19 +07:00
hash_add(blocked_hash, &waiter->fl_link, posix_owner_key(waiter));
locks: encapsulate the fl_link list handling Move the fl_link list handling routines into a separate set of helpers. Also ensure that locks and requests are always put on global lists last (after fully initializing them) and are taken off before unintializing them. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:14 +07:00
}
locks: remove "inline" qualifier from fl_link manipulation functions It's best to let the compiler decide that. Acked-by: J. Bruce Fields <bfields@fieldses.org> Reported-by: Stephen Rothwell <sfr@canb.auug.org.au> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:07 +07:00
static void locks_delete_global_blocked(struct file_lock *waiter)
locks: encapsulate the fl_link list handling Move the fl_link list handling routines into a separate set of helpers. Also ensure that locks and requests are always put on global lists last (after fully initializing them) and are taken off before unintializing them. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:14 +07:00
{
locks: Add lockdep assertion for blocked_lock_lock Annonate insert, remove and iterate function that we need blocked_lock_lock held. Signed-off-by: Daniel Wagner <daniel.wagner@bmw-carit.de> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-04-03 20:04:03 +07:00
lockdep_assert_held(&blocked_lock_lock);
locks: turn the blocked_list into a hashtable Break up the blocked_list into a hashtable, using the fl_owner as a key. This speeds up searching the hash chains, which is especially significant for deadlock detection. Note that the initial implementation assumes that hashing on fl_owner is sufficient. In most cases it should be, with the notable exception being server-side lockd, which compares ownership using a tuple of the nlm_host and the pid sent in the lock request. So, this may degrade to a single hash bucket when you only have a single NFS client. That will be addressed in a later patch. The careful observer may note that this patch leaves the file_lock_list alone. There's much less of a case for turning the file_lock_list into a hashtable. The only user of that list is the code that generates /proc/locks, and it always walks the entire list. Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:18 +07:00
hash_del(&waiter->fl_link);
locks: encapsulate the fl_link list handling Move the fl_link list handling routines into a separate set of helpers. Also ensure that locks and requests are always put on global lists last (after fully initializing them) and are taken off before unintializing them. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:14 +07:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
/* Remove waiter from blocker's block list.
* When blocker ends up pointing to itself then the list is empty.
locks: protect most of the file_lock handling with i_lock Having a global lock that protects all of this code is a clear scalability problem. Instead of doing that, move most of the code to be protected by the i_lock instead. The exceptions are the global lists that the ->fl_link sits on, and the ->fl_block list. ->fl_link is what connects these structures to the global lists, so we must ensure that we hold those locks when iterating over or updating these lists. Furthermore, sound deadlock detection requires that we hold the blocked_list state steady while checking for loops. We also must ensure that the search and update to the list are atomic. For the checking and insertion side of the blocked_list, push the acquisition of the global lock into __posix_lock_file and ensure that checking and update of the blocked_list is done without dropping the lock in between. On the removal side, when waking up blocked lock waiters, take the global lock before walking the blocked list and dequeue the waiters from the global list prior to removal from the fl_block list. With this, deadlock detection should be race free while we minimize excessive file_lock_lock thrashing. Finally, in order to avoid a lock inversion problem when handling /proc/locks output we must ensure that manipulations of the fl_block list are also protected by the file_lock_lock. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:15 +07:00
*
locks: give the blocked_hash its own spinlock There's no reason we have to protect the blocked_hash and file_lock_list with the same spinlock. With the tests I have, breaking it in two gives a barely measurable performance benefit, but it seems reasonable to make this locking as granular as possible. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:20 +07:00
* Must be called with blocked_lock_lock held.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
*/
[PATCH] tiny: Uninline some fslocks.c functions uninline some file locking functions add/remove: 3/0 grow/shrink: 0/15 up/down: 256/-1525 (-1269) function old new delta locks_free_lock - 134 +134 posix_same_owner - 69 +69 __locks_delete_block - 53 +53 posix_locks_conflict 126 108 -18 locks_remove_posix 266 237 -29 locks_wake_up_blocks 121 87 -34 locks_block_on_timeout 83 47 -36 locks_insert_block 157 120 -37 locks_delete_block 62 23 -39 posix_unblock_lock 104 59 -45 posix_locks_deadlock 162 100 -62 locks_delete_lock 228 119 -109 sys_flock 338 217 -121 __break_lease 600 474 -126 lease_init 252 122 -130 fcntl_setlk64 793 649 -144 fcntl_setlk 793 649 -144 __posix_lock_file 1477 1026 -451 Signed-off-by: Matt Mackall <mpm@selenic.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-08 16:05:22 +07:00
static void __locks_delete_block(struct file_lock *waiter)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
locks: encapsulate the fl_link list handling Move the fl_link list handling routines into a separate set of helpers. Also ensure that locks and requests are always put on global lists last (after fully initializing them) and are taken off before unintializing them. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:14 +07:00
locks_delete_global_blocked(waiter);
fs/locks: rename some lists and pointers. struct file lock contains an 'fl_next' pointer which is used to point to the lock that this request is blocked waiting for. So rename it to fl_blocker. The fl_blocked list_head in an active lock is the head of a list of blocked requests. In a request it is a node in that list. These are two distinct uses, so replace with two list_heads with different names. fl_blocked_requests is the head of a list of blocked requests fl_blocked_member is a node in a member of that list. The two different list_heads are never used at the same time, but that will change in a future patch. Note that a tracepoint is changed to report fl_blocker instead of fl_next. Signed-off-by: NeilBrown <neilb@suse.com> Reviewed-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Jeff Layton <jlayton@kernel.org>
2018-11-30 06:04:08 +07:00
list_del_init(&waiter->fl_blocked_member);
waiter->fl_blocker = NULL;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
fs/locks: split out __locks_wake_up_blocks(). This functionality will be useful in future patches, so split it out from locks_wake_up_blocks(). Signed-off-by: NeilBrown <neilb@suse.com> Reviewed-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Jeff Layton <jlayton@kernel.org>
2018-11-30 06:04:08 +07:00
static void __locks_wake_up_blocks(struct file_lock *blocker)
{
while (!list_empty(&blocker->fl_blocked_requests)) {
struct file_lock *waiter;
waiter = list_first_entry(&blocker->fl_blocked_requests,
struct file_lock, fl_blocked_member);
__locks_delete_block(waiter);
if (waiter->fl_lmops && waiter->fl_lmops->lm_notify)
waiter->fl_lmops->lm_notify(waiter);
else
wake_up(&waiter->fl_wait);
}
}
cifs: use posix_unblock_lock instead of locks_delete_block commit 66189be74 (CIFS: Fix VFS lock usage for oplocked files) exported the locks_delete_block symbol. There's already an exported helper function that provides this capability however, so make cifs use that instead and turn locks_delete_block back into a static function. Note that if fl->fl_next == NULL then this lock has already been through locks_delete_block(), so we should be OK to ignore an ENOENT error here and simply not retry the lock. Cc: Pavel Shilovsky <piastryyy@gmail.com> Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:10 +07:00
static void locks_delete_block(struct file_lock *waiter)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
locks: give the blocked_hash its own spinlock There's no reason we have to protect the blocked_hash and file_lock_list with the same spinlock. With the tests I have, breaking it in two gives a barely measurable performance benefit, but it seems reasonable to make this locking as granular as possible. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:20 +07:00
spin_lock(&blocked_lock_lock);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
__locks_delete_block(waiter);
locks: give the blocked_hash its own spinlock There's no reason we have to protect the blocked_hash and file_lock_list with the same spinlock. With the tests I have, breaking it in two gives a barely measurable performance benefit, but it seems reasonable to make this locking as granular as possible. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:20 +07:00
spin_unlock(&blocked_lock_lock);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
/* Insert waiter into blocker's block list.
* We use a circular list so that processes can be easily woken up in
* the order they blocked. The documentation doesn't require this but
* it seems like the reasonable thing to do.
locks: protect most of the file_lock handling with i_lock Having a global lock that protects all of this code is a clear scalability problem. Instead of doing that, move most of the code to be protected by the i_lock instead. The exceptions are the global lists that the ->fl_link sits on, and the ->fl_block list. ->fl_link is what connects these structures to the global lists, so we must ensure that we hold those locks when iterating over or updating these lists. Furthermore, sound deadlock detection requires that we hold the blocked_list state steady while checking for loops. We also must ensure that the search and update to the list are atomic. For the checking and insertion side of the blocked_list, push the acquisition of the global lock into __posix_lock_file and ensure that checking and update of the blocked_list is done without dropping the lock in between. On the removal side, when waking up blocked lock waiters, take the global lock before walking the blocked list and dequeue the waiters from the global list prior to removal from the fl_block list. With this, deadlock detection should be race free while we minimize excessive file_lock_lock thrashing. Finally, in order to avoid a lock inversion problem when handling /proc/locks output we must ensure that manipulations of the fl_block list are also protected by the file_lock_lock. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:15 +07:00
*
locks: add a dedicated spinlock to protect i_flctx lists We can now add a dedicated spinlock without expanding struct inode. Change to using that to protect the various i_flctx lists. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
* Must be called with both the flc_lock and blocked_lock_lock held. The
fs/locks: rename some lists and pointers. struct file lock contains an 'fl_next' pointer which is used to point to the lock that this request is blocked waiting for. So rename it to fl_blocker. The fl_blocked list_head in an active lock is the head of a list of blocked requests. In a request it is a node in that list. These are two distinct uses, so replace with two list_heads with different names. fl_blocked_requests is the head of a list of blocked requests fl_blocked_member is a node in a member of that list. The two different list_heads are never used at the same time, but that will change in a future patch. Note that a tracepoint is changed to report fl_blocker instead of fl_next. Signed-off-by: NeilBrown <neilb@suse.com> Reviewed-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Jeff Layton <jlayton@kernel.org>
2018-11-30 06:04:08 +07:00
* fl_blocked_requests list itself is protected by the blocked_lock_lock,
* but by ensuring that the flc_lock is also held on insertions we can avoid
* taking the blocked_lock_lock in some cases when we see that the
* fl_blocked_requests list is empty.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
*/
locks: protect most of the file_lock handling with i_lock Having a global lock that protects all of this code is a clear scalability problem. Instead of doing that, move most of the code to be protected by the i_lock instead. The exceptions are the global lists that the ->fl_link sits on, and the ->fl_block list. ->fl_link is what connects these structures to the global lists, so we must ensure that we hold those locks when iterating over or updating these lists. Furthermore, sound deadlock detection requires that we hold the blocked_list state steady while checking for loops. We also must ensure that the search and update to the list are atomic. For the checking and insertion side of the blocked_list, push the acquisition of the global lock into __posix_lock_file and ensure that checking and update of the blocked_list is done without dropping the lock in between. On the removal side, when waking up blocked lock waiters, take the global lock before walking the blocked list and dequeue the waiters from the global list prior to removal from the fl_block list. With this, deadlock detection should be race free while we minimize excessive file_lock_lock thrashing. Finally, in order to avoid a lock inversion problem when handling /proc/locks output we must ensure that manipulations of the fl_block list are also protected by the file_lock_lock. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:15 +07:00
static void __locks_insert_block(struct file_lock *blocker,
struct file_lock *waiter)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
fs/locks: rename some lists and pointers. struct file lock contains an 'fl_next' pointer which is used to point to the lock that this request is blocked waiting for. So rename it to fl_blocker. The fl_blocked list_head in an active lock is the head of a list of blocked requests. In a request it is a node in that list. These are two distinct uses, so replace with two list_heads with different names. fl_blocked_requests is the head of a list of blocked requests fl_blocked_member is a node in a member of that list. The two different list_heads are never used at the same time, but that will change in a future patch. Note that a tracepoint is changed to report fl_blocker instead of fl_next. Signed-off-by: NeilBrown <neilb@suse.com> Reviewed-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Jeff Layton <jlayton@kernel.org>
2018-11-30 06:04:08 +07:00
BUG_ON(!list_empty(&waiter->fl_blocked_member));
waiter->fl_blocker = blocker;
list_add_tail(&waiter->fl_blocked_member, &blocker->fl_blocked_requests);
locks: rename FL_FILE_PVT and IS_FILE_PVT to use "*_OFDLCK" instead File-private locks have been re-christened as "open file description" locks. Finish the symbol name cleanup in the internal implementation. Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-04-22 19:24:32 +07:00
if (IS_POSIX(blocker) && !IS_OFDLCK(blocker))
locks: protect most of the file_lock handling with i_lock Having a global lock that protects all of this code is a clear scalability problem. Instead of doing that, move most of the code to be protected by the i_lock instead. The exceptions are the global lists that the ->fl_link sits on, and the ->fl_block list. ->fl_link is what connects these structures to the global lists, so we must ensure that we hold those locks when iterating over or updating these lists. Furthermore, sound deadlock detection requires that we hold the blocked_list state steady while checking for loops. We also must ensure that the search and update to the list are atomic. For the checking and insertion side of the blocked_list, push the acquisition of the global lock into __posix_lock_file and ensure that checking and update of the blocked_list is done without dropping the lock in between. On the removal side, when waking up blocked lock waiters, take the global lock before walking the blocked list and dequeue the waiters from the global list prior to removal from the fl_block list. With this, deadlock detection should be race free while we minimize excessive file_lock_lock thrashing. Finally, in order to avoid a lock inversion problem when handling /proc/locks output we must ensure that manipulations of the fl_block list are also protected by the file_lock_lock. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:15 +07:00
locks_insert_global_blocked(waiter);
}
locks: add a dedicated spinlock to protect i_flctx lists We can now add a dedicated spinlock without expanding struct inode. Change to using that to protect the various i_flctx lists. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
/* Must be called with flc_lock held. */
locks: protect most of the file_lock handling with i_lock Having a global lock that protects all of this code is a clear scalability problem. Instead of doing that, move most of the code to be protected by the i_lock instead. The exceptions are the global lists that the ->fl_link sits on, and the ->fl_block list. ->fl_link is what connects these structures to the global lists, so we must ensure that we hold those locks when iterating over or updating these lists. Furthermore, sound deadlock detection requires that we hold the blocked_list state steady while checking for loops. We also must ensure that the search and update to the list are atomic. For the checking and insertion side of the blocked_list, push the acquisition of the global lock into __posix_lock_file and ensure that checking and update of the blocked_list is done without dropping the lock in between. On the removal side, when waking up blocked lock waiters, take the global lock before walking the blocked list and dequeue the waiters from the global list prior to removal from the fl_block list. With this, deadlock detection should be race free while we minimize excessive file_lock_lock thrashing. Finally, in order to avoid a lock inversion problem when handling /proc/locks output we must ensure that manipulations of the fl_block list are also protected by the file_lock_lock. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:15 +07:00
static void locks_insert_block(struct file_lock *blocker,
struct file_lock *waiter)
{
locks: give the blocked_hash its own spinlock There's no reason we have to protect the blocked_hash and file_lock_list with the same spinlock. With the tests I have, breaking it in two gives a barely measurable performance benefit, but it seems reasonable to make this locking as granular as possible. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:20 +07:00
spin_lock(&blocked_lock_lock);
locks: protect most of the file_lock handling with i_lock Having a global lock that protects all of this code is a clear scalability problem. Instead of doing that, move most of the code to be protected by the i_lock instead. The exceptions are the global lists that the ->fl_link sits on, and the ->fl_block list. ->fl_link is what connects these structures to the global lists, so we must ensure that we hold those locks when iterating over or updating these lists. Furthermore, sound deadlock detection requires that we hold the blocked_list state steady while checking for loops. We also must ensure that the search and update to the list are atomic. For the checking and insertion side of the blocked_list, push the acquisition of the global lock into __posix_lock_file and ensure that checking and update of the blocked_list is done without dropping the lock in between. On the removal side, when waking up blocked lock waiters, take the global lock before walking the blocked list and dequeue the waiters from the global list prior to removal from the fl_block list. With this, deadlock detection should be race free while we minimize excessive file_lock_lock thrashing. Finally, in order to avoid a lock inversion problem when handling /proc/locks output we must ensure that manipulations of the fl_block list are also protected by the file_lock_lock. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:15 +07:00
__locks_insert_block(blocker, waiter);
locks: give the blocked_hash its own spinlock There's no reason we have to protect the blocked_hash and file_lock_list with the same spinlock. With the tests I have, breaking it in two gives a barely measurable performance benefit, but it seems reasonable to make this locking as granular as possible. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:20 +07:00
spin_unlock(&blocked_lock_lock);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
locks: comment cleanups and clarifications Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:12 +07:00
/*
* Wake up processes blocked waiting for blocker.
*
locks: add a dedicated spinlock to protect i_flctx lists We can now add a dedicated spinlock without expanding struct inode. Change to using that to protect the various i_flctx lists. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
* Must be called with the inode->flc_lock held!
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
*/
static void locks_wake_up_blocks(struct file_lock *blocker)
{
locks: avoid taking global lock if possible when waking up blocked waiters Since we always hold the i_lock when inserting a new waiter onto the fl_block list, we can avoid taking the global lock at all if we find that it's empty when we go to wake up blocked waiters. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:16 +07:00
/*
* Avoid taking global lock if list is empty. This is safe since new
locks: add a dedicated spinlock to protect i_flctx lists We can now add a dedicated spinlock without expanding struct inode. Change to using that to protect the various i_flctx lists. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
* blocked requests are only added to the list under the flc_lock, and
fs/locks: rename some lists and pointers. struct file lock contains an 'fl_next' pointer which is used to point to the lock that this request is blocked waiting for. So rename it to fl_blocker. The fl_blocked list_head in an active lock is the head of a list of blocked requests. In a request it is a node in that list. These are two distinct uses, so replace with two list_heads with different names. fl_blocked_requests is the head of a list of blocked requests fl_blocked_member is a node in a member of that list. The two different list_heads are never used at the same time, but that will change in a future patch. Note that a tracepoint is changed to report fl_blocker instead of fl_next. Signed-off-by: NeilBrown <neilb@suse.com> Reviewed-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Jeff Layton <jlayton@kernel.org>
2018-11-30 06:04:08 +07:00
* the flc_lock is always held here. Note that removal from the
* fl_blocked_requests list does not require the flc_lock, so we must
* recheck list_empty() after acquiring the blocked_lock_lock.
locks: avoid taking global lock if possible when waking up blocked waiters Since we always hold the i_lock when inserting a new waiter onto the fl_block list, we can avoid taking the global lock at all if we find that it's empty when we go to wake up blocked waiters. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:16 +07:00
*/
fs/locks: rename some lists and pointers. struct file lock contains an 'fl_next' pointer which is used to point to the lock that this request is blocked waiting for. So rename it to fl_blocker. The fl_blocked list_head in an active lock is the head of a list of blocked requests. In a request it is a node in that list. These are two distinct uses, so replace with two list_heads with different names. fl_blocked_requests is the head of a list of blocked requests fl_blocked_member is a node in a member of that list. The two different list_heads are never used at the same time, but that will change in a future patch. Note that a tracepoint is changed to report fl_blocker instead of fl_next. Signed-off-by: NeilBrown <neilb@suse.com> Reviewed-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Jeff Layton <jlayton@kernel.org>
2018-11-30 06:04:08 +07:00
if (list_empty(&blocker->fl_blocked_requests))
locks: avoid taking global lock if possible when waking up blocked waiters Since we always hold the i_lock when inserting a new waiter onto the fl_block list, we can avoid taking the global lock at all if we find that it's empty when we go to wake up blocked waiters. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:16 +07:00
return;
locks: give the blocked_hash its own spinlock There's no reason we have to protect the blocked_hash and file_lock_list with the same spinlock. With the tests I have, breaking it in two gives a barely measurable performance benefit, but it seems reasonable to make this locking as granular as possible. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:20 +07:00
spin_lock(&blocked_lock_lock);
fs/locks: split out __locks_wake_up_blocks(). This functionality will be useful in future patches, so split it out from locks_wake_up_blocks(). Signed-off-by: NeilBrown <neilb@suse.com> Reviewed-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Jeff Layton <jlayton@kernel.org>
2018-11-30 06:04:08 +07:00
__locks_wake_up_blocks(blocker);
locks: give the blocked_hash its own spinlock There's no reason we have to protect the blocked_hash and file_lock_list with the same spinlock. With the tests I have, breaking it in two gives a barely measurable performance benefit, but it seems reasonable to make this locking as granular as possible. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:20 +07:00
spin_unlock(&blocked_lock_lock);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
locks: move flock locks to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
static void
Revert "locks: keep a count of locks on the flctx lists" This reverts commit 9bd0f45b7037fcfa8b575c7e27d0431d6e6dc3bb. Linus rightly pointed out that I failed to initialize the counters when adding them, so they don't work as expected. Just revert this patch for now. Reported-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-02-17 02:32:03 +07:00
locks_insert_lock_ctx(struct file_lock *fl, struct list_head *before)
locks: move flock locks to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
{
list_add_tail(&fl->fl_list, before);
locks_insert_global_locks(fl);
}
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
static void
Revert "locks: keep a count of locks on the flctx lists" This reverts commit 9bd0f45b7037fcfa8b575c7e27d0431d6e6dc3bb. Linus rightly pointed out that I failed to initialize the counters when adding them, so they don't work as expected. Just revert this patch for now. Reported-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-02-17 02:32:03 +07:00
locks_unlink_lock_ctx(struct file_lock *fl)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
locks: encapsulate the fl_link list handling Move the fl_link list handling routines into a separate set of helpers. Also ensure that locks and requests are always put on global lists last (after fully initializing them) and are taken off before unintializing them. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:14 +07:00
locks_delete_global_locks(fl);
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
list_del_init(&fl->fl_list);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
locks_wake_up_blocks(fl);
locks: close potential race between setlease and open As Al Viro points out, there is an unlikely, but possible race between opening a file and setting a lease on it. generic_add_lease is done with the i_lock held, but the inode->i_flock check in break_lease is lockless. It's possible for another task doing an open to do the entire pathwalk and call break_lease between the point where generic_add_lease checks for a conflicting open and adds the lease to the list. If this occurs, we can end up with a lease set on the file with a conflicting open. To guard against that, check again for a conflicting open after adding the lease to the i_flock list. If the above race occurs, then we can simply unwind the lease setting and return -EAGAIN. Because we take dentry references and acquire write access on the file before calling break_lease, we know that if the i_flock list is empty when the open caller goes to check it then the necessary refcounts have already been incremented. Thus the additional check for a conflicting open will see that there is one and the setlease call will fail. Cc: Bruce Fields <bfields@fieldses.org> Cc: David Howells <dhowells@redhat.com> Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Reported-by: Al Viro <viro@ZenIV.linux.org.uk> Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@fieldses.org>
2014-02-04 00:13:06 +07:00
}
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
static void
Revert "locks: keep a count of locks on the flctx lists" This reverts commit 9bd0f45b7037fcfa8b575c7e27d0431d6e6dc3bb. Linus rightly pointed out that I failed to initialize the counters when adding them, so they don't work as expected. Just revert this patch for now. Reported-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-02-17 02:32:03 +07:00
locks_delete_lock_ctx(struct file_lock *fl, struct list_head *dispose)
locks: close potential race between setlease and open As Al Viro points out, there is an unlikely, but possible race between opening a file and setting a lease on it. generic_add_lease is done with the i_lock held, but the inode->i_flock check in break_lease is lockless. It's possible for another task doing an open to do the entire pathwalk and call break_lease between the point where generic_add_lease checks for a conflicting open and adds the lease to the list. If this occurs, we can end up with a lease set on the file with a conflicting open. To guard against that, check again for a conflicting open after adding the lease to the i_flock list. If the above race occurs, then we can simply unwind the lease setting and return -EAGAIN. Because we take dentry references and acquire write access on the file before calling break_lease, we know that if the i_flock list is empty when the open caller goes to check it then the necessary refcounts have already been incremented. Thus the additional check for a conflicting open will see that there is one and the setlease call will fail. Cc: Bruce Fields <bfields@fieldses.org> Cc: David Howells <dhowells@redhat.com> Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Reported-by: Al Viro <viro@ZenIV.linux.org.uk> Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@fieldses.org>
2014-02-04 00:13:06 +07:00
{
Revert "locks: keep a count of locks on the flctx lists" This reverts commit 9bd0f45b7037fcfa8b575c7e27d0431d6e6dc3bb. Linus rightly pointed out that I failed to initialize the counters when adding them, so they don't work as expected. Just revert this patch for now. Reported-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-02-17 02:32:03 +07:00
locks_unlink_lock_ctx(fl);
locks: defer freeing locks in locks_delete_lock until after i_lock has been dropped In commit 72f98e72551fa (locks: turn lock_flocks into a spinlock), we moved from using the BKL to a global spinlock. With this change, we lost the ability to block in the fl_release_private operation. This is problematic for NFS (and probably some other filesystems as well). Add a new list_head argument to locks_delete_lock. If that argument is non-NULL, then queue any locks that we want to free to the list instead of freeing them. Then, add a new locks_dispose_list function that will walk such a list and call locks_free_lock on them after the i_lock has been dropped. Finally, change all of the callers of locks_delete_lock to pass in a list_head, except for lease_modify. That function can be called long after the i_lock has been acquired. Deferring the freeing of a lease after unlocking it in that function is non-trivial until we overhaul some of the spinlocking in the lease code. Currently though, no filesystem that sets fl_release_private supports leases, so this is not currently a problem. We'll eventually want to make the same change in the lease code, but it needs a lot more work before we can reasonably do so. Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-08-12 01:20:31 +07:00
if (dispose)
locks: add new struct list_head to struct file_lock ...that we can use to queue file_locks to per-ctx list_heads. Go ahead and convert locks_delete_lock and locks_dispose_list to use it instead of the fl_block list. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:54 +07:00
list_add(&fl->fl_list, dispose);
locks: defer freeing locks in locks_delete_lock until after i_lock has been dropped In commit 72f98e72551fa (locks: turn lock_flocks into a spinlock), we moved from using the BKL to a global spinlock. With this change, we lost the ability to block in the fl_release_private operation. This is problematic for NFS (and probably some other filesystems as well). Add a new list_head argument to locks_delete_lock. If that argument is non-NULL, then queue any locks that we want to free to the list instead of freeing them. Then, add a new locks_dispose_list function that will walk such a list and call locks_free_lock on them after the i_lock has been dropped. Finally, change all of the callers of locks_delete_lock to pass in a list_head, except for lease_modify. That function can be called long after the i_lock has been acquired. Deferring the freeing of a lease after unlocking it in that function is non-trivial until we overhaul some of the spinlocking in the lease code. Currently though, no filesystem that sets fl_release_private supports leases, so this is not currently a problem. We'll eventually want to make the same change in the lease code, but it needs a lot more work before we can reasonably do so. Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-08-12 01:20:31 +07:00
else
locks_free_lock(fl);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
/* Determine if lock sys_fl blocks lock caller_fl. Common functionality
* checks for shared/exclusive status of overlapping locks.
*/
static int locks_conflict(struct file_lock *caller_fl, struct file_lock *sys_fl)
{
if (sys_fl->fl_type == F_WRLCK)
return 1;
if (caller_fl->fl_type == F_WRLCK)
return 1;
return 0;
}
/* Determine if lock sys_fl blocks lock caller_fl. POSIX specific
* checking before calling the locks_conflict().
*/
static int posix_locks_conflict(struct file_lock *caller_fl, struct file_lock *sys_fl)
{
/* POSIX locks owned by the same process do not conflict with
* each other.
*/
locks: remove extraneous IS_POSIX and IS_FLOCK tests We know that the locks being passed into this function are of the correct type, now that they live on their own lists. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-04-03 20:04:02 +07:00
if (posix_same_owner(caller_fl, sys_fl))
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
return (0);
/* Check whether they overlap */
if (!locks_overlap(caller_fl, sys_fl))
return 0;
return (locks_conflict(caller_fl, sys_fl));
}
/* Determine if lock sys_fl blocks lock caller_fl. FLOCK specific
* checking before calling the locks_conflict().
*/
static int flock_locks_conflict(struct file_lock *caller_fl, struct file_lock *sys_fl)
{
/* FLOCK locks referring to the same filp do not conflict with
* each other.
*/
locks: remove extraneous IS_POSIX and IS_FLOCK tests We know that the locks being passed into this function are of the correct type, now that they live on their own lists. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-04-03 20:04:02 +07:00
if (caller_fl->fl_file == sys_fl->fl_file)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
return (0);
if ((caller_fl->fl_type & LOCK_MAND) || (sys_fl->fl_type & LOCK_MAND))
return 0;
return (locks_conflict(caller_fl, sys_fl));
}
locks: make posix_test_lock() interface more consistent Since posix_test_lock(), like fcntl() and ->lock(), indicates absence or presence of a conflict lock by setting fl_type to, respectively, F_UNLCK or something other than F_UNLCK, the return value is no longer needed. Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu>
2007-05-12 03:09:32 +07:00
void
locks: give posix_test_lock same interface as ->lock posix_test_lock() and ->lock() do the same job but have gratuitously different interfaces. Modify posix_test_lock() so the two agree, simplifying some code in the process. Signed-off-by: Marc Eshel <eshel@almaden.ibm.com> Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu>
2007-02-21 12:55:18 +07:00
posix_test_lock(struct file *filp, struct file_lock *fl)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
struct file_lock *cfl;
locks: convert posix locks to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
struct file_lock_context *ctx;
locks: fix file locking on overlayfs This patch allows flock, posix locks, ofd locks and leases to work correctly on overlayfs. Instead of using the underlying inode for storing lock context use the overlay inode. This allows locks to be persistent across copy-up. This is done by introducing locks_inode() helper and using it instead of file_inode() to get the inode in locking code. For non-overlayfs the two are equivalent, except for an extra pointer dereference in locks_inode(). Since lock operations are in "struct file_operations" we must also make sure not to call underlying filesystem's lock operations. Introcude a super block flag MS_NOREMOTELOCK to this effect. Signed-off-by: Miklos Szeredi <mszeredi@redhat.com> Acked-by: Jeff Layton <jlayton@poochiereds.net> Cc: "J. Bruce Fields" <bfields@fieldses.org>
2016-09-16 17:44:20 +07:00
struct inode *inode = locks_inode(filp);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
fs: fix data races on inode->i_flctx locks_get_lock_context() uses cmpxchg() to install i_flctx. cmpxchg() is a release operation which is correct. But it uses a plain load to load i_flctx. This is incorrect. Subsequent loads from i_flctx can hoist above the load of i_flctx pointer itself and observe uninitialized garbage there. This in turn can lead to corruption of ctx->flc_lock and other members. Documentation/memory-barriers.txt explicitly requires to use a barrier in such context: "A load-load control dependency requires a full read memory barrier". Use smp_load_acquire() in locks_get_lock_context() and in bunch of other functions that can proceed concurrently with locks_get_lock_context(). The data race was found with KernelThreadSanitizer (KTSAN). Signed-off-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-09-21 14:43:06 +07:00
ctx = smp_load_acquire(&inode->i_flctx);
locks: convert posix locks to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
if (!ctx || list_empty_careful(&ctx->flc_posix)) {
fl->fl_type = F_UNLCK;
return;
}
locks: add a dedicated spinlock to protect i_flctx lists We can now add a dedicated spinlock without expanding struct inode. Change to using that to protect the various i_flctx lists. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
spin_lock(&ctx->flc_lock);
locks: convert posix locks to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
list_for_each_entry(cfl, &ctx->flc_posix, fl_list) {
if (posix_locks_conflict(fl, cfl)) {
locks_copy_conflock(fl, cfl);
goto out;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
locks: convert posix locks to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
fl->fl_type = F_UNLCK;
out:
locks: add a dedicated spinlock to protect i_flctx lists We can now add a dedicated spinlock without expanding struct inode. Change to using that to protect the various i_flctx lists. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
spin_unlock(&ctx->flc_lock);
locks: make posix_test_lock() interface more consistent Since posix_test_lock(), like fcntl() and ->lock(), indicates absence or presence of a conflict lock by setting fl_type to, respectively, F_UNLCK or something other than F_UNLCK, the return value is no longer needed. Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu>
2007-05-12 03:09:32 +07:00
return;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
EXPORT_SYMBOL(posix_test_lock);
locks: clarify posix_locks_deadlock For such a short function (with such a long comment), posix_locks_deadlock() seems to cause a lot of confusion. Attempt to make it a bit clearer: - Remove the initial posix_same_owner() check, which can never pass (since this is only called in the case that block_fl and caller_fl conflict) - Use an explicit loop (and a helper function) instead of a goto. - Rewrite the comment, attempting a clearer explanation, and removing some uninteresting historical detail. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2007-10-27 05:05:40 +07:00
/*
* Deadlock detection:
*
* We attempt to detect deadlocks that are due purely to posix file
* locks.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
*
locks: clarify posix_locks_deadlock For such a short function (with such a long comment), posix_locks_deadlock() seems to cause a lot of confusion. Attempt to make it a bit clearer: - Remove the initial posix_same_owner() check, which can never pass (since this is only called in the case that block_fl and caller_fl conflict) - Use an explicit loop (and a helper function) instead of a goto. - Rewrite the comment, attempting a clearer explanation, and removing some uninteresting historical detail. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2007-10-27 05:05:40 +07:00
* We assume that a task can be waiting for at most one lock at a time.
* So for any acquired lock, the process holding that lock may be
* waiting on at most one other lock. That lock in turns may be held by
* someone waiting for at most one other lock. Given a requested lock
* caller_fl which is about to wait for a conflicting lock block_fl, we
* follow this chain of waiters to ensure we are not about to create a
* cycle.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
*
locks: clarify posix_locks_deadlock For such a short function (with such a long comment), posix_locks_deadlock() seems to cause a lot of confusion. Attempt to make it a bit clearer: - Remove the initial posix_same_owner() check, which can never pass (since this is only called in the case that block_fl and caller_fl conflict) - Use an explicit loop (and a helper function) instead of a goto. - Rewrite the comment, attempting a clearer explanation, and removing some uninteresting historical detail. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2007-10-27 05:05:40 +07:00
* Since we do this before we ever put a process to sleep on a lock, we
* are ensured that there is never a cycle; that is what guarantees that
* the while() loop in posix_locks_deadlock() eventually completes.
locks: fix possible infinite loop in posix deadlock detection It's currently possible to send posix_locks_deadlock() into an infinite loop (under the BKL). For now, fix this just by bailing out after a few iterations. We may want to fix this in a way that better clarifies the semantics of deadlock detection. But that will take more time, and this minimal fix is probably adequate for any realistic scenario, and is simple enough to be appropriate for applying to stable kernels now. Thanks to George Davis for reporting the problem. Cc: "George G. Davis" <gdavis@mvista.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Acked-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-10-30 22:20:02 +07:00
*
locks: clarify posix_locks_deadlock For such a short function (with such a long comment), posix_locks_deadlock() seems to cause a lot of confusion. Attempt to make it a bit clearer: - Remove the initial posix_same_owner() check, which can never pass (since this is only called in the case that block_fl and caller_fl conflict) - Use an explicit loop (and a helper function) instead of a goto. - Rewrite the comment, attempting a clearer explanation, and removing some uninteresting historical detail. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2007-10-27 05:05:40 +07:00
* Note: the above assumption may not be true when handling lock
* requests from a broken NFS client. It may also fail in the presence
* of tasks (such as posix threads) sharing the same open file table.
* To handle those cases, we just bail out after a few iterations.
locks: skip deadlock detection on FL_FILE_PVT locks It's not really feasible to do deadlock detection with FL_FILE_PVT locks since they aren't owned by a single task, per-se. Deadlock detection also tends to be rather expensive so just skip it for these sorts of locks. Also, add a FIXME comment about adding more limited deadlock detection that just applies to ro -> rw upgrades, per Andy's request. Cc: Andy Lutomirski <luto@amacapital.net> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:09 +07:00
*
locks: rename FL_FILE_PVT and IS_FILE_PVT to use "*_OFDLCK" instead File-private locks have been re-christened as "open file description" locks. Finish the symbol name cleanup in the internal implementation. Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-04-22 19:24:32 +07:00
* For FL_OFDLCK locks, the owner is the filp, not the files_struct.
locks: skip deadlock detection on FL_FILE_PVT locks It's not really feasible to do deadlock detection with FL_FILE_PVT locks since they aren't owned by a single task, per-se. Deadlock detection also tends to be rather expensive so just skip it for these sorts of locks. Also, add a FIXME comment about adding more limited deadlock detection that just applies to ro -> rw upgrades, per Andy's request. Cc: Andy Lutomirski <luto@amacapital.net> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:09 +07:00
* Because the owner is not even nominally tied to a thread of
* execution, the deadlock detection below can't reasonably work well. Just
* skip it for those.
*
locks: rename FL_FILE_PVT and IS_FILE_PVT to use "*_OFDLCK" instead File-private locks have been re-christened as "open file description" locks. Finish the symbol name cleanup in the internal implementation. Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-04-22 19:24:32 +07:00
* In principle, we could do a more limited deadlock detection on FL_OFDLCK
locks: skip deadlock detection on FL_FILE_PVT locks It's not really feasible to do deadlock detection with FL_FILE_PVT locks since they aren't owned by a single task, per-se. Deadlock detection also tends to be rather expensive so just skip it for these sorts of locks. Also, add a FIXME comment about adding more limited deadlock detection that just applies to ro -> rw upgrades, per Andy's request. Cc: Andy Lutomirski <luto@amacapital.net> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:09 +07:00
* locks that just checks for the case where two tasks are attempting to
* upgrade from read to write locks on the same inode.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
*/
locks: fix possible infinite loop in posix deadlock detection It's currently possible to send posix_locks_deadlock() into an infinite loop (under the BKL). For now, fix this just by bailing out after a few iterations. We may want to fix this in a way that better clarifies the semantics of deadlock detection. But that will take more time, and this minimal fix is probably adequate for any realistic scenario, and is simple enough to be appropriate for applying to stable kernels now. Thanks to George Davis for reporting the problem. Cc: "George G. Davis" <gdavis@mvista.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Acked-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-10-30 22:20:02 +07:00
#define MAX_DEADLK_ITERATIONS 10
locks: clarify posix_locks_deadlock For such a short function (with such a long comment), posix_locks_deadlock() seems to cause a lot of confusion. Attempt to make it a bit clearer: - Remove the initial posix_same_owner() check, which can never pass (since this is only called in the case that block_fl and caller_fl conflict) - Use an explicit loop (and a helper function) instead of a goto. - Rewrite the comment, attempting a clearer explanation, and removing some uninteresting historical detail. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2007-10-27 05:05:40 +07:00
/* Find a lock that the owner of the given block_fl is blocking on. */
static struct file_lock *what_owner_is_waiting_for(struct file_lock *block_fl)
{
struct file_lock *fl;
locks: add a new "lm_owner_key" lock operation Currently, the hashing that the locking code uses to add these values to the blocked_hash is simply calculated using fl_owner field. That's valid in most cases except for server-side lockd, which validates the owner of a lock based on fl_owner and fl_pid. In the case where you have a small number of NFS clients doing a lot of locking between different processes, you could end up with all the blocked requests sitting in a very small number of hash buckets. Add a new lm_owner_key operation to the lock_manager_operations that will generate an unsigned long to use as the key in the hashtable. That function is only implemented for server-side lockd, and simply XORs the fl_owner and fl_pid. Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:19 +07:00
hash_for_each_possible(blocked_hash, fl, fl_link, posix_owner_key(block_fl)) {
locks: clarify posix_locks_deadlock For such a short function (with such a long comment), posix_locks_deadlock() seems to cause a lot of confusion. Attempt to make it a bit clearer: - Remove the initial posix_same_owner() check, which can never pass (since this is only called in the case that block_fl and caller_fl conflict) - Use an explicit loop (and a helper function) instead of a goto. - Rewrite the comment, attempting a clearer explanation, and removing some uninteresting historical detail. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2007-10-27 05:05:40 +07:00
if (posix_same_owner(fl, block_fl))
fs/locks: rename some lists and pointers. struct file lock contains an 'fl_next' pointer which is used to point to the lock that this request is blocked waiting for. So rename it to fl_blocker. The fl_blocked list_head in an active lock is the head of a list of blocked requests. In a request it is a node in that list. These are two distinct uses, so replace with two list_heads with different names. fl_blocked_requests is the head of a list of blocked requests fl_blocked_member is a node in a member of that list. The two different list_heads are never used at the same time, but that will change in a future patch. Note that a tracepoint is changed to report fl_blocker instead of fl_next. Signed-off-by: NeilBrown <neilb@suse.com> Reviewed-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Jeff Layton <jlayton@kernel.org>
2018-11-30 06:04:08 +07:00
return fl->fl_blocker;
locks: clarify posix_locks_deadlock For such a short function (with such a long comment), posix_locks_deadlock() seems to cause a lot of confusion. Attempt to make it a bit clearer: - Remove the initial posix_same_owner() check, which can never pass (since this is only called in the case that block_fl and caller_fl conflict) - Use an explicit loop (and a helper function) instead of a goto. - Rewrite the comment, attempting a clearer explanation, and removing some uninteresting historical detail. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2007-10-27 05:05:40 +07:00
}
return NULL;
}
locks: give the blocked_hash its own spinlock There's no reason we have to protect the blocked_hash and file_lock_list with the same spinlock. With the tests I have, breaking it in two gives a barely measurable performance benefit, but it seems reasonable to make this locking as granular as possible. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:20 +07:00
/* Must be called with the blocked_lock_lock held! */
[PATCH] fs/locks.c: make posix_locks_deadlock() static We can now make posix_locks_deadlock() static. Signed-off-by: Adrian Bunk <bunk@stusta.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>
2006-06-23 16:05:13 +07:00
static int posix_locks_deadlock(struct file_lock *caller_fl,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
struct file_lock *block_fl)
{
locks: fix possible infinite loop in posix deadlock detection It's currently possible to send posix_locks_deadlock() into an infinite loop (under the BKL). For now, fix this just by bailing out after a few iterations. We may want to fix this in a way that better clarifies the semantics of deadlock detection. But that will take more time, and this minimal fix is probably adequate for any realistic scenario, and is simple enough to be appropriate for applying to stable kernels now. Thanks to George Davis for reporting the problem. Cc: "George G. Davis" <gdavis@mvista.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Acked-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-10-30 22:20:02 +07:00
int i = 0;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
locks: Add lockdep assertion for blocked_lock_lock Annonate insert, remove and iterate function that we need blocked_lock_lock held. Signed-off-by: Daniel Wagner <daniel.wagner@bmw-carit.de> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-04-03 20:04:03 +07:00
lockdep_assert_held(&blocked_lock_lock);
locks: skip deadlock detection on FL_FILE_PVT locks It's not really feasible to do deadlock detection with FL_FILE_PVT locks since they aren't owned by a single task, per-se. Deadlock detection also tends to be rather expensive so just skip it for these sorts of locks. Also, add a FIXME comment about adding more limited deadlock detection that just applies to ro -> rw upgrades, per Andy's request. Cc: Andy Lutomirski <luto@amacapital.net> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:09 +07:00
/*
* This deadlock detector can't reasonably detect deadlocks with
locks: rename FL_FILE_PVT and IS_FILE_PVT to use "*_OFDLCK" instead File-private locks have been re-christened as "open file description" locks. Finish the symbol name cleanup in the internal implementation. Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-04-22 19:24:32 +07:00
* FL_OFDLCK locks, since they aren't owned by a process, per-se.
locks: skip deadlock detection on FL_FILE_PVT locks It's not really feasible to do deadlock detection with FL_FILE_PVT locks since they aren't owned by a single task, per-se. Deadlock detection also tends to be rather expensive so just skip it for these sorts of locks. Also, add a FIXME comment about adding more limited deadlock detection that just applies to ro -> rw upgrades, per Andy's request. Cc: Andy Lutomirski <luto@amacapital.net> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:09 +07:00
*/
locks: rename FL_FILE_PVT and IS_FILE_PVT to use "*_OFDLCK" instead File-private locks have been re-christened as "open file description" locks. Finish the symbol name cleanup in the internal implementation. Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-04-22 19:24:32 +07:00
if (IS_OFDLCK(caller_fl))
locks: skip deadlock detection on FL_FILE_PVT locks It's not really feasible to do deadlock detection with FL_FILE_PVT locks since they aren't owned by a single task, per-se. Deadlock detection also tends to be rather expensive so just skip it for these sorts of locks. Also, add a FIXME comment about adding more limited deadlock detection that just applies to ro -> rw upgrades, per Andy's request. Cc: Andy Lutomirski <luto@amacapital.net> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:09 +07:00
return 0;
locks: clarify posix_locks_deadlock For such a short function (with such a long comment), posix_locks_deadlock() seems to cause a lot of confusion. Attempt to make it a bit clearer: - Remove the initial posix_same_owner() check, which can never pass (since this is only called in the case that block_fl and caller_fl conflict) - Use an explicit loop (and a helper function) instead of a goto. - Rewrite the comment, attempting a clearer explanation, and removing some uninteresting historical detail. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2007-10-27 05:05:40 +07:00
while ((block_fl = what_owner_is_waiting_for(block_fl))) {
if (i++ > MAX_DEADLK_ITERATIONS)
return 0;
if (posix_same_owner(caller_fl, block_fl))
return 1;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
return 0;
}
/* Try to create a FLOCK lock on filp. We always insert new FLOCK locks
locks: fix flock_lock_file() comment This comment wasn't updated when lease support was added, and it makes essentially the same mistake that the code made before a recent bugfix. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2007-09-13 02:45:07 +07:00
* after any leases, but before any posix locks.
VFS: Allow caller to determine if BSD or posix locks were actually freed Change posix_lock_file_conf(), and flock_lock_file() so that if called with an F_UNLCK argument, and the FL_EXISTS flag they will indicate whether or not any locks were actually freed by returning 0 or -ENOENT. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-06-30 03:38:32 +07:00
*
* Note that if called with an FL_EXISTS argument, the caller may determine
* whether or not a lock was successfully freed by testing the return
* value for -ENOENT.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
*/
locks: have flock_lock_file take an inode pointer instead of a filp ...and rename it to better describe how it works. In order to fix a use-after-free in NFS, we need to be able to remove locks from an inode after the filp associated with them may have already been freed. flock_lock_file already only dereferences the filp to get to the inode, so just change it so the callers do that. All of the callers already pass in a lock request that has the fl_file set properly, so we don't need to pass it in individually. With that change it now only dereferences the filp to get to the inode, so just push that out to the callers. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Reviewed-by: "J. Bruce Fields" <bfields@fieldses.org> Tested-by: "J. Bruce Fields" <bfields@fieldses.org>
2015-07-11 17:43:02 +07:00
static int flock_lock_inode(struct inode *inode, struct file_lock *request)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
[PATCH] fs/locks.c: Fix sys_flock() race sys_flock() currently has a race which can result in a double free in the multi-thread case. Thread 1 Thread 2 sys_flock(file, LOCK_EX) sys_flock(file, LOCK_UN) If Thread 2 removes the lock from inode->i_lock before Thread 1 tests for list_empty(&lock->fl_link) at the end of sys_flock, then both threads will end up calling locks_free_lock for the same lock. Fix is to make flock_lock_file() do the same as posix_lock_file(), namely to make a copy of the request, so that the caller can always free the lock. This also has the side-effect of fixing up a reference problem in the lockd handling of flock. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-03-31 17:30:55 +07:00
struct file_lock *new_fl = NULL;
locks: move flock locks to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
struct file_lock *fl;
struct file_lock_context *ctx;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
int error = 0;
locks: move flock locks to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
bool found = false;
locks: defer freeing locks in locks_delete_lock until after i_lock has been dropped In commit 72f98e72551fa (locks: turn lock_flocks into a spinlock), we moved from using the BKL to a global spinlock. With this change, we lost the ability to block in the fl_release_private operation. This is problematic for NFS (and probably some other filesystems as well). Add a new list_head argument to locks_delete_lock. If that argument is non-NULL, then queue any locks that we want to free to the list instead of freeing them. Then, add a new locks_dispose_list function that will walk such a list and call locks_free_lock on them after the i_lock has been dropped. Finally, change all of the callers of locks_delete_lock to pass in a list_head, except for lease_modify. That function can be called long after the i_lock has been acquired. Deferring the freeing of a lease after unlocking it in that function is non-trivial until we overhaul some of the spinlocking in the lease code. Currently though, no filesystem that sets fl_release_private supports leases, so this is not currently a problem. We'll eventually want to make the same change in the lease code, but it needs a lot more work before we can reasonably do so. Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-08-12 01:20:31 +07:00
LIST_HEAD(dispose);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
locks: don't allocate a lock context for an F_UNLCK request In the event that we get an F_UNLCK request on an inode that has no lock context, there is no reason to allocate one. Change locks_get_lock_context to take a "type" pointer and avoid allocating a new context if it's F_UNLCK. Then, fix the callers to return appropriately if that function returns NULL. Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2015-04-03 20:04:03 +07:00
ctx = locks_get_lock_context(inode, request->fl_type);
if (!ctx) {
if (request->fl_type != F_UNLCK)
return -ENOMEM;
return (request->fl_flags & FL_EXISTS) ? -ENOENT : 0;
}
locks: move flock locks to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
fs/locks.c: prepare for BKL removal This prepares the removal of the big kernel lock from the file locking code. We still use the BKL as long as fs/lockd uses it and ceph might sleep, but we can flip the definition to a private spinlock as soon as that's done. All users outside of fs/lockd get converted to use lock_flocks() instead of lock_kernel() where appropriate. Based on an earlier patch to use a spinlock from Matthew Wilcox, who has attempted this a few times before, the earliest patch from over 10 years ago turned it into a semaphore, which ended up being slower than the BKL and was subsequently reverted. Someone should do some serious performance testing when this becomes a spinlock, since this has caused problems before. Using a spinlock should be at least as good as the BKL in theory, but who knows... Signed-off-by: Arnd Bergmann <arnd@arndb.de> Acked-by: Matthew Wilcox <willy@linux.intel.com> Cc: Christoph Hellwig <hch@lst.de> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Miklos Szeredi <mszeredi@suse.cz> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: John Kacur <jkacur@redhat.com> Cc: Sage Weil <sage@newdream.net> Cc: linux-kernel@vger.kernel.org Cc: linux-fsdevel@vger.kernel.org
2010-09-18 20:09:31 +07:00
if (!(request->fl_flags & FL_ACCESS) && (request->fl_type != F_UNLCK)) {
Memory shortage can result in inconsistent flocks state When the flock_lock_file() is called to change the flock from F_RDLCK to F_WRLCK or vice versa the existing flock can be removed without appropriate warning. Look: for_each_lock(inode, before) { struct file_lock *fl = *before; if (IS_POSIX(fl)) break; if (IS_LEASE(fl)) continue; if (filp != fl->fl_file) continue; if (request->fl_type == fl->fl_type) goto out; found = 1; locks_delete_lock(before); <<<<<< ! break; } if after this point the subsequent locks_alloc_lock() will fail the return code will be -ENOMEM, but the existing lock is already removed. This is a known feature that such "re-locking" is not atomic, but in the racy case the file should stay locked (although by some other process), but in this case the file will be unlocked. The proposal is to prepare the lock in advance keeping no chance to fail in the future code. Found during making the flocks pid-namespaces aware. (Note: Thanks to Reuben Farrelly for finding a bug in an earlier version of this patch.) Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Cc: Reuben Farrelly <reuben-linuxkernel@reub.net>
2007-09-11 19:38:13 +07:00
new_fl = locks_alloc_lock();
fs/locks.c: prepare for BKL removal This prepares the removal of the big kernel lock from the file locking code. We still use the BKL as long as fs/lockd uses it and ceph might sleep, but we can flip the definition to a private spinlock as soon as that's done. All users outside of fs/lockd get converted to use lock_flocks() instead of lock_kernel() where appropriate. Based on an earlier patch to use a spinlock from Matthew Wilcox, who has attempted this a few times before, the earliest patch from over 10 years ago turned it into a semaphore, which ended up being slower than the BKL and was subsequently reverted. Someone should do some serious performance testing when this becomes a spinlock, since this has caused problems before. Using a spinlock should be at least as good as the BKL in theory, but who knows... Signed-off-by: Arnd Bergmann <arnd@arndb.de> Acked-by: Matthew Wilcox <willy@linux.intel.com> Cc: Christoph Hellwig <hch@lst.de> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Miklos Szeredi <mszeredi@suse.cz> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: John Kacur <jkacur@redhat.com> Cc: Sage Weil <sage@newdream.net> Cc: linux-kernel@vger.kernel.org Cc: linux-fsdevel@vger.kernel.org
2010-09-18 20:09:31 +07:00
if (!new_fl)
return -ENOMEM;
Memory shortage can result in inconsistent flocks state When the flock_lock_file() is called to change the flock from F_RDLCK to F_WRLCK or vice versa the existing flock can be removed without appropriate warning. Look: for_each_lock(inode, before) { struct file_lock *fl = *before; if (IS_POSIX(fl)) break; if (IS_LEASE(fl)) continue; if (filp != fl->fl_file) continue; if (request->fl_type == fl->fl_type) goto out; found = 1; locks_delete_lock(before); <<<<<< ! break; } if after this point the subsequent locks_alloc_lock() will fail the return code will be -ENOMEM, but the existing lock is already removed. This is a known feature that such "re-locking" is not atomic, but in the racy case the file should stay locked (although by some other process), but in this case the file will be unlocked. The proposal is to prepare the lock in advance keeping no chance to fail in the future code. Found during making the flocks pid-namespaces aware. (Note: Thanks to Reuben Farrelly for finding a bug in an earlier version of this patch.) Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Cc: Reuben Farrelly <reuben-linuxkernel@reub.net>
2007-09-11 19:38:13 +07:00
}
fs/locks: Use percpu_down_read_preempt_disable() Avoid spurious preemption. Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Al Viro <viro@ZenIV.linux.org.uk> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: dave@stgolabs.net Cc: der.herr@hofr.at Cc: paulmck@linux.vnet.ibm.com Cc: riel@redhat.com Cc: tj@kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2016-05-30 21:48:35 +07:00
percpu_down_read_preempt_disable(&file_rwsem);
locks: add a dedicated spinlock to protect i_flctx lists We can now add a dedicated spinlock without expanding struct inode. Change to using that to protect the various i_flctx lists. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
spin_lock(&ctx->flc_lock);
fs/locks.c: prepare for BKL removal This prepares the removal of the big kernel lock from the file locking code. We still use the BKL as long as fs/lockd uses it and ceph might sleep, but we can flip the definition to a private spinlock as soon as that's done. All users outside of fs/lockd get converted to use lock_flocks() instead of lock_kernel() where appropriate. Based on an earlier patch to use a spinlock from Matthew Wilcox, who has attempted this a few times before, the earliest patch from over 10 years ago turned it into a semaphore, which ended up being slower than the BKL and was subsequently reverted. Someone should do some serious performance testing when this becomes a spinlock, since this has caused problems before. Using a spinlock should be at least as good as the BKL in theory, but who knows... Signed-off-by: Arnd Bergmann <arnd@arndb.de> Acked-by: Matthew Wilcox <willy@linux.intel.com> Cc: Christoph Hellwig <hch@lst.de> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Miklos Szeredi <mszeredi@suse.cz> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: John Kacur <jkacur@redhat.com> Cc: Sage Weil <sage@newdream.net> Cc: linux-kernel@vger.kernel.org Cc: linux-fsdevel@vger.kernel.org
2010-09-18 20:09:31 +07:00
if (request->fl_flags & FL_ACCESS)
goto find_conflict;
locks: move flock locks to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
list_for_each_entry(fl, &ctx->flc_flock, fl_list) {
locks: have flock_lock_file take an inode pointer instead of a filp ...and rename it to better describe how it works. In order to fix a use-after-free in NFS, we need to be able to remove locks from an inode after the filp associated with them may have already been freed. flock_lock_file already only dereferences the filp to get to the inode, so just change it so the callers do that. All of the callers already pass in a lock request that has the fl_file set properly, so we don't need to pass it in individually. With that change it now only dereferences the filp to get to the inode, so just push that out to the callers. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Reviewed-by: "J. Bruce Fields" <bfields@fieldses.org> Tested-by: "J. Bruce Fields" <bfields@fieldses.org>
2015-07-11 17:43:02 +07:00
if (request->fl_file != fl->fl_file)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
continue;
[PATCH] fs/locks.c: Fix sys_flock() race sys_flock() currently has a race which can result in a double free in the multi-thread case. Thread 1 Thread 2 sys_flock(file, LOCK_EX) sys_flock(file, LOCK_UN) If Thread 2 removes the lock from inode->i_lock before Thread 1 tests for list_empty(&lock->fl_link) at the end of sys_flock, then both threads will end up calling locks_free_lock for the same lock. Fix is to make flock_lock_file() do the same as posix_lock_file(), namely to make a copy of the request, so that the caller can always free the lock. This also has the side-effect of fixing up a reference problem in the lockd handling of flock. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-03-31 17:30:55 +07:00
if (request->fl_type == fl->fl_type)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
goto out;
locks: move flock locks to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
found = true;
Revert "locks: keep a count of locks on the flctx lists" This reverts commit 9bd0f45b7037fcfa8b575c7e27d0431d6e6dc3bb. Linus rightly pointed out that I failed to initialize the counters when adding them, so they don't work as expected. Just revert this patch for now. Reported-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-02-17 02:32:03 +07:00
locks_delete_lock_ctx(fl, &dispose);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
break;
}
VFS: Allow caller to determine if BSD or posix locks were actually freed Change posix_lock_file_conf(), and flock_lock_file() so that if called with an F_UNLCK argument, and the FL_EXISTS flag they will indicate whether or not any locks were actually freed by returning 0 or -ENOENT. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-06-30 03:38:32 +07:00
if (request->fl_type == F_UNLCK) {
if ((request->fl_flags & FL_EXISTS) && !found)
error = -ENOENT;
[PATCH] fs/locks.c: Fix sys_flock() race sys_flock() currently has a race which can result in a double free in the multi-thread case. Thread 1 Thread 2 sys_flock(file, LOCK_EX) sys_flock(file, LOCK_UN) If Thread 2 removes the lock from inode->i_lock before Thread 1 tests for list_empty(&lock->fl_link) at the end of sys_flock, then both threads will end up calling locks_free_lock for the same lock. Fix is to make flock_lock_file() do the same as posix_lock_file(), namely to make a copy of the request, so that the caller can always free the lock. This also has the side-effect of fixing up a reference problem in the lockd handling of flock. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-03-31 17:30:55 +07:00
goto out;
VFS: Allow caller to determine if BSD or posix locks were actually freed Change posix_lock_file_conf(), and flock_lock_file() so that if called with an F_UNLCK argument, and the FL_EXISTS flag they will indicate whether or not any locks were actually freed by returning 0 or -ENOENT. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-06-30 03:38:32 +07:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
VFS: Add support for the FL_ACCESS flag to flock_lock_file() Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-06-30 03:38:37 +07:00
find_conflict:
locks: move flock locks to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
list_for_each_entry(fl, &ctx->flc_flock, fl_list) {
[PATCH] fs/locks.c: Fix sys_flock() race sys_flock() currently has a race which can result in a double free in the multi-thread case. Thread 1 Thread 2 sys_flock(file, LOCK_EX) sys_flock(file, LOCK_UN) If Thread 2 removes the lock from inode->i_lock before Thread 1 tests for list_empty(&lock->fl_link) at the end of sys_flock, then both threads will end up calling locks_free_lock for the same lock. Fix is to make flock_lock_file() do the same as posix_lock_file(), namely to make a copy of the request, so that the caller can always free the lock. This also has the side-effect of fixing up a reference problem in the lockd handling of flock. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-03-31 17:30:55 +07:00
if (!flock_locks_conflict(request, fl))
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
continue;
error = -EAGAIN;
locks: add special return value for asynchronous locks Use a special error value FILE_LOCK_DEFERRED to mean that a locking operation returned asynchronously. This is returned by posix_lock_file() for sleeping locks to mean that the lock has been queued on the block list, and will be woken up when it might become available and needs to be retried (either fl_lmops->fl_notify() is called or fl_wait is woken up). f_op->lock() to mean either the above, or that the filesystem will call back with fl_lmops->fl_grant() when the result of the locking operation is known. The filesystem can do this for sleeping as well as non-sleeping locks. This is to make sure, that return values of -EAGAIN and -EINPROGRESS by filesystems are not mistaken to mean an asynchronous locking. This also makes error handling in fs/locks.c and lockd/svclock.c slightly cleaner. Signed-off-by: Miklos Szeredi <mszeredi@suse.cz> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Matthew Wilcox <matthew@wil.cx> Cc: David Teigland <teigland@redhat.com> Cc: Christoph Hellwig <hch@lst.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-07-25 15:48:57 +07:00
if (!(request->fl_flags & FL_SLEEP))
goto out;
error = FILE_LOCK_DEFERRED;
locks_insert_block(fl, request);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
goto out;
}
VFS: Add support for the FL_ACCESS flag to flock_lock_file() Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-06-30 03:38:37 +07:00
if (request->fl_flags & FL_ACCESS)
goto out;
[PATCH] fs/locks.c: Fix sys_flock() race sys_flock() currently has a race which can result in a double free in the multi-thread case. Thread 1 Thread 2 sys_flock(file, LOCK_EX) sys_flock(file, LOCK_UN) If Thread 2 removes the lock from inode->i_lock before Thread 1 tests for list_empty(&lock->fl_link) at the end of sys_flock, then both threads will end up calling locks_free_lock for the same lock. Fix is to make flock_lock_file() do the same as posix_lock_file(), namely to make a copy of the request, so that the caller can always free the lock. This also has the side-effect of fixing up a reference problem in the lockd handling of flock. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-03-31 17:30:55 +07:00
locks_copy_lock(new_fl, request);
Revert "locks: keep a count of locks on the flctx lists" This reverts commit 9bd0f45b7037fcfa8b575c7e27d0431d6e6dc3bb. Linus rightly pointed out that I failed to initialize the counters when adding them, so they don't work as expected. Just revert this patch for now. Reported-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-02-17 02:32:03 +07:00
locks_insert_lock_ctx(new_fl, &ctx->flc_flock);
[PATCH] fs/locks.c: Fix sys_flock() race sys_flock() currently has a race which can result in a double free in the multi-thread case. Thread 1 Thread 2 sys_flock(file, LOCK_EX) sys_flock(file, LOCK_UN) If Thread 2 removes the lock from inode->i_lock before Thread 1 tests for list_empty(&lock->fl_link) at the end of sys_flock, then both threads will end up calling locks_free_lock for the same lock. Fix is to make flock_lock_file() do the same as posix_lock_file(), namely to make a copy of the request, so that the caller can always free the lock. This also has the side-effect of fixing up a reference problem in the lockd handling of flock. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-03-31 17:30:55 +07:00
new_fl = NULL;
[PATCH] Return error in case flock_lock_file failure If flock_lock_file() failed to allocate flock with locks_alloc_lock() then "error = 0" is returned. Need to return some non-zero. Signed-off-by: Pavel Emelianov <xemul@openvz.org> Signed-off-by: Kirill Korotaev <dev@openvz.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-06-14 20:59:35 +07:00
error = 0;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
out:
locks: add a dedicated spinlock to protect i_flctx lists We can now add a dedicated spinlock without expanding struct inode. Change to using that to protect the various i_flctx lists. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
spin_unlock(&ctx->flc_lock);
fs/locks: Use percpu_down_read_preempt_disable() Avoid spurious preemption. Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Al Viro <viro@ZenIV.linux.org.uk> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: dave@stgolabs.net Cc: der.herr@hofr.at Cc: paulmck@linux.vnet.ibm.com Cc: riel@redhat.com Cc: tj@kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2016-05-30 21:48:35 +07:00
percpu_up_read_preempt_enable(&file_rwsem);
[PATCH] fs/locks.c: Fix sys_flock() race sys_flock() currently has a race which can result in a double free in the multi-thread case. Thread 1 Thread 2 sys_flock(file, LOCK_EX) sys_flock(file, LOCK_UN) If Thread 2 removes the lock from inode->i_lock before Thread 1 tests for list_empty(&lock->fl_link) at the end of sys_flock, then both threads will end up calling locks_free_lock for the same lock. Fix is to make flock_lock_file() do the same as posix_lock_file(), namely to make a copy of the request, so that the caller can always free the lock. This also has the side-effect of fixing up a reference problem in the lockd handling of flock. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-03-31 17:30:55 +07:00
if (new_fl)
locks_free_lock(new_fl);
locks: defer freeing locks in locks_delete_lock until after i_lock has been dropped In commit 72f98e72551fa (locks: turn lock_flocks into a spinlock), we moved from using the BKL to a global spinlock. With this change, we lost the ability to block in the fl_release_private operation. This is problematic for NFS (and probably some other filesystems as well). Add a new list_head argument to locks_delete_lock. If that argument is non-NULL, then queue any locks that we want to free to the list instead of freeing them. Then, add a new locks_dispose_list function that will walk such a list and call locks_free_lock on them after the i_lock has been dropped. Finally, change all of the callers of locks_delete_lock to pass in a list_head, except for lease_modify. That function can be called long after the i_lock has been acquired. Deferring the freeing of a lease after unlocking it in that function is non-trivial until we overhaul some of the spinlocking in the lease code. Currently though, no filesystem that sets fl_release_private supports leases, so this is not currently a problem. We'll eventually want to make the same change in the lease code, but it needs a lot more work before we can reasonably do so. Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-08-12 01:20:31 +07:00
locks_dispose_list(&dispose);
locks: add tracepoint in flock codepath Signed-off-by: Jeff Layton <jlayton@kernel.org>
2018-07-30 18:54:56 +07:00
trace_flock_lock_inode(inode, request, error);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
return error;
}
locks: rename __posix_lock_file to posix_lock_inode ...a more descriptive name and we can drop the double underscore prefix. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Acked-by: "J. Bruce Fields" <bfields@fieldses.org>
2016-01-08 06:27:42 +07:00
static int posix_lock_inode(struct inode *inode, struct file_lock *request,
struct file_lock *conflock)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
locks: convert posix locks to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
struct file_lock *fl, *tmp;
[PATCH] locks: don't do unnecessary allocations posix_lock_file() always allocates new locks in advance, even if it's easy to determine that no allocations will be needed. Optimize these cases: - FL_ACCESS flag is set - Unlocking the whole range Signed-off-by: Miklos Szeredi <miklos@szeredi.hu> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-06-23 16:05:10 +07:00
struct file_lock *new_fl = NULL;
struct file_lock *new_fl2 = NULL;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
struct file_lock *left = NULL;
struct file_lock *right = NULL;
locks: convert posix locks to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
struct file_lock_context *ctx;
locks: make "added" in __posix_lock_file a bool Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:13 +07:00
int error;
bool added = false;
locks: defer freeing locks in locks_delete_lock until after i_lock has been dropped In commit 72f98e72551fa (locks: turn lock_flocks into a spinlock), we moved from using the BKL to a global spinlock. With this change, we lost the ability to block in the fl_release_private operation. This is problematic for NFS (and probably some other filesystems as well). Add a new list_head argument to locks_delete_lock. If that argument is non-NULL, then queue any locks that we want to free to the list instead of freeing them. Then, add a new locks_dispose_list function that will walk such a list and call locks_free_lock on them after the i_lock has been dropped. Finally, change all of the callers of locks_delete_lock to pass in a list_head, except for lease_modify. That function can be called long after the i_lock has been acquired. Deferring the freeing of a lease after unlocking it in that function is non-trivial until we overhaul some of the spinlocking in the lease code. Currently though, no filesystem that sets fl_release_private supports leases, so this is not currently a problem. We'll eventually want to make the same change in the lease code, but it needs a lot more work before we can reasonably do so. Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-08-12 01:20:31 +07:00
LIST_HEAD(dispose);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
locks: don't allocate a lock context for an F_UNLCK request In the event that we get an F_UNLCK request on an inode that has no lock context, there is no reason to allocate one. Change locks_get_lock_context to take a "type" pointer and avoid allocating a new context if it's F_UNLCK. Then, fix the callers to return appropriately if that function returns NULL. Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2015-04-03 20:04:03 +07:00
ctx = locks_get_lock_context(inode, request->fl_type);
locks: convert posix locks to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
if (!ctx)
locks: don't allocate a lock context for an F_UNLCK request In the event that we get an F_UNLCK request on an inode that has no lock context, there is no reason to allocate one. Change locks_get_lock_context to take a "type" pointer and avoid allocating a new context if it's F_UNLCK. Then, fix the callers to return appropriately if that function returns NULL. Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2015-04-03 20:04:03 +07:00
return (request->fl_type == F_UNLCK) ? 0 : -ENOMEM;
locks: convert posix locks to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
/*
* We may need two file_lock structures for this operation,
* so we get them in advance to avoid races.
[PATCH] locks: don't do unnecessary allocations posix_lock_file() always allocates new locks in advance, even if it's easy to determine that no allocations will be needed. Optimize these cases: - FL_ACCESS flag is set - Unlocking the whole range Signed-off-by: Miklos Szeredi <miklos@szeredi.hu> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-06-23 16:05:10 +07:00
*
* In some cases we can be sure, that no new locks will be needed
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
*/
[PATCH] locks: don't do unnecessary allocations posix_lock_file() always allocates new locks in advance, even if it's easy to determine that no allocations will be needed. Optimize these cases: - FL_ACCESS flag is set - Unlocking the whole range Signed-off-by: Miklos Szeredi <miklos@szeredi.hu> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-06-23 16:05:10 +07:00
if (!(request->fl_flags & FL_ACCESS) &&
(request->fl_type != F_UNLCK ||
request->fl_start != 0 || request->fl_end != OFFSET_MAX)) {
new_fl = locks_alloc_lock();
new_fl2 = locks_alloc_lock();
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
fs/locks: Use percpu_down_read_preempt_disable() Avoid spurious preemption. Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Al Viro <viro@ZenIV.linux.org.uk> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: dave@stgolabs.net Cc: der.herr@hofr.at Cc: paulmck@linux.vnet.ibm.com Cc: riel@redhat.com Cc: tj@kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2016-05-30 21:48:35 +07:00
percpu_down_read_preempt_disable(&file_rwsem);
locks: add a dedicated spinlock to protect i_flctx lists We can now add a dedicated spinlock without expanding struct inode. Change to using that to protect the various i_flctx lists. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
spin_lock(&ctx->flc_lock);
locks: comment cleanups and clarifications Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:12 +07:00
/*
* New lock request. Walk all POSIX locks and look for conflicts. If
* there are any, either return error or put the request on the
locks: turn the blocked_list into a hashtable Break up the blocked_list into a hashtable, using the fl_owner as a key. This speeds up searching the hash chains, which is especially significant for deadlock detection. Note that the initial implementation assumes that hashing on fl_owner is sufficient. In most cases it should be, with the notable exception being server-side lockd, which compares ownership using a tuple of the nlm_host and the pid sent in the lock request. So, this may degrade to a single hash bucket when you only have a single NFS client. That will be addressed in a later patch. The careful observer may note that this patch leaves the file_lock_list alone. There's much less of a case for turning the file_lock_list into a hashtable. The only user of that list is the code that generates /proc/locks, and it always walks the entire list. Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:18 +07:00
* blocker's list of waiters and the global blocked_hash.
locks: comment cleanups and clarifications Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:12 +07:00
*/
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
if (request->fl_type != F_UNLCK) {
locks: convert posix locks to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
list_for_each_entry(fl, &ctx->flc_posix, fl_list) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
if (!posix_locks_conflict(request, fl))
continue;
[PATCH] VFS,fs/locks.c,NFSD4: add race_free posix_lock_file_conf() interface Lockd and the NFSv4 server both exercise a race condition where posix_test_lock() is called either before or after posix_lock_file() to deal with a denied lock request due to a conflicting lock. Remove the race condition for the NFSv4 server by adding a new conflicting lock parameter to __posix_lock_file() , changing the name to __posix_lock_file_conf(). Keep posix_lock_file() interface, add posix_lock_conf() interface, both call __posix_lock_file_conf(). [akpm@osdl.org: Put the EXPORT_SYMBOL() where it belongs] Signed-off-by: Andy Adamson <andros@citi.umich.edu> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-03-26 16:37:26 +07:00
if (conflock)
locks: Rename __locks_copy_lock() to locks_copy_conflock() Jeff advice, " Right now __locks_copy_lock is only used to copy conflocks. It would be good to rename that to something more distinct (i.e.locks_copy_conflock), to make it clear that we're generating a conflock there." v5: change order from 3/6 to 2/6 v4: new patch only renaming function name Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-08-22 21:18:42 +07:00
locks_copy_conflock(conflock, fl);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
error = -EAGAIN;
if (!(request->fl_flags & FL_SLEEP))
goto out;
locks: protect most of the file_lock handling with i_lock Having a global lock that protects all of this code is a clear scalability problem. Instead of doing that, move most of the code to be protected by the i_lock instead. The exceptions are the global lists that the ->fl_link sits on, and the ->fl_block list. ->fl_link is what connects these structures to the global lists, so we must ensure that we hold those locks when iterating over or updating these lists. Furthermore, sound deadlock detection requires that we hold the blocked_list state steady while checking for loops. We also must ensure that the search and update to the list are atomic. For the checking and insertion side of the blocked_list, push the acquisition of the global lock into __posix_lock_file and ensure that checking and update of the blocked_list is done without dropping the lock in between. On the removal side, when waking up blocked lock waiters, take the global lock before walking the blocked list and dequeue the waiters from the global list prior to removal from the fl_block list. With this, deadlock detection should be race free while we minimize excessive file_lock_lock thrashing. Finally, in order to avoid a lock inversion problem when handling /proc/locks output we must ensure that manipulations of the fl_block list are also protected by the file_lock_lock. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:15 +07:00
/*
* Deadlock detection and insertion into the blocked
* locks list must be done while holding the same lock!
*/
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
error = -EDEADLK;
locks: give the blocked_hash its own spinlock There's no reason we have to protect the blocked_hash and file_lock_list with the same spinlock. With the tests I have, breaking it in two gives a barely measurable performance benefit, but it seems reasonable to make this locking as granular as possible. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:20 +07:00
spin_lock(&blocked_lock_lock);
locks: protect most of the file_lock handling with i_lock Having a global lock that protects all of this code is a clear scalability problem. Instead of doing that, move most of the code to be protected by the i_lock instead. The exceptions are the global lists that the ->fl_link sits on, and the ->fl_block list. ->fl_link is what connects these structures to the global lists, so we must ensure that we hold those locks when iterating over or updating these lists. Furthermore, sound deadlock detection requires that we hold the blocked_list state steady while checking for loops. We also must ensure that the search and update to the list are atomic. For the checking and insertion side of the blocked_list, push the acquisition of the global lock into __posix_lock_file and ensure that checking and update of the blocked_list is done without dropping the lock in between. On the removal side, when waking up blocked lock waiters, take the global lock before walking the blocked list and dequeue the waiters from the global list prior to removal from the fl_block list. With this, deadlock detection should be race free while we minimize excessive file_lock_lock thrashing. Finally, in order to avoid a lock inversion problem when handling /proc/locks output we must ensure that manipulations of the fl_block list are also protected by the file_lock_lock. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:15 +07:00
if (likely(!posix_locks_deadlock(request, fl))) {
error = FILE_LOCK_DEFERRED;
__locks_insert_block(fl, request);
}
locks: give the blocked_hash its own spinlock There's no reason we have to protect the blocked_hash and file_lock_list with the same spinlock. With the tests I have, breaking it in two gives a barely measurable performance benefit, but it seems reasonable to make this locking as granular as possible. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:20 +07:00
spin_unlock(&blocked_lock_lock);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
goto out;
}
}
/* If we're just looking for a conflict, we're done. */
error = 0;
if (request->fl_flags & FL_ACCESS)
goto out;
locks: convert posix locks to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
/* Find the first old lock with the same owner as the new lock */
list_for_each_entry(fl, &ctx->flc_posix, fl_list) {
if (posix_same_owner(request, fl))
break;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
locks: comment cleanups and clarifications Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:12 +07:00
/* Process locks with this owner. */
locks: convert posix locks to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
list_for_each_entry_safe_from(fl, tmp, &ctx->flc_posix, fl_list) {
if (!posix_same_owner(request, fl))
break;
/* Detect adjacent or overlapping regions (if same lock type) */
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
if (request->fl_type == fl->fl_type) {
From: Olaf Kirch <okir@suse.de> [PATCH] Fix miscompare in __posix_lock_file If an application requests the same lock twice, the kernel should just leave the existing lock in place. Currently, it will install a second lock of the same type. Signed-off-by: Olaf Kirch <okir@suse.de> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2005-08-26 06:25:35 +07:00
/* In all comparisons of start vs end, use
* "start - 1" rather than "end + 1". If end
* is OFFSET_MAX, end + 1 will become negative.
*/
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
if (fl->fl_end < request->fl_start - 1)
locks: convert posix locks to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
continue;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
/* If the next lock in the list has entirely bigger
* addresses than the new one, insert the lock here.
*/
From: Olaf Kirch <okir@suse.de> [PATCH] Fix miscompare in __posix_lock_file If an application requests the same lock twice, the kernel should just leave the existing lock in place. Currently, it will install a second lock of the same type. Signed-off-by: Olaf Kirch <okir@suse.de> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2005-08-26 06:25:35 +07:00
if (fl->fl_start - 1 > request->fl_end)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
break;
/* If we come here, the new and old lock are of the
* same type and adjacent or overlapping. Make one
* lock yielding from the lower start address of both
* locks to the higher end address.
*/
if (fl->fl_start > request->fl_start)
fl->fl_start = request->fl_start;
else
request->fl_start = fl->fl_start;
if (fl->fl_end < request->fl_end)
fl->fl_end = request->fl_end;
else
request->fl_end = fl->fl_end;
if (added) {
Revert "locks: keep a count of locks on the flctx lists" This reverts commit 9bd0f45b7037fcfa8b575c7e27d0431d6e6dc3bb. Linus rightly pointed out that I failed to initialize the counters when adding them, so they don't work as expected. Just revert this patch for now. Reported-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-02-17 02:32:03 +07:00
locks_delete_lock_ctx(fl, &dispose);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
continue;
}
request = fl;
locks: make "added" in __posix_lock_file a bool Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:13 +07:00
added = true;
locks: convert posix locks to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
} else {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
/* Processing for different lock types is a bit
* more complex.
*/
if (fl->fl_end < request->fl_start)
locks: convert posix locks to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
continue;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
if (fl->fl_start > request->fl_end)
break;
if (request->fl_type == F_UNLCK)
locks: make "added" in __posix_lock_file a bool Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:13 +07:00
added = true;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
if (fl->fl_start < request->fl_start)
left = fl;
/* If the next lock in the list has a higher end
* address than the new one, insert the new one here.
*/
if (fl->fl_end > request->fl_end) {
right = fl;
break;
}
if (fl->fl_start >= request->fl_start) {
/* The new lock completely replaces an old
* one (This may happen several times).
*/
if (added) {
Revert "locks: keep a count of locks on the flctx lists" This reverts commit 9bd0f45b7037fcfa8b575c7e27d0431d6e6dc3bb. Linus rightly pointed out that I failed to initialize the counters when adding them, so they don't work as expected. Just revert this patch for now. Reported-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-02-17 02:32:03 +07:00
locks_delete_lock_ctx(fl, &dispose);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
continue;
}
locks: don't reuse file_lock in __posix_lock_file Currently in the case where a new file lock completely replaces the old one, we end up overwriting the existing lock with the new info. This means that we have to call fl_release_private inside i_lock. Change the code to instead copy the info to new_fl, insert that lock into the correct spot and then delete the old lock. In a later patch, we'll defer the freeing of the old lock until after the i_lock has been dropped. Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-08-12 19:03:49 +07:00
/*
* Replace the old lock with new_fl, and
* remove the old one. It's safe to do the
* insert here since we know that we won't be
* using new_fl later, and that the lock is
* just replacing an existing lock.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
*/
locks: don't reuse file_lock in __posix_lock_file Currently in the case where a new file lock completely replaces the old one, we end up overwriting the existing lock with the new info. This means that we have to call fl_release_private inside i_lock. Change the code to instead copy the info to new_fl, insert that lock into the correct spot and then delete the old lock. In a later patch, we'll defer the freeing of the old lock until after the i_lock has been dropped. Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-08-12 19:03:49 +07:00
error = -ENOLCK;
if (!new_fl)
goto out;
locks_copy_lock(new_fl, request);
request = new_fl;
new_fl = NULL;
Revert "locks: keep a count of locks on the flctx lists" This reverts commit 9bd0f45b7037fcfa8b575c7e27d0431d6e6dc3bb. Linus rightly pointed out that I failed to initialize the counters when adding them, so they don't work as expected. Just revert this patch for now. Reported-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-02-17 02:32:03 +07:00
locks_insert_lock_ctx(request, &fl->fl_list);
locks_delete_lock_ctx(fl, &dispose);
locks: make "added" in __posix_lock_file a bool Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:13 +07:00
added = true;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
}
}
[PATCH] locks: don't unnecessarily fail posix lock operations posix_lock_file() was too cautious, failing operations on OOM, even if they didn't actually require an allocation. This has the disadvantage, that a failing unlock on process exit could lead to a memory leak. There are two possibilites for this: - filesystem implements .lock() and calls back to posix_lock_file(). On cleanup of files_struct locks_remove_posix() is called which should remove all locks belonging to files_struct. However if filesystem calls posix_lock_file() which fails, then those locks will never be freed. - if a file is closed while a lock is blocked, then after acquiring fcntl_setlk() will undo the lock. But this unlock itself might fail on OOM, again possibly leaking the lock. The solution is to move the checking of the allocations until after it is sure that they will be needed. This will solve the above problem since unlock will always succeed unless it splits an existing region. Signed-off-by: Miklos Szeredi <miklos@szeredi.hu> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-06-23 16:05:09 +07:00
/*
locks: comment cleanups and clarifications Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:12 +07:00
* The above code only modifies existing locks in case of merging or
* replacing. If new lock(s) need to be inserted all modifications are
* done below this, so it's safe yet to bail out.
[PATCH] locks: don't unnecessarily fail posix lock operations posix_lock_file() was too cautious, failing operations on OOM, even if they didn't actually require an allocation. This has the disadvantage, that a failing unlock on process exit could lead to a memory leak. There are two possibilites for this: - filesystem implements .lock() and calls back to posix_lock_file(). On cleanup of files_struct locks_remove_posix() is called which should remove all locks belonging to files_struct. However if filesystem calls posix_lock_file() which fails, then those locks will never be freed. - if a file is closed while a lock is blocked, then after acquiring fcntl_setlk() will undo the lock. But this unlock itself might fail on OOM, again possibly leaking the lock. The solution is to move the checking of the allocations until after it is sure that they will be needed. This will solve the above problem since unlock will always succeed unless it splits an existing region. Signed-off-by: Miklos Szeredi <miklos@szeredi.hu> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-06-23 16:05:09 +07:00
*/
error = -ENOLCK; /* "no luck" */
if (right && left == right && !new_fl2)
goto out;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
error = 0;
if (!added) {
VFS: Allow caller to determine if BSD or posix locks were actually freed Change posix_lock_file_conf(), and flock_lock_file() so that if called with an F_UNLCK argument, and the FL_EXISTS flag they will indicate whether or not any locks were actually freed by returning 0 or -ENOENT. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-06-30 03:38:32 +07:00
if (request->fl_type == F_UNLCK) {
if (request->fl_flags & FL_EXISTS)
error = -ENOENT;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
goto out;
VFS: Allow caller to determine if BSD or posix locks were actually freed Change posix_lock_file_conf(), and flock_lock_file() so that if called with an F_UNLCK argument, and the FL_EXISTS flag they will indicate whether or not any locks were actually freed by returning 0 or -ENOENT. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-06-30 03:38:32 +07:00
}
[PATCH] locks: don't unnecessarily fail posix lock operations posix_lock_file() was too cautious, failing operations on OOM, even if they didn't actually require an allocation. This has the disadvantage, that a failing unlock on process exit could lead to a memory leak. There are two possibilites for this: - filesystem implements .lock() and calls back to posix_lock_file(). On cleanup of files_struct locks_remove_posix() is called which should remove all locks belonging to files_struct. However if filesystem calls posix_lock_file() which fails, then those locks will never be freed. - if a file is closed while a lock is blocked, then after acquiring fcntl_setlk() will undo the lock. But this unlock itself might fail on OOM, again possibly leaking the lock. The solution is to move the checking of the allocations until after it is sure that they will be needed. This will solve the above problem since unlock will always succeed unless it splits an existing region. Signed-off-by: Miklos Szeredi <miklos@szeredi.hu> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-06-23 16:05:09 +07:00
if (!new_fl) {
error = -ENOLCK;
goto out;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
locks_copy_lock(new_fl, request);
Revert "locks: keep a count of locks on the flctx lists" This reverts commit 9bd0f45b7037fcfa8b575c7e27d0431d6e6dc3bb. Linus rightly pointed out that I failed to initialize the counters when adding them, so they don't work as expected. Just revert this patch for now. Reported-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-02-17 02:32:03 +07:00
locks_insert_lock_ctx(new_fl, &fl->fl_list);
locks: fix list insertion when lock is split in two In the case where we're splitting a lock in two, the current code the new "left" lock in the incorrect spot. It's inserted just before "right" when it should instead be inserted just before the new lock. When we add a new lock, set "fl" to that value so that we can add "left" before it. Reported-by: Al Viro <viro@ZenIV.linux.org.uk> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-02-18 05:08:23 +07:00
fl = new_fl;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
new_fl = NULL;
}
if (right) {
if (left == right) {
/* The new lock breaks the old one in two pieces,
* so we have to use the second new lock.
*/
left = new_fl2;
new_fl2 = NULL;
locks_copy_lock(left, right);
Revert "locks: keep a count of locks on the flctx lists" This reverts commit 9bd0f45b7037fcfa8b575c7e27d0431d6e6dc3bb. Linus rightly pointed out that I failed to initialize the counters when adding them, so they don't work as expected. Just revert this patch for now. Reported-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-02-17 02:32:03 +07:00
locks_insert_lock_ctx(left, &fl->fl_list);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
right->fl_start = request->fl_end + 1;
locks_wake_up_blocks(right);
}
if (left) {
left->fl_end = request->fl_start - 1;
locks_wake_up_blocks(left);
}
out:
locks: add a dedicated spinlock to protect i_flctx lists We can now add a dedicated spinlock without expanding struct inode. Change to using that to protect the various i_flctx lists. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
spin_unlock(&ctx->flc_lock);
fs/locks: Use percpu_down_read_preempt_disable() Avoid spurious preemption. Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Al Viro <viro@ZenIV.linux.org.uk> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: dave@stgolabs.net Cc: der.herr@hofr.at Cc: paulmck@linux.vnet.ibm.com Cc: riel@redhat.com Cc: tj@kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2016-05-30 21:48:35 +07:00
percpu_up_read_preempt_enable(&file_rwsem);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
/*
* Free any unused locks.
*/
if (new_fl)
locks_free_lock(new_fl);
if (new_fl2)
locks_free_lock(new_fl2);
locks: defer freeing locks in locks_delete_lock until after i_lock has been dropped In commit 72f98e72551fa (locks: turn lock_flocks into a spinlock), we moved from using the BKL to a global spinlock. With this change, we lost the ability to block in the fl_release_private operation. This is problematic for NFS (and probably some other filesystems as well). Add a new list_head argument to locks_delete_lock. If that argument is non-NULL, then queue any locks that we want to free to the list instead of freeing them. Then, add a new locks_dispose_list function that will walk such a list and call locks_free_lock on them after the i_lock has been dropped. Finally, change all of the callers of locks_delete_lock to pass in a list_head, except for lease_modify. That function can be called long after the i_lock has been acquired. Deferring the freeing of a lease after unlocking it in that function is non-trivial until we overhaul some of the spinlocking in the lease code. Currently though, no filesystem that sets fl_release_private supports leases, so this is not currently a problem. We'll eventually want to make the same change in the lease code, but it needs a lot more work before we can reasonably do so. Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-08-12 01:20:31 +07:00
locks_dispose_list(&dispose);
locks: sprinkle some tracepoints around the file locking code Add some tracepoints around the POSIX locking code. These were useful when tracking down problems when handling the race between setlk and close. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Acked-by: "J. Bruce Fields" <bfields@fieldses.org>
2016-01-07 09:26:10 +07:00
trace_posix_lock_inode(inode, request, error);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
return error;
}
/**
* posix_lock_file - Apply a POSIX-style lock to a file
* @filp: The file to apply the lock to
* @fl: The lock to be applied
locks: allow {vfs,posix}_lock_file to return conflicting lock The nfsv4 protocol's lock operation, in the case of a conflict, returns information about the conflicting lock. It's unclear how clients can use this, so for now we're not going so far as to add a filesystem method that can return a conflicting lock, but we may as well return something in the local case when it's easy to. Signed-off-by: Marc Eshel <eshel@almaden.ibm.com> Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu>
2007-01-19 04:15:35 +07:00
* @conflock: Place to return a copy of the conflicting lock, if found.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
*
* Add a POSIX style lock to a file.
* We merge adjacent & overlapping locks whenever possible.
* POSIX locks are sorted by owner task, then by starting address
VFS: Allow caller to determine if BSD or posix locks were actually freed Change posix_lock_file_conf(), and flock_lock_file() so that if called with an F_UNLCK argument, and the FL_EXISTS flag they will indicate whether or not any locks were actually freed by returning 0 or -ENOENT. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-06-30 03:38:32 +07:00
*
* Note that if called with an FL_EXISTS argument, the caller may determine
* whether or not a lock was successfully freed by testing the return
* value for -ENOENT.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
*/
locks: allow {vfs,posix}_lock_file to return conflicting lock The nfsv4 protocol's lock operation, in the case of a conflict, returns information about the conflicting lock. It's unclear how clients can use this, so for now we're not going so far as to add a filesystem method that can return a conflicting lock, but we may as well return something in the local case when it's easy to. Signed-off-by: Marc Eshel <eshel@almaden.ibm.com> Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu>
2007-01-19 04:15:35 +07:00
int posix_lock_file(struct file *filp, struct file_lock *fl,
[PATCH] VFS,fs/locks.c,NFSD4: add race_free posix_lock_file_conf() interface Lockd and the NFSv4 server both exercise a race condition where posix_test_lock() is called either before or after posix_lock_file() to deal with a denied lock request due to a conflicting lock. Remove the race condition for the NFSv4 server by adding a new conflicting lock parameter to __posix_lock_file() , changing the name to __posix_lock_file_conf(). Keep posix_lock_file() interface, add posix_lock_conf() interface, both call __posix_lock_file_conf(). [akpm@osdl.org: Put the EXPORT_SYMBOL() where it belongs] Signed-off-by: Andy Adamson <andros@citi.umich.edu> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-03-26 16:37:26 +07:00
struct file_lock *conflock)
{
locks: fix file locking on overlayfs This patch allows flock, posix locks, ofd locks and leases to work correctly on overlayfs. Instead of using the underlying inode for storing lock context use the overlay inode. This allows locks to be persistent across copy-up. This is done by introducing locks_inode() helper and using it instead of file_inode() to get the inode in locking code. For non-overlayfs the two are equivalent, except for an extra pointer dereference in locks_inode(). Since lock operations are in "struct file_operations" we must also make sure not to call underlying filesystem's lock operations. Introcude a super block flag MS_NOREMOTELOCK to this effect. Signed-off-by: Miklos Szeredi <mszeredi@redhat.com> Acked-by: Jeff Layton <jlayton@poochiereds.net> Cc: "J. Bruce Fields" <bfields@fieldses.org>
2016-09-16 17:44:20 +07:00
return posix_lock_inode(locks_inode(filp), fl, conflock);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
locks: allow {vfs,posix}_lock_file to return conflicting lock The nfsv4 protocol's lock operation, in the case of a conflict, returns information about the conflicting lock. It's unclear how clients can use this, so for now we're not going so far as to add a filesystem method that can return a conflicting lock, but we may as well return something in the local case when it's easy to. Signed-off-by: Marc Eshel <eshel@almaden.ibm.com> Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu>
2007-01-19 04:15:35 +07:00
EXPORT_SYMBOL(posix_lock_file);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
/**
locks: new helpers - flock_lock_inode_wait and posix_lock_inode_wait Allow callers to pass in an inode instead of a filp. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Reviewed-by: "J. Bruce Fields" <bfields@fieldses.org> Tested-by: "J. Bruce Fields" <bfields@fieldses.org>
2015-07-11 17:43:02 +07:00
* posix_lock_inode_wait - Apply a POSIX-style lock to a file
* @inode: inode of file to which lock request should be applied
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
* @fl: The lock to be applied
*
locks: cleanup posix_lock_inode_wait and flock_lock_inode_wait All callers use locks_lock_inode_wait() instead. Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-10-23 00:38:15 +07:00
* Apply a POSIX style lock request to an inode.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
*/
locks: cleanup posix_lock_inode_wait and flock_lock_inode_wait All callers use locks_lock_inode_wait() instead. Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-10-23 00:38:15 +07:00
static int posix_lock_inode_wait(struct inode *inode, struct file_lock *fl)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
int error;
might_sleep ();
for (;;) {
locks: rename __posix_lock_file to posix_lock_inode ...a more descriptive name and we can drop the double underscore prefix. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Acked-by: "J. Bruce Fields" <bfields@fieldses.org>
2016-01-08 06:27:42 +07:00
error = posix_lock_inode(inode, fl, NULL);
locks: add special return value for asynchronous locks Use a special error value FILE_LOCK_DEFERRED to mean that a locking operation returned asynchronously. This is returned by posix_lock_file() for sleeping locks to mean that the lock has been queued on the block list, and will be woken up when it might become available and needs to be retried (either fl_lmops->fl_notify() is called or fl_wait is woken up). f_op->lock() to mean either the above, or that the filesystem will call back with fl_lmops->fl_grant() when the result of the locking operation is known. The filesystem can do this for sleeping as well as non-sleeping locks. This is to make sure, that return values of -EAGAIN and -EINPROGRESS by filesystems are not mistaken to mean an asynchronous locking. This also makes error handling in fs/locks.c and lockd/svclock.c slightly cleaner. Signed-off-by: Miklos Szeredi <mszeredi@suse.cz> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Matthew Wilcox <matthew@wil.cx> Cc: David Teigland <teigland@redhat.com> Cc: Christoph Hellwig <hch@lst.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-07-25 15:48:57 +07:00
if (error != FILE_LOCK_DEFERRED)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
break;
fs/locks: rename some lists and pointers. struct file lock contains an 'fl_next' pointer which is used to point to the lock that this request is blocked waiting for. So rename it to fl_blocker. The fl_blocked list_head in an active lock is the head of a list of blocked requests. In a request it is a node in that list. These are two distinct uses, so replace with two list_heads with different names. fl_blocked_requests is the head of a list of blocked requests fl_blocked_member is a node in a member of that list. The two different list_heads are never used at the same time, but that will change in a future patch. Note that a tracepoint is changed to report fl_blocker instead of fl_next. Signed-off-by: NeilBrown <neilb@suse.com> Reviewed-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Jeff Layton <jlayton@kernel.org>
2018-11-30 06:04:08 +07:00
error = wait_event_interruptible(fl->fl_wait, !fl->fl_blocker);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
if (!error)
continue;
locks_delete_block(fl);
break;
}
return error;
}
locks: new helpers - flock_lock_inode_wait and posix_lock_inode_wait Allow callers to pass in an inode instead of a filp. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Reviewed-by: "J. Bruce Fields" <bfields@fieldses.org> Tested-by: "J. Bruce Fields" <bfields@fieldses.org>
2015-07-11 17:43:02 +07:00
locks: Allow disabling mandatory locking at compile time Mandatory locking appears to be almost unused and buggy and there appears no real interest in doing anything with it. Since effectively no one uses the code and since the code is buggy let's allow it to be disabled at compile time. I would just suggest removing the code but undoubtedly that will break some piece of userspace code somewhere. For the distributions that don't care about this piece of code this gives a nice starting point to make mandatory locking go away. Cc: Benjamin Coddington <bcodding@redhat.com> Cc: Dmitry Vyukov <dvyukov@google.com> Cc: Jeff Layton <jeff.layton@primarydata.com> Cc: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-11-16 21:49:34 +07:00
#ifdef CONFIG_MANDATORY_FILE_LOCKING
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
/**
* locks_mandatory_locked - Check for an active lock
locks: fix locks_mandatory_locked to respect file-private locks As Trond pointed out, you can currently deadlock yourself by setting a file-private lock on a file that requires mandatory locking and then trying to do I/O on it. Avoid this problem by plumbing some knowledge of file-private locks into the mandatory locking code. In order to do this, we must pass down information about the struct file that's being used to locks_verify_locked. Reported-by: Trond Myklebust <trond.myklebust@primarydata.com> Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@redhat.com>
2014-03-10 20:54:15 +07:00
* @file: the file to check
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
*
* Searches the inode's list of locks to find any POSIX locks which conflict.
* This function is called from locks_verify_locked() only.
*/
locks: fix locks_mandatory_locked to respect file-private locks As Trond pointed out, you can currently deadlock yourself by setting a file-private lock on a file that requires mandatory locking and then trying to do I/O on it. Avoid this problem by plumbing some knowledge of file-private locks into the mandatory locking code. In order to do this, we must pass down information about the struct file that's being used to locks_verify_locked. Reported-by: Trond Myklebust <trond.myklebust@primarydata.com> Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@redhat.com>
2014-03-10 20:54:15 +07:00
int locks_mandatory_locked(struct file *file)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
locks: convert posix locks to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
int ret;
locks: fix file locking on overlayfs This patch allows flock, posix locks, ofd locks and leases to work correctly on overlayfs. Instead of using the underlying inode for storing lock context use the overlay inode. This allows locks to be persistent across copy-up. This is done by introducing locks_inode() helper and using it instead of file_inode() to get the inode in locking code. For non-overlayfs the two are equivalent, except for an extra pointer dereference in locks_inode(). Since lock operations are in "struct file_operations" we must also make sure not to call underlying filesystem's lock operations. Introcude a super block flag MS_NOREMOTELOCK to this effect. Signed-off-by: Miklos Szeredi <mszeredi@redhat.com> Acked-by: Jeff Layton <jlayton@poochiereds.net> Cc: "J. Bruce Fields" <bfields@fieldses.org>
2016-09-16 17:44:20 +07:00
struct inode *inode = locks_inode(file);
locks: convert posix locks to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
struct file_lock_context *ctx;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
struct file_lock *fl;
fs: fix data races on inode->i_flctx locks_get_lock_context() uses cmpxchg() to install i_flctx. cmpxchg() is a release operation which is correct. But it uses a plain load to load i_flctx. This is incorrect. Subsequent loads from i_flctx can hoist above the load of i_flctx pointer itself and observe uninitialized garbage there. This in turn can lead to corruption of ctx->flc_lock and other members. Documentation/memory-barriers.txt explicitly requires to use a barrier in such context: "A load-load control dependency requires a full read memory barrier". Use smp_load_acquire() in locks_get_lock_context() and in bunch of other functions that can proceed concurrently with locks_get_lock_context(). The data race was found with KernelThreadSanitizer (KTSAN). Signed-off-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-09-21 14:43:06 +07:00
ctx = smp_load_acquire(&inode->i_flctx);
locks: convert posix locks to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
if (!ctx || list_empty_careful(&ctx->flc_posix))
return 0;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
/*
* Search the lock list for this inode for any POSIX locks.
*/
locks: add a dedicated spinlock to protect i_flctx lists We can now add a dedicated spinlock without expanding struct inode. Change to using that to protect the various i_flctx lists. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
spin_lock(&ctx->flc_lock);
locks: convert posix locks to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
ret = 0;
list_for_each_entry(fl, &ctx->flc_posix, fl_list) {
locks: purge fl_owner_t from fs/locks.c Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-07-13 22:00:38 +07:00
if (fl->fl_owner != current->files &&
locks: convert posix locks to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
fl->fl_owner != file) {
ret = -EAGAIN;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
break;
locks: convert posix locks to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
locks: add a dedicated spinlock to protect i_flctx lists We can now add a dedicated spinlock without expanding struct inode. Change to using that to protect the various i_flctx lists. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
spin_unlock(&ctx->flc_lock);
locks: convert posix locks to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
return ret;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
/**
* locks_mandatory_area - Check for a conflicting lock
locks: new locks_mandatory_area calling convention Pass a loff_t end for the last byte instead of the 32-bit count parameter to allow full file clones even on 32-bit architectures. While we're at it also simplify the read/write selection. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2015-12-03 18:59:49 +07:00
* @inode: the file to check
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
* @filp: how the file was opened (if it was)
locks: new locks_mandatory_area calling convention Pass a loff_t end for the last byte instead of the 32-bit count parameter to allow full file clones even on 32-bit architectures. While we're at it also simplify the read/write selection. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2015-12-03 18:59:49 +07:00
* @start: first byte in the file to check
* @end: lastbyte in the file to check
* @type: %F_WRLCK for a write lock, else %F_RDLCK
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
*
* Searches the inode's list of locks to find any POSIX locks which conflict.
*/
locks: new locks_mandatory_area calling convention Pass a loff_t end for the last byte instead of the 32-bit count parameter to allow full file clones even on 32-bit architectures. While we're at it also simplify the read/write selection. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2015-12-03 18:59:49 +07:00
int locks_mandatory_area(struct inode *inode, struct file *filp, loff_t start,
loff_t end, unsigned char type)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
struct file_lock fl;
int error;
locks: make locks_mandatory_area check for file-private locks Allow locks_mandatory_area() to handle file-private locks correctly. If there is a file-private lock set on an open file and we're doing I/O via the same, then that should not cause anything to block. Handle this by first doing a non-blocking FL_ACCESS check for a file-private lock, and then fall back to checking for a classic POSIX lock (and possibly blocking). Note that this approach is subject to the same races that have always plagued mandatory locking on Linux. Reported-by: Trond Myklebust <trond.myklebust@primarydata.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-03-10 20:54:19 +07:00
bool sleep = false;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
locks_init_lock(&fl);
fl.fl_pid = current->tgid;
fl.fl_file = filp;
fl.fl_flags = FL_POSIX | FL_ACCESS;
if (filp && !(filp->f_flags & O_NONBLOCK))
locks: make locks_mandatory_area check for file-private locks Allow locks_mandatory_area() to handle file-private locks correctly. If there is a file-private lock set on an open file and we're doing I/O via the same, then that should not cause anything to block. Handle this by first doing a non-blocking FL_ACCESS check for a file-private lock, and then fall back to checking for a classic POSIX lock (and possibly blocking). Note that this approach is subject to the same races that have always plagued mandatory locking on Linux. Reported-by: Trond Myklebust <trond.myklebust@primarydata.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-03-10 20:54:19 +07:00
sleep = true;
locks: new locks_mandatory_area calling convention Pass a loff_t end for the last byte instead of the 32-bit count parameter to allow full file clones even on 32-bit architectures. While we're at it also simplify the read/write selection. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2015-12-03 18:59:49 +07:00
fl.fl_type = type;
fl.fl_start = start;
fl.fl_end = end;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
for (;;) {
locks: make locks_mandatory_area check for file-private locks Allow locks_mandatory_area() to handle file-private locks correctly. If there is a file-private lock set on an open file and we're doing I/O via the same, then that should not cause anything to block. Handle this by first doing a non-blocking FL_ACCESS check for a file-private lock, and then fall back to checking for a classic POSIX lock (and possibly blocking). Note that this approach is subject to the same races that have always plagued mandatory locking on Linux. Reported-by: Trond Myklebust <trond.myklebust@primarydata.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-03-10 20:54:19 +07:00
if (filp) {
locks: purge fl_owner_t from fs/locks.c Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-07-13 22:00:38 +07:00
fl.fl_owner = filp;
locks: make locks_mandatory_area check for file-private locks Allow locks_mandatory_area() to handle file-private locks correctly. If there is a file-private lock set on an open file and we're doing I/O via the same, then that should not cause anything to block. Handle this by first doing a non-blocking FL_ACCESS check for a file-private lock, and then fall back to checking for a classic POSIX lock (and possibly blocking). Note that this approach is subject to the same races that have always plagued mandatory locking on Linux. Reported-by: Trond Myklebust <trond.myklebust@primarydata.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-03-10 20:54:19 +07:00
fl.fl_flags &= ~FL_SLEEP;
locks: rename __posix_lock_file to posix_lock_inode ...a more descriptive name and we can drop the double underscore prefix. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Acked-by: "J. Bruce Fields" <bfields@fieldses.org>
2016-01-08 06:27:42 +07:00
error = posix_lock_inode(inode, &fl, NULL);
locks: make locks_mandatory_area check for file-private locks Allow locks_mandatory_area() to handle file-private locks correctly. If there is a file-private lock set on an open file and we're doing I/O via the same, then that should not cause anything to block. Handle this by first doing a non-blocking FL_ACCESS check for a file-private lock, and then fall back to checking for a classic POSIX lock (and possibly blocking). Note that this approach is subject to the same races that have always plagued mandatory locking on Linux. Reported-by: Trond Myklebust <trond.myklebust@primarydata.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-03-10 20:54:19 +07:00
if (!error)
break;
}
if (sleep)
fl.fl_flags |= FL_SLEEP;
fl.fl_owner = current->files;
locks: rename __posix_lock_file to posix_lock_inode ...a more descriptive name and we can drop the double underscore prefix. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Acked-by: "J. Bruce Fields" <bfields@fieldses.org>
2016-01-08 06:27:42 +07:00
error = posix_lock_inode(inode, &fl, NULL);
locks: add special return value for asynchronous locks Use a special error value FILE_LOCK_DEFERRED to mean that a locking operation returned asynchronously. This is returned by posix_lock_file() for sleeping locks to mean that the lock has been queued on the block list, and will be woken up when it might become available and needs to be retried (either fl_lmops->fl_notify() is called or fl_wait is woken up). f_op->lock() to mean either the above, or that the filesystem will call back with fl_lmops->fl_grant() when the result of the locking operation is known. The filesystem can do this for sleeping as well as non-sleeping locks. This is to make sure, that return values of -EAGAIN and -EINPROGRESS by filesystems are not mistaken to mean an asynchronous locking. This also makes error handling in fs/locks.c and lockd/svclock.c slightly cleaner. Signed-off-by: Miklos Szeredi <mszeredi@suse.cz> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Matthew Wilcox <matthew@wil.cx> Cc: David Teigland <teigland@redhat.com> Cc: Christoph Hellwig <hch@lst.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-07-25 15:48:57 +07:00
if (error != FILE_LOCK_DEFERRED)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
break;
fs/locks: rename some lists and pointers. struct file lock contains an 'fl_next' pointer which is used to point to the lock that this request is blocked waiting for. So rename it to fl_blocker. The fl_blocked list_head in an active lock is the head of a list of blocked requests. In a request it is a node in that list. These are two distinct uses, so replace with two list_heads with different names. fl_blocked_requests is the head of a list of blocked requests fl_blocked_member is a node in a member of that list. The two different list_heads are never used at the same time, but that will change in a future patch. Note that a tracepoint is changed to report fl_blocker instead of fl_next. Signed-off-by: NeilBrown <neilb@suse.com> Reviewed-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Jeff Layton <jlayton@kernel.org>
2018-11-30 06:04:08 +07:00
error = wait_event_interruptible(fl.fl_wait, !fl.fl_blocker);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
if (!error) {
/*
* If we've been sleeping someone might have
* changed the permissions behind our back.
*/
Cleanup macros for distinguishing mandatory locks The combination of S_ISGID bit set and S_IXGRP bit unset is used to mark the inode as "mandatory lockable" and there's a macro for this check called MANDATORY_LOCK(inode). However, fs/locks.c and some filesystems still perform the explicit i_mode checking. Besides, Andrew pointed out, that this macro is buggy itself, as it dereferences the inode arg twice. Convert this macro into static inline function and switch its users to it, making the code shorter and more readable. The __mandatory_lock() helper is to be used in places where the IS_MANDLOCK() for superblock is already known to be true. Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: David Howells <dhowells@redhat.com> Cc: Eric Van Hensbergen <ericvh@gmail.com> Cc: Ron Minnich <rminnich@sandia.gov> Cc: Latchesar Ionkov <lucho@ionkov.net> Cc: Steven Whitehouse <swhiteho@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2007-10-02 04:41:11 +07:00
if (__mandatory_lock(inode))
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
continue;
}
locks_delete_block(&fl);
break;
}
return error;
}
EXPORT_SYMBOL(locks_mandatory_area);
locks: Allow disabling mandatory locking at compile time Mandatory locking appears to be almost unused and buggy and there appears no real interest in doing anything with it. Since effectively no one uses the code and since the code is buggy let's allow it to be disabled at compile time. I would just suggest removing the code but undoubtedly that will break some piece of userspace code somewhere. For the distributions that don't care about this piece of code this gives a nice starting point to make mandatory locking go away. Cc: Benjamin Coddington <bcodding@redhat.com> Cc: Dmitry Vyukov <dvyukov@google.com> Cc: Jeff Layton <jeff.layton@primarydata.com> Cc: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-11-16 21:49:34 +07:00
#endif /* CONFIG_MANDATORY_FILE_LOCKING */
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
locks: fix tracking of inprogress lease breaks We currently use a bit in fl_flags to record whether a lease is being broken, and set fl_type to the type (RDLCK or UNLCK) that it will eventually have. This means that once the lease break starts, we forget what the lease's type *used* to be. Breaking a read lease will then result in blocking read opens, even though there's no conflict--because the lease type is now F_UNLCK and we can no longer tell whether it was previously a read or write lease. So, instead keep fl_type as the original type (the type which we enforce), and keep track of whether we're unlocking or merely downgrading by replacing the single FL_INPROGRESS flag by FL_UNLOCK_PENDING and FL_DOWNGRADE_PENDING flags. To get this right we also need to track separate downgrade and break times, to handle the case where a write-leased file gets conflicting opens first for read, then later for write. (I first considered just eliminating the downgrade behavior completely--nfsv4 doesn't need it, and nobody as far as I can tell actually uses it currently--but Jeremy Allison tells me that Windows oplocks do behave this way, so Samba will probably use this some day.) Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-27 05:25:49 +07:00
static void lease_clear_pending(struct file_lock *fl, int arg)
{
switch (arg) {
case F_UNLCK:
fl->fl_flags &= ~FL_UNLOCK_PENDING;
/* fall through: */
case F_RDLCK:
fl->fl_flags &= ~FL_DOWNGRADE_PENDING;
}
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
/* We already had a lease on this file; just change its type */
locks: clean up the lm_change prototype Now that we use standard list_heads for tracking leases, we can have lm_change take a pointer to the lease to be modified instead of a double pointer. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
int lease_modify(struct file_lock *fl, int arg, struct list_head *dispose)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
int error = assign_type(fl, arg);
if (error)
return error;
locks: fix tracking of inprogress lease breaks We currently use a bit in fl_flags to record whether a lease is being broken, and set fl_type to the type (RDLCK or UNLCK) that it will eventually have. This means that once the lease break starts, we forget what the lease's type *used* to be. Breaking a read lease will then result in blocking read opens, even though there's no conflict--because the lease type is now F_UNLCK and we can no longer tell whether it was previously a read or write lease. So, instead keep fl_type as the original type (the type which we enforce), and keep track of whether we're unlocking or merely downgrading by replacing the single FL_INPROGRESS flag by FL_UNLOCK_PENDING and FL_DOWNGRADE_PENDING flags. To get this right we also need to track separate downgrade and break times, to handle the case where a write-leased file gets conflicting opens first for read, then later for write. (I first considered just eliminating the downgrade behavior completely--nfsv4 doesn't need it, and nobody as far as I can tell actually uses it currently--but Jeremy Allison tells me that Windows oplocks do behave this way, so Samba will probably use this some day.) Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-27 05:25:49 +07:00
lease_clear_pending(fl, arg);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
locks_wake_up_blocks(fl);
locks: prevent side-effects of locks_release_private before file_lock is initialized When calling fcntl(fd, F_SETLEASE, lck) [with lck=F_WRLCK or F_RDLCK], the custom signal or owner (if any were previously set using F_SETSIG or F_SETOWN fcntls) would be reset when F_SETLEASE was called for the second time on the same file descriptor. This bug is a regression of 2.6.37 and is described here: https://bugzilla.kernel.org/show_bug.cgi?id=43336 This patch reverts a commit from Oct 2004 (with subject "nfs4 lease: move the f_delown processing") which originally introduced the lm_release_private callback. Signed-off-by: Filipe Brandenburger <filbranden@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-07-27 11:42:52 +07:00
if (arg == F_UNLCK) {
struct file *filp = fl->fl_file;
f_delown(filp);
filp->f_owner.signum = 0;
locks: move lease-specific code out of locks_delete_lock No point putting something only used by one caller into common code. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-07-28 03:18:00 +07:00
fasync_helper(0, fl->fl_file, 0, &fl->fl_fasync);
if (fl->fl_fasync != NULL) {
printk(KERN_ERR "locks_delete_lock: fasync == %p\n", fl->fl_fasync);
fl->fl_fasync = NULL;
}
Revert "locks: keep a count of locks on the flctx lists" This reverts commit 9bd0f45b7037fcfa8b575c7e27d0431d6e6dc3bb. Linus rightly pointed out that I failed to initialize the counters when adding them, so they don't work as expected. Just revert this patch for now. Reported-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-02-17 02:32:03 +07:00
locks_delete_lock_ctx(fl, dispose);
locks: prevent side-effects of locks_release_private before file_lock is initialized When calling fcntl(fd, F_SETLEASE, lck) [with lck=F_WRLCK or F_RDLCK], the custom signal or owner (if any were previously set using F_SETSIG or F_SETOWN fcntls) would be reset when F_SETLEASE was called for the second time on the same file descriptor. This bug is a regression of 2.6.37 and is described here: https://bugzilla.kernel.org/show_bug.cgi?id=43336 This patch reverts a commit from Oct 2004 (with subject "nfs4 lease: move the f_delown processing") which originally introduced the lm_release_private callback. Signed-off-by: Filipe Brandenburger <filbranden@gmail.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-07-27 11:42:52 +07:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
return 0;
}
EXPORT_SYMBOL(lease_modify);
locks: fix tracking of inprogress lease breaks We currently use a bit in fl_flags to record whether a lease is being broken, and set fl_type to the type (RDLCK or UNLCK) that it will eventually have. This means that once the lease break starts, we forget what the lease's type *used* to be. Breaking a read lease will then result in blocking read opens, even though there's no conflict--because the lease type is now F_UNLCK and we can no longer tell whether it was previously a read or write lease. So, instead keep fl_type as the original type (the type which we enforce), and keep track of whether we're unlocking or merely downgrading by replacing the single FL_INPROGRESS flag by FL_UNLOCK_PENDING and FL_DOWNGRADE_PENDING flags. To get this right we also need to track separate downgrade and break times, to handle the case where a write-leased file gets conflicting opens first for read, then later for write. (I first considered just eliminating the downgrade behavior completely--nfsv4 doesn't need it, and nobody as far as I can tell actually uses it currently--but Jeremy Allison tells me that Windows oplocks do behave this way, so Samba will probably use this some day.) Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-27 05:25:49 +07:00
static bool past_time(unsigned long then)
{
if (!then)
/* 0 is a special value meaning "this never expires": */
return false;
return time_after(jiffies, then);
}
locks: move freeing of leases outside of i_lock There was only one place where we still could free a file_lock while holding the i_lock -- lease_modify. Add a new list_head argument to the lm_change operation, pass in a private list when calling it, and fix those callers to dispose of the list once the lock has been dropped. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-09-01 18:12:07 +07:00
static void time_out_leases(struct inode *inode, struct list_head *dispose)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
struct file_lock_context *ctx = inode->i_flctx;
struct file_lock *fl, *tmp;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
locks: add a dedicated spinlock to protect i_flctx lists We can now add a dedicated spinlock without expanding struct inode. Change to using that to protect the various i_flctx lists. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
lockdep_assert_held(&ctx->flc_lock);
locks: move i_lock acquisition into generic_*_lease handlers Now that we have a saner internal API for managing leases, we no longer need to mandate that the inode->i_lock be held over most of the lease code. Push it down into generic_add_lease and generic_delete_lease. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-08-23 05:50:48 +07:00
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
list_for_each_entry_safe(fl, tmp, &ctx->flc_lease, fl_list) {
locks: add some tracepoints in the lease handling code v2: add a __break_lease tracepoint for non-blocking case Recently, I needed these to help track down a softlockup when recalling a delegation, but they might be helpful in other situations as well. Cc: "J. Bruce Fields" <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@poochiereds.net>
2014-05-10 01:13:05 +07:00
trace_time_out_leases(inode, fl);
locks: fix tracking of inprogress lease breaks We currently use a bit in fl_flags to record whether a lease is being broken, and set fl_type to the type (RDLCK or UNLCK) that it will eventually have. This means that once the lease break starts, we forget what the lease's type *used* to be. Breaking a read lease will then result in blocking read opens, even though there's no conflict--because the lease type is now F_UNLCK and we can no longer tell whether it was previously a read or write lease. So, instead keep fl_type as the original type (the type which we enforce), and keep track of whether we're unlocking or merely downgrading by replacing the single FL_INPROGRESS flag by FL_UNLOCK_PENDING and FL_DOWNGRADE_PENDING flags. To get this right we also need to track separate downgrade and break times, to handle the case where a write-leased file gets conflicting opens first for read, then later for write. (I first considered just eliminating the downgrade behavior completely--nfsv4 doesn't need it, and nobody as far as I can tell actually uses it currently--but Jeremy Allison tells me that Windows oplocks do behave this way, so Samba will probably use this some day.) Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-27 05:25:49 +07:00
if (past_time(fl->fl_downgrade_time))
locks: clean up the lm_change prototype Now that we use standard list_heads for tracking leases, we can have lm_change take a pointer to the lease to be modified instead of a double pointer. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
lease_modify(fl, F_RDLCK, dispose);
locks: fix tracking of inprogress lease breaks We currently use a bit in fl_flags to record whether a lease is being broken, and set fl_type to the type (RDLCK or UNLCK) that it will eventually have. This means that once the lease break starts, we forget what the lease's type *used* to be. Breaking a read lease will then result in blocking read opens, even though there's no conflict--because the lease type is now F_UNLCK and we can no longer tell whether it was previously a read or write lease. So, instead keep fl_type as the original type (the type which we enforce), and keep track of whether we're unlocking or merely downgrading by replacing the single FL_INPROGRESS flag by FL_UNLOCK_PENDING and FL_DOWNGRADE_PENDING flags. To get this right we also need to track separate downgrade and break times, to handle the case where a write-leased file gets conflicting opens first for read, then later for write. (I first considered just eliminating the downgrade behavior completely--nfsv4 doesn't need it, and nobody as far as I can tell actually uses it currently--but Jeremy Allison tells me that Windows oplocks do behave this way, so Samba will probably use this some day.) Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-27 05:25:49 +07:00
if (past_time(fl->fl_break_time))
locks: clean up the lm_change prototype Now that we use standard list_heads for tracking leases, we can have lm_change take a pointer to the lease to be modified instead of a double pointer. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
lease_modify(fl, F_UNLCK, dispose);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
}
locks: implement delegations Implement NFSv4 delegations at the vfs level using the new FL_DELEG lock type. Note nfsd is the only delegation user and is only using read delegations. Warn on any attempt to set a write delegation for now. We'll come back to that case later. Acked-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2012-03-06 01:18:59 +07:00
static bool leases_conflict(struct file_lock *lease, struct file_lock *breaker)
{
fs: add FL_LAYOUT lease type This (ab-)uses the file locking code to allow filesystems to recall outstanding pNFS layouts on a file. This new lease type is similar but not quite the same as FL_DELEG. A FL_LAYOUT lease can always be granted, an a per-filesystem lock (XFS iolock for the initial implementation) ensures not FL_LAYOUT leases granted when we would need to recall them. Also included are changes that allow multiple outstanding read leases of different types on the same file as long as they have a differnt owner. This wasn't a problem until now as nfsd never set FL_LEASE leases, and no one else used FL_DELEG leases, but given that nfsd will also issues FL_LAYOUT leases we will have to handle it now. Signed-off-by: Christoph Hellwig <hch@lst.de>
2015-01-22 01:17:03 +07:00
if ((breaker->fl_flags & FL_LAYOUT) != (lease->fl_flags & FL_LAYOUT))
return false;
locks: implement delegations Implement NFSv4 delegations at the vfs level using the new FL_DELEG lock type. Note nfsd is the only delegation user and is only using read delegations. Warn on any attempt to set a write delegation for now. We'll come back to that case later. Acked-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2012-03-06 01:18:59 +07:00
if ((breaker->fl_flags & FL_DELEG) && (lease->fl_flags & FL_LEASE))
return false;
return locks_conflict(breaker, lease);
}
locks: __break_lease cleanup in preparation of allowing direct removal of leases Eliminate an unneeded "flock" variable. We can use "fl" as a loop cursor everywhere. Add a any_leases_conflict helper function as well to consolidate a bit of code. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-09-02 01:53:41 +07:00
static bool
any_leases_conflict(struct inode *inode, struct file_lock *breaker)
{
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
struct file_lock_context *ctx = inode->i_flctx;
locks: __break_lease cleanup in preparation of allowing direct removal of leases Eliminate an unneeded "flock" variable. We can use "fl" as a loop cursor everywhere. Add a any_leases_conflict helper function as well to consolidate a bit of code. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-09-02 01:53:41 +07:00
struct file_lock *fl;
locks: add a dedicated spinlock to protect i_flctx lists We can now add a dedicated spinlock without expanding struct inode. Change to using that to protect the various i_flctx lists. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
lockdep_assert_held(&ctx->flc_lock);
locks: __break_lease cleanup in preparation of allowing direct removal of leases Eliminate an unneeded "flock" variable. We can use "fl" as a loop cursor everywhere. Add a any_leases_conflict helper function as well to consolidate a bit of code. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-09-02 01:53:41 +07:00
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
list_for_each_entry(fl, &ctx->flc_lease, fl_list) {
locks: __break_lease cleanup in preparation of allowing direct removal of leases Eliminate an unneeded "flock" variable. We can use "fl" as a loop cursor everywhere. Add a any_leases_conflict helper function as well to consolidate a bit of code. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-09-02 01:53:41 +07:00
if (leases_conflict(fl, breaker))
return true;
}
return false;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
/**
* __break_lease - revoke all outstanding leases on file
* @inode: the inode of the file to return
locks: implement delegations Implement NFSv4 delegations at the vfs level using the new FL_DELEG lock type. Note nfsd is the only delegation user and is only using read delegations. Warn on any attempt to set a write delegation for now. We'll come back to that case later. Acked-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2012-03-06 01:18:59 +07:00
* @mode: O_RDONLY: break only write leases; O_WRONLY or O_RDWR:
* break all leases
* @type: FL_LEASE: break leases and delegations; FL_DELEG: break
* only delegations
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
*
leases: minor break_lease() comment clarification clarify that break_lease() checks for presence of any lock, not just leases. Signed-off-by: David M. Richter <richterd@citi.umich.edu> Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu>
2007-05-10 03:10:27 +07:00
* break_lease (inlined for speed) has checked there already is at least
* some kind of lock (maybe a lease) on this file. Leases are broken on
* a call to open() or truncate(). This function can sleep unless you
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
* specified %O_NONBLOCK to your open().
*/
locks: implement delegations Implement NFSv4 delegations at the vfs level using the new FL_DELEG lock type. Note nfsd is the only delegation user and is only using read delegations. Warn on any attempt to set a write delegation for now. We'll come back to that case later. Acked-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2012-03-06 01:18:59 +07:00
int __break_lease(struct inode *inode, unsigned int mode, unsigned int type)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
locks: fix tracking of inprogress lease breaks We currently use a bit in fl_flags to record whether a lease is being broken, and set fl_type to the type (RDLCK or UNLCK) that it will eventually have. This means that once the lease break starts, we forget what the lease's type *used* to be. Breaking a read lease will then result in blocking read opens, even though there's no conflict--because the lease type is now F_UNLCK and we can no longer tell whether it was previously a read or write lease. So, instead keep fl_type as the original type (the type which we enforce), and keep track of whether we're unlocking or merely downgrading by replacing the single FL_INPROGRESS flag by FL_UNLOCK_PENDING and FL_DOWNGRADE_PENDING flags. To get this right we also need to track separate downgrade and break times, to handle the case where a write-leased file gets conflicting opens first for read, then later for write. (I first considered just eliminating the downgrade behavior completely--nfsv4 doesn't need it, and nobody as far as I can tell actually uses it currently--but Jeremy Allison tells me that Windows oplocks do behave this way, so Samba will probably use this some day.) Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-27 05:25:49 +07:00
int error = 0;
fs: fix data races on inode->i_flctx locks_get_lock_context() uses cmpxchg() to install i_flctx. cmpxchg() is a release operation which is correct. But it uses a plain load to load i_flctx. This is incorrect. Subsequent loads from i_flctx can hoist above the load of i_flctx pointer itself and observe uninitialized garbage there. This in turn can lead to corruption of ctx->flc_lock and other members. Documentation/memory-barriers.txt explicitly requires to use a barrier in such context: "A load-load control dependency requires a full read memory barrier". Use smp_load_acquire() in locks_get_lock_context() and in bunch of other functions that can proceed concurrently with locks_get_lock_context(). The data race was found with KernelThreadSanitizer (KTSAN). Signed-off-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-09-21 14:43:06 +07:00
struct file_lock_context *ctx;
locks: fix file_lock deletion inside loop locks_delete_lock_ctx() is called inside the loop, so we should use list_for_each_entry_safe. Fixes: 8634b51f6ca2 (locks: convert lease handling to file_lock_context) Signed-off-by: "Yan, Zheng" <zyan@redhat.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-03-27 09:34:20 +07:00
struct file_lock *new_fl, *fl, *tmp;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
unsigned long break_time;
Switch may_open() and break_lease() to passing O_... ... instead of mixing FMODE_ and O_ Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2009-12-24 18:47:55 +07:00
int want_write = (mode & O_ACCMODE) != O_RDONLY;
locks: move freeing of leases outside of i_lock There was only one place where we still could free a file_lock while holding the i_lock -- lease_modify. Add a new list_head argument to the lm_change operation, pass in a private list when calling it, and fix those callers to dispose of the list once the lock has been dropped. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-09-01 18:12:07 +07:00
LIST_HEAD(dispose);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
Switch may_open() and break_lease() to passing O_... ... instead of mixing FMODE_ and O_ Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2009-12-24 18:47:55 +07:00
new_fl = lease_alloc(NULL, want_write ? F_WRLCK : F_RDLCK);
vfs: fix handling of lock allocation failure in lease-break case Bruce Fields notes that commit 778fc546f749 ("locks: fix tracking of inprogress lease breaks") introduced a possible error pointer dereference on failure to allocate memory. locks_conflict() will dereference the passed-in new lease lock structure that may be an error pointer. This means an open (without O_NONBLOCK set) on a file with a lease applied (generally only done when Samba or nfsd (with v4) is running) could crash if a kmalloc() fails. So instead of playing games with IS_ERROR() all over the place, just check the allocation failure early. That makes the code more straightforward, and avoids this possible bad pointer dereference. Based-on-patch-by: J. Bruce Fields <bfields@redhat.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-12-27 01:25:26 +07:00
if (IS_ERR(new_fl))
return PTR_ERR(new_fl);
locks: implement delegations Implement NFSv4 delegations at the vfs level using the new FL_DELEG lock type. Note nfsd is the only delegation user and is only using read delegations. Warn on any attempt to set a write delegation for now. We'll come back to that case later. Acked-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2012-03-06 01:18:59 +07:00
new_fl->fl_flags = type;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
/* typically we will check that ctx is non-NULL before calling */
fs: fix data races on inode->i_flctx locks_get_lock_context() uses cmpxchg() to install i_flctx. cmpxchg() is a release operation which is correct. But it uses a plain load to load i_flctx. This is incorrect. Subsequent loads from i_flctx can hoist above the load of i_flctx pointer itself and observe uninitialized garbage there. This in turn can lead to corruption of ctx->flc_lock and other members. Documentation/memory-barriers.txt explicitly requires to use a barrier in such context: "A load-load control dependency requires a full read memory barrier". Use smp_load_acquire() in locks_get_lock_context() and in bunch of other functions that can proceed concurrently with locks_get_lock_context(). The data race was found with KernelThreadSanitizer (KTSAN). Signed-off-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-09-21 14:43:06 +07:00
ctx = smp_load_acquire(&inode->i_flctx);
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
if (!ctx) {
WARN_ON_ONCE(1);
return error;
}
fs/locks: Use percpu_down_read_preempt_disable() Avoid spurious preemption. Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Al Viro <viro@ZenIV.linux.org.uk> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: dave@stgolabs.net Cc: der.herr@hofr.at Cc: paulmck@linux.vnet.ibm.com Cc: riel@redhat.com Cc: tj@kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2016-05-30 21:48:35 +07:00
percpu_down_read_preempt_disable(&file_rwsem);
locks: add a dedicated spinlock to protect i_flctx lists We can now add a dedicated spinlock without expanding struct inode. Change to using that to protect the various i_flctx lists. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
spin_lock(&ctx->flc_lock);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
locks: move freeing of leases outside of i_lock There was only one place where we still could free a file_lock while holding the i_lock -- lease_modify. Add a new list_head argument to the lm_change operation, pass in a private list when calling it, and fix those callers to dispose of the list once the lock has been dropped. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-09-01 18:12:07 +07:00
time_out_leases(inode, &dispose);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
locks: __break_lease cleanup in preparation of allowing direct removal of leases Eliminate an unneeded "flock" variable. We can use "fl" as a loop cursor everywhere. Add a any_leases_conflict helper function as well to consolidate a bit of code. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-09-02 01:53:41 +07:00
if (!any_leases_conflict(inode, new_fl))
locks: fix tracking of inprogress lease breaks We currently use a bit in fl_flags to record whether a lease is being broken, and set fl_type to the type (RDLCK or UNLCK) that it will eventually have. This means that once the lease break starts, we forget what the lease's type *used* to be. Breaking a read lease will then result in blocking read opens, even though there's no conflict--because the lease type is now F_UNLCK and we can no longer tell whether it was previously a read or write lease. So, instead keep fl_type as the original type (the type which we enforce), and keep track of whether we're unlocking or merely downgrading by replacing the single FL_INPROGRESS flag by FL_UNLOCK_PENDING and FL_DOWNGRADE_PENDING flags. To get this right we also need to track separate downgrade and break times, to handle the case where a write-leased file gets conflicting opens first for read, then later for write. (I first considered just eliminating the downgrade behavior completely--nfsv4 doesn't need it, and nobody as far as I can tell actually uses it currently--but Jeremy Allison tells me that Windows oplocks do behave this way, so Samba will probably use this some day.) Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-27 05:25:49 +07:00
goto out;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
break_time = 0;
if (lease_break_time > 0) {
break_time = jiffies + lease_break_time * HZ;
if (break_time == 0)
break_time++; /* so that 0 means no break time */
}
locks: fix file_lock deletion inside loop locks_delete_lock_ctx() is called inside the loop, so we should use list_for_each_entry_safe. Fixes: 8634b51f6ca2 (locks: convert lease handling to file_lock_context) Signed-off-by: "Yan, Zheng" <zyan@redhat.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-03-27 09:34:20 +07:00
list_for_each_entry_safe(fl, tmp, &ctx->flc_lease, fl_list) {
locks: implement delegations Implement NFSv4 delegations at the vfs level using the new FL_DELEG lock type. Note nfsd is the only delegation user and is only using read delegations. Warn on any attempt to set a write delegation for now. We'll come back to that case later. Acked-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2012-03-06 01:18:59 +07:00
if (!leases_conflict(fl, new_fl))
continue;
locks: fix tracking of inprogress lease breaks We currently use a bit in fl_flags to record whether a lease is being broken, and set fl_type to the type (RDLCK or UNLCK) that it will eventually have. This means that once the lease break starts, we forget what the lease's type *used* to be. Breaking a read lease will then result in blocking read opens, even though there's no conflict--because the lease type is now F_UNLCK and we can no longer tell whether it was previously a read or write lease. So, instead keep fl_type as the original type (the type which we enforce), and keep track of whether we're unlocking or merely downgrading by replacing the single FL_INPROGRESS flag by FL_UNLOCK_PENDING and FL_DOWNGRADE_PENDING flags. To get this right we also need to track separate downgrade and break times, to handle the case where a write-leased file gets conflicting opens first for read, then later for write. (I first considered just eliminating the downgrade behavior completely--nfsv4 doesn't need it, and nobody as far as I can tell actually uses it currently--but Jeremy Allison tells me that Windows oplocks do behave this way, so Samba will probably use this some day.) Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-27 05:25:49 +07:00
if (want_write) {
if (fl->fl_flags & FL_UNLOCK_PENDING)
continue;
fl->fl_flags |= FL_UNLOCK_PENDING;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
fl->fl_break_time = break_time;
locks: fix tracking of inprogress lease breaks We currently use a bit in fl_flags to record whether a lease is being broken, and set fl_type to the type (RDLCK or UNLCK) that it will eventually have. This means that once the lease break starts, we forget what the lease's type *used* to be. Breaking a read lease will then result in blocking read opens, even though there's no conflict--because the lease type is now F_UNLCK and we can no longer tell whether it was previously a read or write lease. So, instead keep fl_type as the original type (the type which we enforce), and keep track of whether we're unlocking or merely downgrading by replacing the single FL_INPROGRESS flag by FL_UNLOCK_PENDING and FL_DOWNGRADE_PENDING flags. To get this right we also need to track separate downgrade and break times, to handle the case where a write-leased file gets conflicting opens first for read, then later for write. (I first considered just eliminating the downgrade behavior completely--nfsv4 doesn't need it, and nobody as far as I can tell actually uses it currently--but Jeremy Allison tells me that Windows oplocks do behave this way, so Samba will probably use this some day.) Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-27 05:25:49 +07:00
} else {
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
if (lease_breaking(fl))
locks: fix tracking of inprogress lease breaks We currently use a bit in fl_flags to record whether a lease is being broken, and set fl_type to the type (RDLCK or UNLCK) that it will eventually have. This means that once the lease break starts, we forget what the lease's type *used* to be. Breaking a read lease will then result in blocking read opens, even though there's no conflict--because the lease type is now F_UNLCK and we can no longer tell whether it was previously a read or write lease. So, instead keep fl_type as the original type (the type which we enforce), and keep track of whether we're unlocking or merely downgrading by replacing the single FL_INPROGRESS flag by FL_UNLOCK_PENDING and FL_DOWNGRADE_PENDING flags. To get this right we also need to track separate downgrade and break times, to handle the case where a write-leased file gets conflicting opens first for read, then later for write. (I first considered just eliminating the downgrade behavior completely--nfsv4 doesn't need it, and nobody as far as I can tell actually uses it currently--but Jeremy Allison tells me that Windows oplocks do behave this way, so Samba will probably use this some day.) Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-27 05:25:49 +07:00
continue;
fl->fl_flags |= FL_DOWNGRADE_PENDING;
fl->fl_downgrade_time = break_time;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
locks: give lm_break a return value Christoph suggests: "Add a return value to lm_break so that the lock manager can tell the core code "you can delete this lease right now". That gets rid of the games with the timeout which require all kinds of race avoidance code in the users." Do that here and have the nfsd lease break routine use it when it detects that there was a race between setting up the lease and it being broken. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-09-02 02:06:54 +07:00
if (fl->fl_lmops->lm_break(fl))
Revert "locks: keep a count of locks on the flctx lists" This reverts commit 9bd0f45b7037fcfa8b575c7e27d0431d6e6dc3bb. Linus rightly pointed out that I failed to initialize the counters when adding them, so they don't work as expected. Just revert this patch for now. Reported-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-02-17 02:32:03 +07:00
locks_delete_lock_ctx(fl, &dispose);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
if (list_empty(&ctx->flc_lease))
locks: give lm_break a return value Christoph suggests: "Add a return value to lm_break so that the lock manager can tell the core code "you can delete this lease right now". That gets rid of the games with the timeout which require all kinds of race avoidance code in the users." Do that here and have the nfsd lease break routine use it when it detects that there was a race between setting up the lease and it being broken. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-09-02 02:06:54 +07:00
goto out;
locks: remove i_have_this_lease check from __break_lease I think that the intent of this code was to ensure that a process won't deadlock if it has one fd open with a lease on it and then breaks that lease by opening another fd. In that case it'll treat the __break_lease call as if it were non-blocking. This seems wrong -- the process could (for instance) be multithreaded and managing different fds via different threads. I also don't see any mention of this limitation in the (somewhat sketchy) documentation. Remove the check and the non-blocking behavior when i_have_this_lease is true. Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-09-02 01:27:43 +07:00
if (mode & O_NONBLOCK) {
locks: add some tracepoints in the lease handling code v2: add a __break_lease tracepoint for non-blocking case Recently, I needed these to help track down a softlockup when recalling a delegation, but they might be helpful in other situations as well. Cc: "J. Bruce Fields" <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@poochiereds.net>
2014-05-10 01:13:05 +07:00
trace_break_lease_noblock(inode, new_fl);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
error = -EWOULDBLOCK;
goto out;
}
restart:
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
fl = list_first_entry(&ctx->flc_lease, struct file_lock, fl_list);
break_time = fl->fl_break_time;
locks: allow __break_lease to sleep even when break_time is 0 A fl->fl_break_time of 0 has a special meaning to the lease break code that basically means "never break the lease". knfsd uses this to ensure that leases don't disappear out from under it. Unfortunately, the code in __break_lease can end up passing this value to wait_event_interruptible as a timeout, which prevents it from going to sleep at all. This makes __break_lease to spin in a tight loop and causes soft lockups. Fix this by ensuring that we pass a minimum value of 1 as a timeout instead. Cc: <stable@vger.kernel.org> Cc: J. Bruce Fields <bfields@fieldses.org> Reported-by: Terry Barnaby <terry1@beam.ltd.uk> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-04-15 17:17:49 +07:00
if (break_time != 0)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
break_time -= jiffies;
locks: allow __break_lease to sleep even when break_time is 0 A fl->fl_break_time of 0 has a special meaning to the lease break code that basically means "never break the lease". knfsd uses this to ensure that leases don't disappear out from under it. Unfortunately, the code in __break_lease can end up passing this value to wait_event_interruptible as a timeout, which prevents it from going to sleep at all. This makes __break_lease to spin in a tight loop and causes soft lockups. Fix this by ensuring that we pass a minimum value of 1 as a timeout instead. Cc: <stable@vger.kernel.org> Cc: J. Bruce Fields <bfields@fieldses.org> Reported-by: Terry Barnaby <terry1@beam.ltd.uk> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-04-15 17:17:49 +07:00
if (break_time == 0)
break_time++;
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
locks_insert_block(fl, new_fl);
locks: add some tracepoints in the lease handling code v2: add a __break_lease tracepoint for non-blocking case Recently, I needed these to help track down a softlockup when recalling a delegation, but they might be helpful in other situations as well. Cc: "J. Bruce Fields" <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@poochiereds.net>
2014-05-10 01:13:05 +07:00
trace_break_lease_block(inode, new_fl);
locks: add a dedicated spinlock to protect i_flctx lists We can now add a dedicated spinlock without expanding struct inode. Change to using that to protect the various i_flctx lists. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
spin_unlock(&ctx->flc_lock);
fs/locks: Use percpu_down_read_preempt_disable() Avoid spurious preemption. Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Al Viro <viro@ZenIV.linux.org.uk> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: dave@stgolabs.net Cc: der.herr@hofr.at Cc: paulmck@linux.vnet.ibm.com Cc: riel@redhat.com Cc: tj@kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2016-05-30 21:48:35 +07:00
percpu_up_read_preempt_enable(&file_rwsem);
fs/locks: Replace lg_global with a percpu-rwsem Replace the global part of the lglock with a percpu-rwsem. Since fcl_lock is a spinlock and itself nests under i_lock, which too is a spinlock we cannot acquire sleeping locks at locks_{insert,remove}_global_locks(). We can however wrap all fcl_lock acquisitions with percpu_down_read such that all invocations of locks_{insert,remove}_global_locks() have that read lock held. This allows us to replace the lg_global part of the lglock with the write side of the rwsem. In the absense of writers, percpu_{down,up}_read() are free of atomic instructions. This further avoids the very long preempt-disable regions caused by lglock on larger machines. Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Al Viro <viro@ZenIV.linux.org.uk> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: dave@stgolabs.net Cc: der.herr@hofr.at Cc: paulmck@linux.vnet.ibm.com Cc: riel@redhat.com Cc: tj@kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-06-22 19:16:33 +07:00
locks: move freeing of leases outside of i_lock There was only one place where we still could free a file_lock while holding the i_lock -- lease_modify. Add a new list_head argument to the lm_change operation, pass in a private list when calling it, and fix those callers to dispose of the list once the lock has been dropped. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-09-01 18:12:07 +07:00
locks_dispose_list(&dispose);
file locks: Use wait_event_interruptible_timeout() interruptible_sleep_on_locked() is just an open-coded wait_event_interruptible_timeout(), with the one difference that interruptible_sleep_on_locked() doesn't bother to check the condition on which it is waiting, depending instead on the BKL to avoid the case where it blocks after the wakeup has already been called. locks_block_on_timeout() is only used in one place, so it's actually simpler to inline it into its caller. Signed-off-by: Matthew Wilcox <willy@linux.intel.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-01-15 11:28:30 +07:00
error = wait_event_interruptible_timeout(new_fl->fl_wait,
fs/locks: rename some lists and pointers. struct file lock contains an 'fl_next' pointer which is used to point to the lock that this request is blocked waiting for. So rename it to fl_blocker. The fl_blocked list_head in an active lock is the head of a list of blocked requests. In a request it is a node in that list. These are two distinct uses, so replace with two list_heads with different names. fl_blocked_requests is the head of a list of blocked requests fl_blocked_member is a node in a member of that list. The two different list_heads are never used at the same time, but that will change in a future patch. Note that a tracepoint is changed to report fl_blocker instead of fl_next. Signed-off-by: NeilBrown <neilb@suse.com> Reviewed-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Jeff Layton <jlayton@kernel.org>
2018-11-30 06:04:08 +07:00
!new_fl->fl_blocker, break_time);
fs/locks: Replace lg_global with a percpu-rwsem Replace the global part of the lglock with a percpu-rwsem. Since fcl_lock is a spinlock and itself nests under i_lock, which too is a spinlock we cannot acquire sleeping locks at locks_{insert,remove}_global_locks(). We can however wrap all fcl_lock acquisitions with percpu_down_read such that all invocations of locks_{insert,remove}_global_locks() have that read lock held. This allows us to replace the lg_global part of the lglock with the write side of the rwsem. In the absense of writers, percpu_{down,up}_read() are free of atomic instructions. This further avoids the very long preempt-disable regions caused by lglock on larger machines. Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Al Viro <viro@ZenIV.linux.org.uk> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: dave@stgolabs.net Cc: der.herr@hofr.at Cc: paulmck@linux.vnet.ibm.com Cc: riel@redhat.com Cc: tj@kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-06-22 19:16:33 +07:00
fs/locks: Use percpu_down_read_preempt_disable() Avoid spurious preemption. Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Al Viro <viro@ZenIV.linux.org.uk> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: dave@stgolabs.net Cc: der.herr@hofr.at Cc: paulmck@linux.vnet.ibm.com Cc: riel@redhat.com Cc: tj@kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2016-05-30 21:48:35 +07:00
percpu_down_read_preempt_disable(&file_rwsem);
locks: add a dedicated spinlock to protect i_flctx lists We can now add a dedicated spinlock without expanding struct inode. Change to using that to protect the various i_flctx lists. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
spin_lock(&ctx->flc_lock);
locks: add some tracepoints in the lease handling code v2: add a __break_lease tracepoint for non-blocking case Recently, I needed these to help track down a softlockup when recalling a delegation, but they might be helpful in other situations as well. Cc: "J. Bruce Fields" <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@poochiereds.net>
2014-05-10 01:13:05 +07:00
trace_break_lease_unblock(inode, new_fl);
locks: protect most of the file_lock handling with i_lock Having a global lock that protects all of this code is a clear scalability problem. Instead of doing that, move most of the code to be protected by the i_lock instead. The exceptions are the global lists that the ->fl_link sits on, and the ->fl_block list. ->fl_link is what connects these structures to the global lists, so we must ensure that we hold those locks when iterating over or updating these lists. Furthermore, sound deadlock detection requires that we hold the blocked_list state steady while checking for loops. We also must ensure that the search and update to the list are atomic. For the checking and insertion side of the blocked_list, push the acquisition of the global lock into __posix_lock_file and ensure that checking and update of the blocked_list is done without dropping the lock in between. On the removal side, when waking up blocked lock waiters, take the global lock before walking the blocked list and dequeue the waiters from the global list prior to removal from the fl_block list. With this, deadlock detection should be race free while we minimize excessive file_lock_lock thrashing. Finally, in order to avoid a lock inversion problem when handling /proc/locks output we must ensure that manipulations of the fl_block list are also protected by the file_lock_lock. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:15 +07:00
locks_delete_block(new_fl);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
if (error >= 0) {
locks: fix tracking of inprogress lease breaks We currently use a bit in fl_flags to record whether a lease is being broken, and set fl_type to the type (RDLCK or UNLCK) that it will eventually have. This means that once the lease break starts, we forget what the lease's type *used* to be. Breaking a read lease will then result in blocking read opens, even though there's no conflict--because the lease type is now F_UNLCK and we can no longer tell whether it was previously a read or write lease. So, instead keep fl_type as the original type (the type which we enforce), and keep track of whether we're unlocking or merely downgrading by replacing the single FL_INPROGRESS flag by FL_UNLOCK_PENDING and FL_DOWNGRADE_PENDING flags. To get this right we also need to track separate downgrade and break times, to handle the case where a write-leased file gets conflicting opens first for read, then later for write. (I first considered just eliminating the downgrade behavior completely--nfsv4 doesn't need it, and nobody as far as I can tell actually uses it currently--but Jeremy Allison tells me that Windows oplocks do behave this way, so Samba will probably use this some day.) Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-27 05:25:49 +07:00
/*
* Wait for the next conflicting lease that has not been
* broken yet
*/
locks: __break_lease cleanup in preparation of allowing direct removal of leases Eliminate an unneeded "flock" variable. We can use "fl" as a loop cursor everywhere. Add a any_leases_conflict helper function as well to consolidate a bit of code. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-09-02 01:53:41 +07:00
if (error == 0)
time_out_leases(inode, &dispose);
if (any_leases_conflict(inode, new_fl))
goto restart;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
error = 0;
}
out:
locks: add a dedicated spinlock to protect i_flctx lists We can now add a dedicated spinlock without expanding struct inode. Change to using that to protect the various i_flctx lists. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
spin_unlock(&ctx->flc_lock);
fs/locks: Use percpu_down_read_preempt_disable() Avoid spurious preemption. Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Al Viro <viro@ZenIV.linux.org.uk> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: dave@stgolabs.net Cc: der.herr@hofr.at Cc: paulmck@linux.vnet.ibm.com Cc: riel@redhat.com Cc: tj@kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2016-05-30 21:48:35 +07:00
percpu_up_read_preempt_enable(&file_rwsem);
locks: move freeing of leases outside of i_lock There was only one place where we still could free a file_lock while holding the i_lock -- lease_modify. Add a new list_head argument to the lm_change operation, pass in a private list when calling it, and fix those callers to dispose of the list once the lock has been dropped. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-09-01 18:12:07 +07:00
locks_dispose_list(&dispose);
vfs: fix handling of lock allocation failure in lease-break case Bruce Fields notes that commit 778fc546f749 ("locks: fix tracking of inprogress lease breaks") introduced a possible error pointer dereference on failure to allocate memory. locks_conflict() will dereference the passed-in new lease lock structure that may be an error pointer. This means an open (without O_NONBLOCK set) on a file with a lease applied (generally only done when Samba or nfsd (with v4) is running) could crash if a kmalloc() fails. So instead of playing games with IS_ERROR() all over the place, just check the allocation failure early. That makes the code more straightforward, and avoids this possible bad pointer dereference. Based-on-patch-by: J. Bruce Fields <bfields@redhat.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-12-27 01:25:26 +07:00
locks_free_lock(new_fl);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
return error;
}
EXPORT_SYMBOL(__break_lease);
/**
nfsd: encode stat->mtime for getattr instead of inode->i_mtime The values of stat->mtime and inode->i_mtime may differ for overlayfs and stat->mtime is the correct value to use when encoding getattr. This is also consistent with the fact that other attr times are also encoded from stat values. Both callers of lease_get_mtime() already have the value of stat->mtime, so the only needed change is that lease_get_mtime() will not overwrite this value with inode->i_mtime in case the inode does not have an exclusive lease. Signed-off-by: Amir Goldstein <amir73il@gmail.com> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-01-03 22:14:34 +07:00
* lease_get_mtime - update modified time of an inode with exclusive lease
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
* @inode: the inode
nfsd: encode stat->mtime for getattr instead of inode->i_mtime The values of stat->mtime and inode->i_mtime may differ for overlayfs and stat->mtime is the correct value to use when encoding getattr. This is also consistent with the fact that other attr times are also encoded from stat values. Both callers of lease_get_mtime() already have the value of stat->mtime, so the only needed change is that lease_get_mtime() will not overwrite this value with inode->i_mtime in case the inode does not have an exclusive lease. Signed-off-by: Amir Goldstein <amir73il@gmail.com> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2018-01-03 22:14:34 +07:00
* @time: pointer to a timespec which contains the last modified time
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
*
* This is to force NFS clients to flush their caches for files with
* exclusive leases. The justification is that if someone has an
fs: fix kernel-doc notation warnings Fix kernel-doc notation warnings in fs/. Warning(mmotm-2008-0314-1449//fs/super.c:560): missing initial short description on line: * mark_files_ro Warning(mmotm-2008-0314-1449//fs/locks.c:1277): missing initial short description on line: * lease_get_mtime Warning(mmotm-2008-0314-1449//fs/locks.c:1277): missing initial short description on line: * lease_get_mtime Warning(mmotm-2008-0314-1449//fs/namei.c:1368): missing initial short description on line: * lookup_one_len: filesystem helper to lookup single pathname component Warning(mmotm-2008-0314-1449//fs/buffer.c:3221): missing initial short description on line: * bh_uptodate_or_lock: Test whether the buffer is uptodate Warning(mmotm-2008-0314-1449//fs/buffer.c:3240): missing initial short description on line: * bh_submit_read: Submit a locked buffer for reading Warning(mmotm-2008-0314-1449//fs/fs-writeback.c:30): missing initial short description on line: * writeback_acquire: attempt to get exclusive writeback access to a device Warning(mmotm-2008-0314-1449//fs/fs-writeback.c:47): missing initial short description on line: * writeback_in_progress: determine whether there is writeback in progress Warning(mmotm-2008-0314-1449//fs/fs-writeback.c:58): missing initial short description on line: * writeback_release: relinquish exclusive writeback access against a device. Warning(mmotm-2008-0314-1449//include/linux/jbd.h:351): contents before sections Warning(mmotm-2008-0314-1449//include/linux/jbd.h:561): contents before sections Warning(mmotm-2008-0314-1449//fs/jbd/transaction.c:1935): missing initial short description on line: * void journal_invalidatepage() Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-03-20 07:01:00 +07:00
* exclusive lease, then they could be modifying it.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
*/
vfs: change inode times to use struct timespec64 struct timespec is not y2038 safe. Transition vfs to use y2038 safe struct timespec64 instead. The change was made with the help of the following cocinelle script. This catches about 80% of the changes. All the header file and logic changes are included in the first 5 rules. The rest are trivial substitutions. I avoid changing any of the function signatures or any other filesystem specific data structures to keep the patch simple for review. The script can be a little shorter by combining different cases. But, this version was sufficient for my usecase. virtual patch @ depends on patch @ identifier now; @@ - struct timespec + struct timespec64 current_time ( ... ) { - struct timespec now = current_kernel_time(); + struct timespec64 now = current_kernel_time64(); ... - return timespec_trunc( + return timespec64_trunc( ... ); } @ depends on patch @ identifier xtime; @@ struct \( iattr \| inode \| kstat \) { ... - struct timespec xtime; + struct timespec64 xtime; ... } @ depends on patch @ identifier t; @@ struct inode_operations { ... int (*update_time) (..., - struct timespec t, + struct timespec64 t, ...); ... } @ depends on patch @ identifier t; identifier fn_update_time =~ "update_time$"; @@ fn_update_time (..., - struct timespec *t, + struct timespec64 *t, ...) { ... } @ depends on patch @ identifier t; @@ lease_get_mtime( ... , - struct timespec *t + struct timespec64 *t ) { ... } @te depends on patch forall@ identifier ts; local idexpression struct inode *inode_node; identifier i_xtime =~ "^i_[acm]time$"; identifier ia_xtime =~ "^ia_[acm]time$"; identifier fn_update_time =~ "update_time$"; identifier fn; expression e, E3; local idexpression struct inode *node1; local idexpression struct inode *node2; local idexpression struct iattr *attr1; local idexpression struct iattr *attr2; local idexpression struct iattr attr; identifier i_xtime1 =~ "^i_[acm]time$"; identifier i_xtime2 =~ "^i_[acm]time$"; identifier ia_xtime1 =~ "^ia_[acm]time$"; identifier ia_xtime2 =~ "^ia_[acm]time$"; @@ ( ( - struct timespec ts; + struct timespec64 ts; | - struct timespec ts = current_time(inode_node); + struct timespec64 ts = current_time(inode_node); ) <+... when != ts ( - timespec_equal(&inode_node->i_xtime, &ts) + timespec64_equal(&inode_node->i_xtime, &ts) | - timespec_equal(&ts, &inode_node->i_xtime) + timespec64_equal(&ts, &inode_node->i_xtime) | - timespec_compare(&inode_node->i_xtime, &ts) + timespec64_compare(&inode_node->i_xtime, &ts) | - timespec_compare(&ts, &inode_node->i_xtime) + timespec64_compare(&ts, &inode_node->i_xtime) | ts = current_time(e) | fn_update_time(..., &ts,...) | inode_node->i_xtime = ts | node1->i_xtime = ts | ts = inode_node->i_xtime | <+... attr1->ia_xtime ...+> = ts | ts = attr1->ia_xtime | ts.tv_sec | ts.tv_nsec | btrfs_set_stack_timespec_sec(..., ts.tv_sec) | btrfs_set_stack_timespec_nsec(..., ts.tv_nsec) | - ts = timespec64_to_timespec( + ts = ... -) | - ts = ktime_to_timespec( + ts = ktime_to_timespec64( ...) | - ts = E3 + ts = timespec_to_timespec64(E3) | - ktime_get_real_ts(&ts) + ktime_get_real_ts64(&ts) | fn(..., - ts + timespec64_to_timespec(ts) ,...) ) ...+> ( <... when != ts - return ts; + return timespec64_to_timespec(ts); ...> ) | - timespec_equal(&node1->i_xtime1, &node2->i_xtime2) + timespec64_equal(&node1->i_xtime2, &node2->i_xtime2) | - timespec_equal(&node1->i_xtime1, &attr2->ia_xtime2) + timespec64_equal(&node1->i_xtime2, &attr2->ia_xtime2) | - timespec_compare(&node1->i_xtime1, &node2->i_xtime2) + timespec64_compare(&node1->i_xtime1, &node2->i_xtime2) | node1->i_xtime1 = - timespec_trunc(attr1->ia_xtime1, + timespec64_trunc(attr1->ia_xtime1, ...) | - attr1->ia_xtime1 = timespec_trunc(attr2->ia_xtime2, + attr1->ia_xtime1 = timespec64_trunc(attr2->ia_xtime2, ...) | - ktime_get_real_ts(&attr1->ia_xtime1) + ktime_get_real_ts64(&attr1->ia_xtime1) | - ktime_get_real_ts(&attr.ia_xtime1) + ktime_get_real_ts64(&attr.ia_xtime1) ) @ depends on patch @ struct inode *node; struct iattr *attr; identifier fn; identifier i_xtime =~ "^i_[acm]time$"; identifier ia_xtime =~ "^ia_[acm]time$"; expression e; @@ ( - fn(node->i_xtime); + fn(timespec64_to_timespec(node->i_xtime)); | fn(..., - node->i_xtime); + timespec64_to_timespec(node->i_xtime)); | - e = fn(attr->ia_xtime); + e = fn(timespec64_to_timespec(attr->ia_xtime)); ) @ depends on patch forall @ struct inode *node; struct iattr *attr; identifier i_xtime =~ "^i_[acm]time$"; identifier ia_xtime =~ "^ia_[acm]time$"; identifier fn; @@ { + struct timespec ts; <+... ( + ts = timespec64_to_timespec(node->i_xtime); fn (..., - &node->i_xtime, + &ts, ...); | + ts = timespec64_to_timespec(attr->ia_xtime); fn (..., - &attr->ia_xtime, + &ts, ...); ) ...+> } @ depends on patch forall @ struct inode *node; struct iattr *attr; struct kstat *stat; identifier ia_xtime =~ "^ia_[acm]time$"; identifier i_xtime =~ "^i_[acm]time$"; identifier xtime =~ "^[acm]time$"; identifier fn, ret; @@ { + struct timespec ts; <+... ( + ts = timespec64_to_timespec(node->i_xtime); ret = fn (..., - &node->i_xtime, + &ts, ...); | + ts = timespec64_to_timespec(node->i_xtime); ret = fn (..., - &node->i_xtime); + &ts); | + ts = timespec64_to_timespec(attr->ia_xtime); ret = fn (..., - &attr->ia_xtime, + &ts, ...); | + ts = timespec64_to_timespec(attr->ia_xtime); ret = fn (..., - &attr->ia_xtime); + &ts); | + ts = timespec64_to_timespec(stat->xtime); ret = fn (..., - &stat->xtime); + &ts); ) ...+> } @ depends on patch @ struct inode *node; struct inode *node2; identifier i_xtime1 =~ "^i_[acm]time$"; identifier i_xtime2 =~ "^i_[acm]time$"; identifier i_xtime3 =~ "^i_[acm]time$"; struct iattr *attrp; struct iattr *attrp2; struct iattr attr ; identifier ia_xtime1 =~ "^ia_[acm]time$"; identifier ia_xtime2 =~ "^ia_[acm]time$"; struct kstat *stat; struct kstat stat1; struct timespec64 ts; identifier xtime =~ "^[acmb]time$"; expression e; @@ ( ( node->i_xtime2 \| attrp->ia_xtime2 \| attr.ia_xtime2 \) = node->i_xtime1 ; | node->i_xtime2 = \( node2->i_xtime1 \| timespec64_trunc(...) \); | node->i_xtime2 = node->i_xtime1 = node->i_xtime3 = \(ts \| current_time(...) \); | node->i_xtime1 = node->i_xtime3 = \(ts \| current_time(...) \); | stat->xtime = node2->i_xtime1; | stat1.xtime = node2->i_xtime1; | ( node->i_xtime2 \| attrp->ia_xtime2 \) = attrp->ia_xtime1 ; | ( attrp->ia_xtime1 \| attr.ia_xtime1 \) = attrp2->ia_xtime2; | - e = node->i_xtime1; + e = timespec64_to_timespec( node->i_xtime1 ); | - e = attrp->ia_xtime1; + e = timespec64_to_timespec( attrp->ia_xtime1 ); | node->i_xtime1 = current_time(...); | node->i_xtime2 = node->i_xtime1 = node->i_xtime3 = - e; + timespec_to_timespec64(e); | node->i_xtime1 = node->i_xtime3 = - e; + timespec_to_timespec64(e); | - node->i_xtime1 = e; + node->i_xtime1 = timespec_to_timespec64(e); ) Signed-off-by: Deepa Dinamani <deepa.kernel@gmail.com> Cc: <anton@tuxera.com> Cc: <balbi@kernel.org> Cc: <bfields@fieldses.org> Cc: <darrick.wong@oracle.com> Cc: <dhowells@redhat.com> Cc: <dsterba@suse.com> Cc: <dwmw2@infradead.org> Cc: <hch@lst.de> Cc: <hirofumi@mail.parknet.co.jp> Cc: <hubcap@omnibond.com> Cc: <jack@suse.com> Cc: <jaegeuk@kernel.org> Cc: <jaharkes@cs.cmu.edu> Cc: <jslaby@suse.com> Cc: <keescook@chromium.org> Cc: <mark@fasheh.com> Cc: <miklos@szeredi.hu> Cc: <nico@linaro.org> Cc: <reiserfs-devel@vger.kernel.org> Cc: <richard@nod.at> Cc: <sage@redhat.com> Cc: <sfrench@samba.org> Cc: <swhiteho@redhat.com> Cc: <tj@kernel.org> Cc: <trond.myklebust@primarydata.com> Cc: <tytso@mit.edu> Cc: <viro@zeniv.linux.org.uk>
2018-05-09 09:36:02 +07:00
void lease_get_mtime(struct inode *inode, struct timespec64 *time)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
locks: close potential race in lease_get_mtime lease_get_mtime is called without the i_lock held, so there's no guarantee about the stability of the list. Between the time when we assign "flock" and then dereference it to check whether it's a lease and for write, the lease could be freed. Ensure that that doesn't occur by taking the i_lock before trying to check the lease. Cc: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-08-22 21:18:44 +07:00
bool has_lease = false;
fs: fix data races on inode->i_flctx locks_get_lock_context() uses cmpxchg() to install i_flctx. cmpxchg() is a release operation which is correct. But it uses a plain load to load i_flctx. This is incorrect. Subsequent loads from i_flctx can hoist above the load of i_flctx pointer itself and observe uninitialized garbage there. This in turn can lead to corruption of ctx->flc_lock and other members. Documentation/memory-barriers.txt explicitly requires to use a barrier in such context: "A load-load control dependency requires a full read memory barrier". Use smp_load_acquire() in locks_get_lock_context() and in bunch of other functions that can proceed concurrently with locks_get_lock_context(). The data race was found with KernelThreadSanitizer (KTSAN). Signed-off-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-09-21 14:43:06 +07:00
struct file_lock_context *ctx;
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
struct file_lock *fl;
locks: close potential race in lease_get_mtime lease_get_mtime is called without the i_lock held, so there's no guarantee about the stability of the list. Between the time when we assign "flock" and then dereference it to check whether it's a lease and for write, the lease could be freed. Ensure that that doesn't occur by taking the i_lock before trying to check the lease. Cc: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-08-22 21:18:44 +07:00
fs: fix data races on inode->i_flctx locks_get_lock_context() uses cmpxchg() to install i_flctx. cmpxchg() is a release operation which is correct. But it uses a plain load to load i_flctx. This is incorrect. Subsequent loads from i_flctx can hoist above the load of i_flctx pointer itself and observe uninitialized garbage there. This in turn can lead to corruption of ctx->flc_lock and other members. Documentation/memory-barriers.txt explicitly requires to use a barrier in such context: "A load-load control dependency requires a full read memory barrier". Use smp_load_acquire() in locks_get_lock_context() and in bunch of other functions that can proceed concurrently with locks_get_lock_context(). The data race was found with KernelThreadSanitizer (KTSAN). Signed-off-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-09-21 14:43:06 +07:00
ctx = smp_load_acquire(&inode->i_flctx);
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
if (ctx && !list_empty_careful(&ctx->flc_lease)) {
locks: add a dedicated spinlock to protect i_flctx lists We can now add a dedicated spinlock without expanding struct inode. Change to using that to protect the various i_flctx lists. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
spin_lock(&ctx->flc_lock);
locks: use list_first_entry_or_null() Simplify the code with list_first_entry_or_null(). Signed-off-by: Geliang Tang <geliangtang@163.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-11-18 20:40:33 +07:00
fl = list_first_entry_or_null(&ctx->flc_lease,
struct file_lock, fl_list);
if (fl && (fl->fl_type == F_WRLCK))
has_lease = true;
locks: add a dedicated spinlock to protect i_flctx lists We can now add a dedicated spinlock without expanding struct inode. Change to using that to protect the various i_flctx lists. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
spin_unlock(&ctx->flc_lock);
locks: close potential race in lease_get_mtime lease_get_mtime is called without the i_lock held, so there's no guarantee about the stability of the list. Between the time when we assign "flock" and then dereference it to check whether it's a lease and for write, the lease could be freed. Ensure that that doesn't occur by taking the i_lock before trying to check the lease. Cc: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-08-22 21:18:44 +07:00
}
if (has_lease)
fs: Replace current_fs_time() with current_time() current_fs_time() uses struct super_block* as an argument. As per Linus's suggestion, this is changed to take struct inode* as a parameter instead. This is because the function is primarily meant for vfs inode timestamps. Also the function was renamed as per Arnd's suggestion. Change all calls to current_fs_time() to use the new current_time() function instead. current_fs_time() will be deleted. Signed-off-by: Deepa Dinamani <deepa.kernel@gmail.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2016-09-14 21:48:06 +07:00
*time = current_time(inode);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
EXPORT_SYMBOL(lease_get_mtime);
/**
* fcntl_getlease - Enquire what lease is currently active
* @filp: the file
*
* The value returned by this function will be one of
* (if no lease break is pending):
*
* %F_RDLCK to indicate a shared lease is held.
*
* %F_WRLCK to indicate an exclusive lease is held.
*
* %F_UNLCK to indicate no lease is held.
*
* (if a lease break is pending):
*
* %F_RDLCK to indicate an exclusive lease needs to be
* changed to a shared lease (or removed).
*
* %F_UNLCK to indicate the lease needs to be removed.
*
* XXX: sfr & willy disagree over whether F_INPROGRESS
* should be returned to userspace.
*/
int fcntl_getlease(struct file *filp)
{
struct file_lock *fl;
locks: fix file locking on overlayfs This patch allows flock, posix locks, ofd locks and leases to work correctly on overlayfs. Instead of using the underlying inode for storing lock context use the overlay inode. This allows locks to be persistent across copy-up. This is done by introducing locks_inode() helper and using it instead of file_inode() to get the inode in locking code. For non-overlayfs the two are equivalent, except for an extra pointer dereference in locks_inode(). Since lock operations are in "struct file_operations" we must also make sure not to call underlying filesystem's lock operations. Introcude a super block flag MS_NOREMOTELOCK to this effect. Signed-off-by: Miklos Szeredi <mszeredi@redhat.com> Acked-by: Jeff Layton <jlayton@poochiereds.net> Cc: "J. Bruce Fields" <bfields@fieldses.org>
2016-09-16 17:44:20 +07:00
struct inode *inode = locks_inode(filp);
fs: fix data races on inode->i_flctx locks_get_lock_context() uses cmpxchg() to install i_flctx. cmpxchg() is a release operation which is correct. But it uses a plain load to load i_flctx. This is incorrect. Subsequent loads from i_flctx can hoist above the load of i_flctx pointer itself and observe uninitialized garbage there. This in turn can lead to corruption of ctx->flc_lock and other members. Documentation/memory-barriers.txt explicitly requires to use a barrier in such context: "A load-load control dependency requires a full read memory barrier". Use smp_load_acquire() in locks_get_lock_context() and in bunch of other functions that can proceed concurrently with locks_get_lock_context(). The data race was found with KernelThreadSanitizer (KTSAN). Signed-off-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-09-21 14:43:06 +07:00
struct file_lock_context *ctx;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
int type = F_UNLCK;
locks: move freeing of leases outside of i_lock There was only one place where we still could free a file_lock while holding the i_lock -- lease_modify. Add a new list_head argument to the lm_change operation, pass in a private list when calling it, and fix those callers to dispose of the list once the lock has been dropped. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-09-01 18:12:07 +07:00
LIST_HEAD(dispose);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
fs: fix data races on inode->i_flctx locks_get_lock_context() uses cmpxchg() to install i_flctx. cmpxchg() is a release operation which is correct. But it uses a plain load to load i_flctx. This is incorrect. Subsequent loads from i_flctx can hoist above the load of i_flctx pointer itself and observe uninitialized garbage there. This in turn can lead to corruption of ctx->flc_lock and other members. Documentation/memory-barriers.txt explicitly requires to use a barrier in such context: "A load-load control dependency requires a full read memory barrier". Use smp_load_acquire() in locks_get_lock_context() and in bunch of other functions that can proceed concurrently with locks_get_lock_context(). The data race was found with KernelThreadSanitizer (KTSAN). Signed-off-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-09-21 14:43:06 +07:00
ctx = smp_load_acquire(&inode->i_flctx);
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
if (ctx && !list_empty_careful(&ctx->flc_lease)) {
locking, fs/locks: Add missing file_sem locks I overlooked a few code-paths that can lead to locks_delete_global_locks(). Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Reviewed-by: Jeff Layton <jlayton@poochiereds.net> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Bruce Fields <bfields@fieldses.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: linux-fsdevel@vger.kernel.org Cc: syzkaller <syzkaller@googlegroups.com> Link: http://lkml.kernel.org/r/20161008081228.GF3142@twins.programming.kicks-ass.net Signed-off-by: Ingo Molnar <mingo@kernel.org>
2016-10-08 15:12:28 +07:00
percpu_down_read_preempt_disable(&file_rwsem);
locks: add a dedicated spinlock to protect i_flctx lists We can now add a dedicated spinlock without expanding struct inode. Change to using that to protect the various i_flctx lists. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
spin_lock(&ctx->flc_lock);
locks: fix file locking on overlayfs This patch allows flock, posix locks, ofd locks and leases to work correctly on overlayfs. Instead of using the underlying inode for storing lock context use the overlay inode. This allows locks to be persistent across copy-up. This is done by introducing locks_inode() helper and using it instead of file_inode() to get the inode in locking code. For non-overlayfs the two are equivalent, except for an extra pointer dereference in locks_inode(). Since lock operations are in "struct file_operations" we must also make sure not to call underlying filesystem's lock operations. Introcude a super block flag MS_NOREMOTELOCK to this effect. Signed-off-by: Miklos Szeredi <mszeredi@redhat.com> Acked-by: Jeff Layton <jlayton@poochiereds.net> Cc: "J. Bruce Fields" <bfields@fieldses.org>
2016-09-16 17:44:20 +07:00
time_out_leases(inode, &dispose);
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
list_for_each_entry(fl, &ctx->flc_lease, fl_list) {
if (fl->fl_file != filp)
continue;
locks: fix tracking of inprogress lease breaks We currently use a bit in fl_flags to record whether a lease is being broken, and set fl_type to the type (RDLCK or UNLCK) that it will eventually have. This means that once the lease break starts, we forget what the lease's type *used* to be. Breaking a read lease will then result in blocking read opens, even though there's no conflict--because the lease type is now F_UNLCK and we can no longer tell whether it was previously a read or write lease. So, instead keep fl_type as the original type (the type which we enforce), and keep track of whether we're unlocking or merely downgrading by replacing the single FL_INPROGRESS flag by FL_UNLOCK_PENDING and FL_DOWNGRADE_PENDING flags. To get this right we also need to track separate downgrade and break times, to handle the case where a write-leased file gets conflicting opens first for read, then later for write. (I first considered just eliminating the downgrade behavior completely--nfsv4 doesn't need it, and nobody as far as I can tell actually uses it currently--but Jeremy Allison tells me that Windows oplocks do behave this way, so Samba will probably use this some day.) Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-27 05:25:49 +07:00
type = target_leasetype(fl);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
break;
}
locks: add a dedicated spinlock to protect i_flctx lists We can now add a dedicated spinlock without expanding struct inode. Change to using that to protect the various i_flctx lists. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
spin_unlock(&ctx->flc_lock);
locking, fs/locks: Add missing file_sem locks I overlooked a few code-paths that can lead to locks_delete_global_locks(). Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Reviewed-by: Jeff Layton <jlayton@poochiereds.net> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Bruce Fields <bfields@fieldses.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: linux-fsdevel@vger.kernel.org Cc: syzkaller <syzkaller@googlegroups.com> Link: http://lkml.kernel.org/r/20161008081228.GF3142@twins.programming.kicks-ass.net Signed-off-by: Ingo Molnar <mingo@kernel.org>
2016-10-08 15:12:28 +07:00
percpu_up_read_preempt_enable(&file_rwsem);
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
locks_dispose_list(&dispose);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
return type;
}
locks: close potential race between setlease and open As Al Viro points out, there is an unlikely, but possible race between opening a file and setting a lease on it. generic_add_lease is done with the i_lock held, but the inode->i_flock check in break_lease is lockless. It's possible for another task doing an open to do the entire pathwalk and call break_lease between the point where generic_add_lease checks for a conflicting open and adds the lease to the list. If this occurs, we can end up with a lease set on the file with a conflicting open. To guard against that, check again for a conflicting open after adding the lease to the i_flock list. If the above race occurs, then we can simply unwind the lease setting and return -EAGAIN. Because we take dentry references and acquire write access on the file before calling break_lease, we know that if the i_flock list is empty when the open caller goes to check it then the necessary refcounts have already been incremented. Thus the additional check for a conflicting open will see that there is one and the setlease call will fail. Cc: Bruce Fields <bfields@fieldses.org> Cc: David Howells <dhowells@redhat.com> Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Reported-by: Al Viro <viro@ZenIV.linux.org.uk> Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@fieldses.org>
2014-02-04 00:13:06 +07:00
/**
* check_conflicting_open - see if the given dentry points to a file that has
* an existing open that would conflict with the
* desired lease.
* @dentry: dentry to check
* @arg: type of lease that we're trying to acquire
fs: fix fs/locks.c kernel-doc warning Fix kernel-doc warnings in fs/locks.c: Warning(..//fs/locks.c:1577): No description found for parameter 'flags' Signed-off-by: Randy Dunlap <rdunlap@infradead.org> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-08-10 08:43:17 +07:00
* @flags: current lock flags
locks: close potential race between setlease and open As Al Viro points out, there is an unlikely, but possible race between opening a file and setting a lease on it. generic_add_lease is done with the i_lock held, but the inode->i_flock check in break_lease is lockless. It's possible for another task doing an open to do the entire pathwalk and call break_lease between the point where generic_add_lease checks for a conflicting open and adds the lease to the list. If this occurs, we can end up with a lease set on the file with a conflicting open. To guard against that, check again for a conflicting open after adding the lease to the i_flock list. If the above race occurs, then we can simply unwind the lease setting and return -EAGAIN. Because we take dentry references and acquire write access on the file before calling break_lease, we know that if the i_flock list is empty when the open caller goes to check it then the necessary refcounts have already been incremented. Thus the additional check for a conflicting open will see that there is one and the setlease call will fail. Cc: Bruce Fields <bfields@fieldses.org> Cc: David Howells <dhowells@redhat.com> Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Reported-by: Al Viro <viro@ZenIV.linux.org.uk> Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@fieldses.org>
2014-02-04 00:13:06 +07:00
*
* Check to see if there's an existing open fd on this file that would
* conflict with the lease we're trying to set.
*/
static int
fs: add FL_LAYOUT lease type This (ab-)uses the file locking code to allow filesystems to recall outstanding pNFS layouts on a file. This new lease type is similar but not quite the same as FL_DELEG. A FL_LAYOUT lease can always be granted, an a per-filesystem lock (XFS iolock for the initial implementation) ensures not FL_LAYOUT leases granted when we would need to recall them. Also included are changes that allow multiple outstanding read leases of different types on the same file as long as they have a differnt owner. This wasn't a problem until now as nfsd never set FL_LEASE leases, and no one else used FL_DELEG leases, but given that nfsd will also issues FL_LAYOUT leases we will have to handle it now. Signed-off-by: Christoph Hellwig <hch@lst.de>
2015-01-22 01:17:03 +07:00
check_conflicting_open(const struct dentry *dentry, const long arg, int flags)
locks: close potential race between setlease and open As Al Viro points out, there is an unlikely, but possible race between opening a file and setting a lease on it. generic_add_lease is done with the i_lock held, but the inode->i_flock check in break_lease is lockless. It's possible for another task doing an open to do the entire pathwalk and call break_lease between the point where generic_add_lease checks for a conflicting open and adds the lease to the list. If this occurs, we can end up with a lease set on the file with a conflicting open. To guard against that, check again for a conflicting open after adding the lease to the i_flock list. If the above race occurs, then we can simply unwind the lease setting and return -EAGAIN. Because we take dentry references and acquire write access on the file before calling break_lease, we know that if the i_flock list is empty when the open caller goes to check it then the necessary refcounts have already been incremented. Thus the additional check for a conflicting open will see that there is one and the setlease call will fail. Cc: Bruce Fields <bfields@fieldses.org> Cc: David Howells <dhowells@redhat.com> Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Reported-by: Al Viro <viro@ZenIV.linux.org.uk> Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@fieldses.org>
2014-02-04 00:13:06 +07:00
{
int ret = 0;
struct inode *inode = dentry->d_inode;
fs: add FL_LAYOUT lease type This (ab-)uses the file locking code to allow filesystems to recall outstanding pNFS layouts on a file. This new lease type is similar but not quite the same as FL_DELEG. A FL_LAYOUT lease can always be granted, an a per-filesystem lock (XFS iolock for the initial implementation) ensures not FL_LAYOUT leases granted when we would need to recall them. Also included are changes that allow multiple outstanding read leases of different types on the same file as long as they have a differnt owner. This wasn't a problem until now as nfsd never set FL_LEASE leases, and no one else used FL_DELEG leases, but given that nfsd will also issues FL_LAYOUT leases we will have to handle it now. Signed-off-by: Christoph Hellwig <hch@lst.de>
2015-01-22 01:17:03 +07:00
if (flags & FL_LAYOUT)
return 0;
Revert "vfs: do get_write_access() on upper layer of overlayfs" This reverts commit 4d0c5ba2ff79ef9f5188998b29fd28fcb05f3667. We now get write access on both overlay and underlying layers so this patch is no longer needed for correct operation. Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
2018-07-18 20:44:43 +07:00
if ((arg == F_RDLCK) && (atomic_read(&inode->i_writecount) > 0))
locks: close potential race between setlease and open As Al Viro points out, there is an unlikely, but possible race between opening a file and setting a lease on it. generic_add_lease is done with the i_lock held, but the inode->i_flock check in break_lease is lockless. It's possible for another task doing an open to do the entire pathwalk and call break_lease between the point where generic_add_lease checks for a conflicting open and adds the lease to the list. If this occurs, we can end up with a lease set on the file with a conflicting open. To guard against that, check again for a conflicting open after adding the lease to the i_flock list. If the above race occurs, then we can simply unwind the lease setting and return -EAGAIN. Because we take dentry references and acquire write access on the file before calling break_lease, we know that if the i_flock list is empty when the open caller goes to check it then the necessary refcounts have already been incremented. Thus the additional check for a conflicting open will see that there is one and the setlease call will fail. Cc: Bruce Fields <bfields@fieldses.org> Cc: David Howells <dhowells@redhat.com> Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Reported-by: Al Viro <viro@ZenIV.linux.org.uk> Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@fieldses.org>
2014-02-04 00:13:06 +07:00
return -EAGAIN;
if ((arg == F_WRLCK) && ((d_count(dentry) > 1) ||
(atomic_read(&inode->i_count) > 1)))
ret = -EAGAIN;
return ret;
}
locks: plumb a "priv" pointer into the setlease routines In later patches, we're going to add a new lock_manager_operation to finish setting up the lease while still holding the i_lock. To do this, we'll need to pass a little bit of info in the fcntl setlease case (primarily an fasync structure). Plumb the extra pointer into there in advance of that. We declare this pointer as a void ** to make it clear that this is private info, and that the caller isn't required to set this unless the lm_setup specifically requires it. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-08-22 21:40:25 +07:00
static int
generic_add_lease(struct file *filp, long arg, struct file_lock **flp, void **priv)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
struct file_lock *fl, *my_fl = NULL, *lease;
[PATCH] VFS: change struct file to use struct path This patch changes struct file to use struct path instead of having independent pointers to struct dentry and struct vfsmount, and converts all users of f_{dentry,vfsmnt} in fs/ to use f_path.{dentry,mnt}. Additionally, it adds two #define's to make the transition easier for users of the f_dentry and f_vfsmnt. Signed-off-by: Josef "Jeff" Sipek <jsipek@cs.sunysb.edu> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-08 17:36:35 +07:00
struct dentry *dentry = filp->f_path.dentry;
locks: fix file locking on overlayfs This patch allows flock, posix locks, ofd locks and leases to work correctly on overlayfs. Instead of using the underlying inode for storing lock context use the overlay inode. This allows locks to be persistent across copy-up. This is done by introducing locks_inode() helper and using it instead of file_inode() to get the inode in locking code. For non-overlayfs the two are equivalent, except for an extra pointer dereference in locks_inode(). Since lock operations are in "struct file_operations" we must also make sure not to call underlying filesystem's lock operations. Introcude a super block flag MS_NOREMOTELOCK to this effect. Signed-off-by: Miklos Szeredi <mszeredi@redhat.com> Acked-by: Jeff Layton <jlayton@poochiereds.net> Cc: "J. Bruce Fields" <bfields@fieldses.org>
2016-09-16 17:44:20 +07:00
struct inode *inode = dentry->d_inode;
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
struct file_lock_context *ctx;
locks: implement delegations Implement NFSv4 delegations at the vfs level using the new FL_DELEG lock type. Note nfsd is the only delegation user and is only using read delegations. Warn on any attempt to set a write delegation for now. We'll come back to that case later. Acked-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2012-03-06 01:18:59 +07:00
bool is_deleg = (*flp)->fl_flags & FL_DELEG;
locks: setlease cleanup There's an incorrect comment here. Also clean up the logic: the "rdlease" and "wrlease" locals are confusingly named, and don't really add anything since we can make a decision as soon as we hit one of these cases. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-19 21:59:49 +07:00
int error;
locks: move freeing of leases outside of i_lock There was only one place where we still could free a file_lock while holding the i_lock -- lease_modify. Add a new list_head argument to the lm_change operation, pass in a private list when calling it, and fix those callers to dispose of the list once the lock has been dropped. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-09-01 18:12:07 +07:00
LIST_HEAD(dispose);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
locks: fix leaks on setlease errors We're depending on setlease to free the passed-in lease on failure. Signed-off-by: J. Bruce Fields <bfields@redhat.com> Acked-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-10-31 04:31:14 +07:00
lease = *flp;
locks: add some tracepoints in the lease handling code v2: add a __break_lease tracepoint for non-blocking case Recently, I needed these to help track down a softlockup when recalling a delegation, but they might be helpful in other situations as well. Cc: "J. Bruce Fields" <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@poochiereds.net>
2014-05-10 01:13:05 +07:00
trace_generic_add_lease(inode, lease);
locks: don't allocate a lock context for an F_UNLCK request In the event that we get an F_UNLCK request on an inode that has no lock context, there is no reason to allocate one. Change locks_get_lock_context to take a "type" pointer and avoid allocating a new context if it's F_UNLCK. Then, fix the callers to return appropriately if that function returns NULL. Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2015-04-03 20:04:03 +07:00
/* Note that arg is never F_UNLCK here */
ctx = locks_get_lock_context(inode, arg);
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
if (!ctx)
return -ENOMEM;
locks: implement delegations Implement NFSv4 delegations at the vfs level using the new FL_DELEG lock type. Note nfsd is the only delegation user and is only using read delegations. Warn on any attempt to set a write delegation for now. We'll come back to that case later. Acked-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2012-03-06 01:18:59 +07:00
/*
* In the delegation case we need mutual exclusion with
* a number of operations that take the i_mutex. We trylock
* because delegations are an optional optimization, and if
* there's some chance of a conflict--we'd rather not
* bother, maybe that's a sign this just isn't a good file to
* hand out a delegation on.
*/
wrappers for ->i_mutex access parallel to mutex_{lock,unlock,trylock,is_locked,lock_nested}, inode_foo(inode) being mutex_foo(&inode->i_mutex). Please, use those for access to ->i_mutex; over the coming cycle ->i_mutex will become rwsem, with ->lookup() done with it held only shared. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2016-01-23 03:40:57 +07:00
if (is_deleg && !inode_trylock(inode))
locks: implement delegations Implement NFSv4 delegations at the vfs level using the new FL_DELEG lock type. Note nfsd is the only delegation user and is only using read delegations. Warn on any attempt to set a write delegation for now. We'll come back to that case later. Acked-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2012-03-06 01:18:59 +07:00
return -EAGAIN;
if (is_deleg && arg == F_WRLCK) {
/* Write delegations are not currently supported: */
wrappers for ->i_mutex access parallel to mutex_{lock,unlock,trylock,is_locked,lock_nested}, inode_foo(inode) being mutex_foo(&inode->i_mutex). Please, use those for access to ->i_mutex; over the coming cycle ->i_mutex will become rwsem, with ->lookup() done with it held only shared. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2016-01-23 03:40:57 +07:00
inode_unlock(inode);
locks: implement delegations Implement NFSv4 delegations at the vfs level using the new FL_DELEG lock type. Note nfsd is the only delegation user and is only using read delegations. Warn on any attempt to set a write delegation for now. We'll come back to that case later. Acked-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2012-03-06 01:18:59 +07:00
WARN_ON_ONCE(1);
return -EINVAL;
}
locks: fix leaks on setlease errors We're depending on setlease to free the passed-in lease on failure. Signed-off-by: J. Bruce Fields <bfields@redhat.com> Acked-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-10-31 04:31:14 +07:00
fs/locks: Use percpu_down_read_preempt_disable() Avoid spurious preemption. Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Al Viro <viro@ZenIV.linux.org.uk> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: dave@stgolabs.net Cc: der.herr@hofr.at Cc: paulmck@linux.vnet.ibm.com Cc: riel@redhat.com Cc: tj@kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2016-05-30 21:48:35 +07:00
percpu_down_read_preempt_disable(&file_rwsem);
locks: add a dedicated spinlock to protect i_flctx lists We can now add a dedicated spinlock without expanding struct inode. Change to using that to protect the various i_flctx lists. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
spin_lock(&ctx->flc_lock);
locks: move freeing of leases outside of i_lock There was only one place where we still could free a file_lock while holding the i_lock -- lease_modify. Add a new list_head argument to the lm_change operation, pass in a private list when calling it, and fix those callers to dispose of the list once the lock has been dropped. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-09-01 18:12:07 +07:00
time_out_leases(inode, &dispose);
fs: add FL_LAYOUT lease type This (ab-)uses the file locking code to allow filesystems to recall outstanding pNFS layouts on a file. This new lease type is similar but not quite the same as FL_DELEG. A FL_LAYOUT lease can always be granted, an a per-filesystem lock (XFS iolock for the initial implementation) ensures not FL_LAYOUT leases granted when we would need to recall them. Also included are changes that allow multiple outstanding read leases of different types on the same file as long as they have a differnt owner. This wasn't a problem until now as nfsd never set FL_LEASE leases, and no one else used FL_DELEG leases, but given that nfsd will also issues FL_LAYOUT leases we will have to handle it now. Signed-off-by: Christoph Hellwig <hch@lst.de>
2015-01-22 01:17:03 +07:00
error = check_conflicting_open(dentry, arg, lease->fl_flags);
locks: close potential race between setlease and open As Al Viro points out, there is an unlikely, but possible race between opening a file and setting a lease on it. generic_add_lease is done with the i_lock held, but the inode->i_flock check in break_lease is lockless. It's possible for another task doing an open to do the entire pathwalk and call break_lease between the point where generic_add_lease checks for a conflicting open and adds the lease to the list. If this occurs, we can end up with a lease set on the file with a conflicting open. To guard against that, check again for a conflicting open after adding the lease to the i_flock list. If the above race occurs, then we can simply unwind the lease setting and return -EAGAIN. Because we take dentry references and acquire write access on the file before calling break_lease, we know that if the i_flock list is empty when the open caller goes to check it then the necessary refcounts have already been incremented. Thus the additional check for a conflicting open will see that there is one and the setlease call will fail. Cc: Bruce Fields <bfields@fieldses.org> Cc: David Howells <dhowells@redhat.com> Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Reported-by: Al Viro <viro@ZenIV.linux.org.uk> Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@fieldses.org>
2014-02-04 00:13:06 +07:00
if (error)
locks: fix leaks on setlease errors We're depending on setlease to free the passed-in lease on failure. Signed-off-by: J. Bruce Fields <bfields@redhat.com> Acked-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-10-31 04:31:14 +07:00
goto out;
locks: share more common lease code Share more code between setlease (used by nfsd) and fcntl. Also some minor cleanup. Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu> Acked-by: Christoph Hellwig <hch@infradead.org>
2007-06-01 04:03:46 +07:00
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
/*
* At this point, we know that if there is an exclusive
* lease on this file, then we hold it on this filp
* (otherwise our open of this file would have blocked).
* And if we are trying to acquire an exclusive lease,
* then the file is not open by anyone (including us)
* except for this filp.
*/
locks: setlease cleanup There's an incorrect comment here. Also clean up the logic: the "rdlease" and "wrlease" locals are confusingly named, and don't really add anything since we can make a decision as soon as we hit one of these cases. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-19 21:59:49 +07:00
error = -EAGAIN;
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
list_for_each_entry(fl, &ctx->flc_lease, fl_list) {
fs: track fl_owner for leases Just like for other lock types we should allow different owners to have a read lease on a file. Currently this can't happen, but with the addition of pNFS layout leases we'll need this feature. Signed-off-by: Christoph Hellwig <hch@lst.de>
2015-01-22 01:14:02 +07:00
if (fl->fl_file == filp &&
fl->fl_owner == lease->fl_owner) {
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
my_fl = fl;
locks: setlease cleanup There's an incorrect comment here. Also clean up the logic: the "rdlease" and "wrlease" locals are confusingly named, and don't really add anything since we can make a decision as soon as we hit one of these cases. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-19 21:59:49 +07:00
continue;
}
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
locks: setlease cleanup There's an incorrect comment here. Also clean up the logic: the "rdlease" and "wrlease" locals are confusingly named, and don't really add anything since we can make a decision as soon as we hit one of these cases. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-08-19 21:59:49 +07:00
/*
* No exclusive leases if someone else has a lease on
* this file:
*/
if (arg == F_WRLCK)
goto out;
/*
* Modifying our existing lease is OK, but no getting a
* new lease if someone else is opening for write:
*/
if (fl->fl_flags & FL_UNLOCK_PENDING)
goto out;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
if (my_fl != NULL) {
locks: fix fasync_struct memory leak in lease upgrade/downgrade handling Commit 8634b51f6ca2 (locks: convert lease handling to file_lock_context) introduced a regression in the handling of lease upgrade/downgrades. In the event that we already have a lease on a file and are going to either upgrade or downgrade it, we skip doing any list insertion or deletion and simply re-call lm_setup on the existing lease. As of commit 8634b51f6ca2 however, we end up calling lm_setup on the lease that was passed in, instead of on the existing lease. This causes us to leak the fasync_struct that was allocated in the event that there was not already an existing one (as it always appeared that there wasn't one). Fixes: 8634b51f6ca2 (locks: convert lease handling to file_lock_context) Reported-and-Tested-by: Daniel Wagner <daniel.wagner@bmw-carit.de> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-03-05 05:34:32 +07:00
lease = my_fl;
error = lease->fl_lmops->lm_change(lease, arg, &dispose);
locks: define a lm_setup handler for leases ...and move the fasync setup into it for fcntl lease calls. At the same time, change the semantics of how the file_lock double-pointer is handled. Up until now, on a successful lease return you got a pointer to the lock on the list. This is bad, since that pointer can no longer be relied on as valid once the inode->i_lock has been released. Change the code to instead just zero out the pointer if the lease we passed in ended up being used. Then the callers can just check to see if it's NULL after the call and free it if it isn't. The priv argument has the same semantics. The lm_setup function can zero the pointer out to signal to the caller that it should not be freed after the function returns. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-08-22 21:55:47 +07:00
if (error)
goto out;
goto out_setup;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
error = -EINVAL;
if (!leases_enable)
goto out;
Revert "locks: keep a count of locks on the flctx lists" This reverts commit 9bd0f45b7037fcfa8b575c7e27d0431d6e6dc3bb. Linus rightly pointed out that I failed to initialize the counters when adding them, so they don't work as expected. Just revert this patch for now. Reported-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-02-17 02:32:03 +07:00
locks_insert_lock_ctx(lease, &ctx->flc_lease);
locks: close potential race between setlease and open As Al Viro points out, there is an unlikely, but possible race between opening a file and setting a lease on it. generic_add_lease is done with the i_lock held, but the inode->i_flock check in break_lease is lockless. It's possible for another task doing an open to do the entire pathwalk and call break_lease between the point where generic_add_lease checks for a conflicting open and adds the lease to the list. If this occurs, we can end up with a lease set on the file with a conflicting open. To guard against that, check again for a conflicting open after adding the lease to the i_flock list. If the above race occurs, then we can simply unwind the lease setting and return -EAGAIN. Because we take dentry references and acquire write access on the file before calling break_lease, we know that if the i_flock list is empty when the open caller goes to check it then the necessary refcounts have already been incremented. Thus the additional check for a conflicting open will see that there is one and the setlease call will fail. Cc: Bruce Fields <bfields@fieldses.org> Cc: David Howells <dhowells@redhat.com> Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Reported-by: Al Viro <viro@ZenIV.linux.org.uk> Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@fieldses.org>
2014-02-04 00:13:06 +07:00
/*
* The check in break_lease() is lockless. It's possible for another
* open to race in after we did the earlier check for a conflicting
* open but before the lease was inserted. Check again for a
* conflicting open and cancel the lease if there is one.
*
* We also add a barrier here to ensure that the insertion of the lock
* precedes these checks.
*/
smp_mb();
fs: add FL_LAYOUT lease type This (ab-)uses the file locking code to allow filesystems to recall outstanding pNFS layouts on a file. This new lease type is similar but not quite the same as FL_DELEG. A FL_LAYOUT lease can always be granted, an a per-filesystem lock (XFS iolock for the initial implementation) ensures not FL_LAYOUT leases granted when we would need to recall them. Also included are changes that allow multiple outstanding read leases of different types on the same file as long as they have a differnt owner. This wasn't a problem until now as nfsd never set FL_LEASE leases, and no one else used FL_DELEG leases, but given that nfsd will also issues FL_LAYOUT leases we will have to handle it now. Signed-off-by: Christoph Hellwig <hch@lst.de>
2015-01-22 01:17:03 +07:00
error = check_conflicting_open(dentry, arg, lease->fl_flags);
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
if (error) {
Revert "locks: keep a count of locks on the flctx lists" This reverts commit 9bd0f45b7037fcfa8b575c7e27d0431d6e6dc3bb. Linus rightly pointed out that I failed to initialize the counters when adding them, so they don't work as expected. Just revert this patch for now. Reported-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-02-17 02:32:03 +07:00
locks_unlink_lock_ctx(lease);
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
goto out;
}
locks: define a lm_setup handler for leases ...and move the fasync setup into it for fcntl lease calls. At the same time, change the semantics of how the file_lock double-pointer is handled. Up until now, on a successful lease return you got a pointer to the lock on the list. This is bad, since that pointer can no longer be relied on as valid once the inode->i_lock has been released. Change the code to instead just zero out the pointer if the lease we passed in ended up being used. Then the callers can just check to see if it's NULL after the call and free it if it isn't. The priv argument has the same semantics. The lm_setup function can zero the pointer out to signal to the caller that it should not be freed after the function returns. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-08-22 21:55:47 +07:00
out_setup:
if (lease->fl_lmops->lm_setup)
lease->fl_lmops->lm_setup(lease, priv);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
out:
locks: add a dedicated spinlock to protect i_flctx lists We can now add a dedicated spinlock without expanding struct inode. Change to using that to protect the various i_flctx lists. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
spin_unlock(&ctx->flc_lock);
fs/locks: Use percpu_down_read_preempt_disable() Avoid spurious preemption. Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Al Viro <viro@ZenIV.linux.org.uk> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: dave@stgolabs.net Cc: der.herr@hofr.at Cc: paulmck@linux.vnet.ibm.com Cc: riel@redhat.com Cc: tj@kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2016-05-30 21:48:35 +07:00
percpu_up_read_preempt_enable(&file_rwsem);
locks: move freeing of leases outside of i_lock There was only one place where we still could free a file_lock while holding the i_lock -- lease_modify. Add a new list_head argument to the lm_change operation, pass in a private list when calling it, and fix those callers to dispose of the list once the lock has been dropped. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-09-01 18:12:07 +07:00
locks_dispose_list(&dispose);
locks: implement delegations Implement NFSv4 delegations at the vfs level using the new FL_DELEG lock type. Note nfsd is the only delegation user and is only using read delegations. Warn on any attempt to set a write delegation for now. We'll come back to that case later. Acked-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2012-03-06 01:18:59 +07:00
if (is_deleg)
wrappers for ->i_mutex access parallel to mutex_{lock,unlock,trylock,is_locked,lock_nested}, inode_foo(inode) being mutex_foo(&inode->i_mutex). Please, use those for access to ->i_mutex; over the coming cycle ->i_mutex will become rwsem, with ->lookup() done with it held only shared. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2016-01-23 03:40:57 +07:00
inode_unlock(inode);
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
if (!error && !my_fl)
locks: define a lm_setup handler for leases ...and move the fasync setup into it for fcntl lease calls. At the same time, change the semantics of how the file_lock double-pointer is handled. Up until now, on a successful lease return you got a pointer to the lock on the list. This is bad, since that pointer can no longer be relied on as valid once the inode->i_lock has been released. Change the code to instead just zero out the pointer if the lease we passed in ended up being used. Then the callers can just check to see if it's NULL after the call and free it if it isn't. The priv argument has the same semantics. The lm_setup function can zero the pointer out to signal to the caller that it should not be freed after the function returns. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-08-22 21:55:47 +07:00
*flp = NULL;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
return error;
}
leases: split up generic_setlease into lock/unlock cases Eventually we should probably do the same thing to the file operations as well. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-09-21 19:34:32 +07:00
fs: track fl_owner for leases Just like for other lock types we should allow different owners to have a read lease on a file. Currently this can't happen, but with the addition of pNFS layout leases we'll need this feature. Signed-off-by: Christoph Hellwig <hch@lst.de>
2015-01-22 01:14:02 +07:00
static int generic_delete_lease(struct file *filp, void *owner)
leases: split up generic_setlease into lock/unlock cases Eventually we should probably do the same thing to the file operations as well. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-09-21 19:34:32 +07:00
{
locks: generic_delete_lease doesn't need a file_lock at all Ensure that it's OK to pass in a NULL file_lock double pointer on a F_UNLCK request and convert the vfs_setlease F_UNLCK callers to do just that. Finally, turn the BUG_ON in generic_setlease into a WARN_ON_ONCE with an error return. That's a problem we can handle without crashing the box if it occurs. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-08-22 21:18:45 +07:00
int error = -EAGAIN;
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
struct file_lock *fl, *victim = NULL;
locks: fix file locking on overlayfs This patch allows flock, posix locks, ofd locks and leases to work correctly on overlayfs. Instead of using the underlying inode for storing lock context use the overlay inode. This allows locks to be persistent across copy-up. This is done by introducing locks_inode() helper and using it instead of file_inode() to get the inode in locking code. For non-overlayfs the two are equivalent, except for an extra pointer dereference in locks_inode(). Since lock operations are in "struct file_operations" we must also make sure not to call underlying filesystem's lock operations. Introcude a super block flag MS_NOREMOTELOCK to this effect. Signed-off-by: Miklos Szeredi <mszeredi@redhat.com> Acked-by: Jeff Layton <jlayton@poochiereds.net> Cc: "J. Bruce Fields" <bfields@fieldses.org>
2016-09-16 17:44:20 +07:00
struct inode *inode = locks_inode(filp);
fs: fix data races on inode->i_flctx locks_get_lock_context() uses cmpxchg() to install i_flctx. cmpxchg() is a release operation which is correct. But it uses a plain load to load i_flctx. This is incorrect. Subsequent loads from i_flctx can hoist above the load of i_flctx pointer itself and observe uninitialized garbage there. This in turn can lead to corruption of ctx->flc_lock and other members. Documentation/memory-barriers.txt explicitly requires to use a barrier in such context: "A load-load control dependency requires a full read memory barrier". Use smp_load_acquire() in locks_get_lock_context() and in bunch of other functions that can proceed concurrently with locks_get_lock_context(). The data race was found with KernelThreadSanitizer (KTSAN). Signed-off-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-09-21 14:43:06 +07:00
struct file_lock_context *ctx;
locks: move freeing of leases outside of i_lock There was only one place where we still could free a file_lock while holding the i_lock -- lease_modify. Add a new list_head argument to the lm_change operation, pass in a private list when calling it, and fix those callers to dispose of the list once the lock has been dropped. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-09-01 18:12:07 +07:00
LIST_HEAD(dispose);
leases: split up generic_setlease into lock/unlock cases Eventually we should probably do the same thing to the file operations as well. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-09-21 19:34:32 +07:00
fs: fix data races on inode->i_flctx locks_get_lock_context() uses cmpxchg() to install i_flctx. cmpxchg() is a release operation which is correct. But it uses a plain load to load i_flctx. This is incorrect. Subsequent loads from i_flctx can hoist above the load of i_flctx pointer itself and observe uninitialized garbage there. This in turn can lead to corruption of ctx->flc_lock and other members. Documentation/memory-barriers.txt explicitly requires to use a barrier in such context: "A load-load control dependency requires a full read memory barrier". Use smp_load_acquire() in locks_get_lock_context() and in bunch of other functions that can proceed concurrently with locks_get_lock_context(). The data race was found with KernelThreadSanitizer (KTSAN). Signed-off-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-09-21 14:43:06 +07:00
ctx = smp_load_acquire(&inode->i_flctx);
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
if (!ctx) {
trace_generic_delete_lease(inode, NULL);
return error;
}
fs/locks: Use percpu_down_read_preempt_disable() Avoid spurious preemption. Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Al Viro <viro@ZenIV.linux.org.uk> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: dave@stgolabs.net Cc: der.herr@hofr.at Cc: paulmck@linux.vnet.ibm.com Cc: riel@redhat.com Cc: tj@kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2016-05-30 21:48:35 +07:00
percpu_down_read_preempt_disable(&file_rwsem);
locks: add a dedicated spinlock to protect i_flctx lists We can now add a dedicated spinlock without expanding struct inode. Change to using that to protect the various i_flctx lists. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
spin_lock(&ctx->flc_lock);
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
list_for_each_entry(fl, &ctx->flc_lease, fl_list) {
fs: track fl_owner for leases Just like for other lock types we should allow different owners to have a read lease on a file. Currently this can't happen, but with the addition of pNFS layout leases we'll need this feature. Signed-off-by: Christoph Hellwig <hch@lst.de>
2015-01-22 01:14:02 +07:00
if (fl->fl_file == filp &&
fl->fl_owner == owner) {
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
victim = fl;
locks: generic_delete_lease doesn't need a file_lock at all Ensure that it's OK to pass in a NULL file_lock double pointer on a F_UNLCK request and convert the vfs_setlease F_UNLCK callers to do just that. Finally, turn the BUG_ON in generic_setlease into a WARN_ON_ONCE with an error return. That's a problem we can handle without crashing the box if it occurs. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-08-22 21:18:45 +07:00
break;
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
}
leases: split up generic_setlease into lock/unlock cases Eventually we should probably do the same thing to the file operations as well. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-09-21 19:34:32 +07:00
}
locks: fix generic_delete_lease tracepoint to use victim pointer It's possible that "fl" won't point at a valid lock at this point, so use "victim" instead which is either a valid lock or NULL. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-03-14 20:45:35 +07:00
trace_generic_delete_lease(inode, victim);
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
if (victim)
locks: clean up the lm_change prototype Now that we use standard list_heads for tracking leases, we can have lm_change take a pointer to the lease to be modified instead of a double pointer. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
error = fl->fl_lmops->lm_change(victim, F_UNLCK, &dispose);
locks: add a dedicated spinlock to protect i_flctx lists We can now add a dedicated spinlock without expanding struct inode. Change to using that to protect the various i_flctx lists. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
spin_unlock(&ctx->flc_lock);
fs/locks: Use percpu_down_read_preempt_disable() Avoid spurious preemption. Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Al Viro <viro@ZenIV.linux.org.uk> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: dave@stgolabs.net Cc: der.herr@hofr.at Cc: paulmck@linux.vnet.ibm.com Cc: riel@redhat.com Cc: tj@kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2016-05-30 21:48:35 +07:00
percpu_up_read_preempt_enable(&file_rwsem);
locks: move freeing of leases outside of i_lock There was only one place where we still could free a file_lock while holding the i_lock -- lease_modify. Add a new list_head argument to the lm_change operation, pass in a private list when calling it, and fix those callers to dispose of the list once the lock has been dropped. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-09-01 18:12:07 +07:00
locks_dispose_list(&dispose);
locks: generic_delete_lease doesn't need a file_lock at all Ensure that it's OK to pass in a NULL file_lock double pointer on a F_UNLCK request and convert the vfs_setlease F_UNLCK callers to do just that. Finally, turn the BUG_ON in generic_setlease into a WARN_ON_ONCE with an error return. That's a problem we can handle without crashing the box if it occurs. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-08-22 21:18:45 +07:00
return error;
leases: split up generic_setlease into lock/unlock cases Eventually we should probably do the same thing to the file operations as well. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-09-21 19:34:32 +07:00
}
/**
* generic_setlease - sets a lease on an open file
locks: define a lm_setup handler for leases ...and move the fasync setup into it for fcntl lease calls. At the same time, change the semantics of how the file_lock double-pointer is handled. Up until now, on a successful lease return you got a pointer to the lock on the list. This is bad, since that pointer can no longer be relied on as valid once the inode->i_lock has been released. Change the code to instead just zero out the pointer if the lease we passed in ended up being used. Then the callers can just check to see if it's NULL after the call and free it if it isn't. The priv argument has the same semantics. The lm_setup function can zero the pointer out to signal to the caller that it should not be freed after the function returns. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-08-22 21:55:47 +07:00
* @filp: file pointer
* @arg: type of lease to obtain
* @flp: input - file_lock to use, output - file_lock inserted
* @priv: private data for lm_setup (may be NULL if lm_setup
* doesn't require it)
leases: split up generic_setlease into lock/unlock cases Eventually we should probably do the same thing to the file operations as well. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-09-21 19:34:32 +07:00
*
* The (input) flp->fl_lmops->lm_break function is required
* by break_lease().
*/
locks: plumb a "priv" pointer into the setlease routines In later patches, we're going to add a new lock_manager_operation to finish setting up the lease while still holding the i_lock. To do this, we'll need to pass a little bit of info in the fcntl setlease case (primarily an fasync structure). Plumb the extra pointer into there in advance of that. We declare this pointer as a void ** to make it clear that this is private info, and that the caller isn't required to set this unless the lm_setup specifically requires it. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-08-22 21:40:25 +07:00
int generic_setlease(struct file *filp, long arg, struct file_lock **flp,
void **priv)
leases: split up generic_setlease into lock/unlock cases Eventually we should probably do the same thing to the file operations as well. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-09-21 19:34:32 +07:00
{
locks: fix file locking on overlayfs This patch allows flock, posix locks, ofd locks and leases to work correctly on overlayfs. Instead of using the underlying inode for storing lock context use the overlay inode. This allows locks to be persistent across copy-up. This is done by introducing locks_inode() helper and using it instead of file_inode() to get the inode in locking code. For non-overlayfs the two are equivalent, except for an extra pointer dereference in locks_inode(). Since lock operations are in "struct file_operations" we must also make sure not to call underlying filesystem's lock operations. Introcude a super block flag MS_NOREMOTELOCK to this effect. Signed-off-by: Miklos Szeredi <mszeredi@redhat.com> Acked-by: Jeff Layton <jlayton@poochiereds.net> Cc: "J. Bruce Fields" <bfields@fieldses.org>
2016-09-16 17:44:20 +07:00
struct inode *inode = locks_inode(filp);
leases: split up generic_setlease into lock/unlock cases Eventually we should probably do the same thing to the file operations as well. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-09-21 19:34:32 +07:00
int error;
userns: Use uid_eq gid_eq helpers when comparing kuids and kgids in the vfs Acked-by: Serge Hallyn <serge.hallyn@canonical.com> Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
2012-03-04 12:17:15 +07:00
if ((!uid_eq(current_fsuid(), inode->i_uid)) && !capable(CAP_LEASE))
leases: split up generic_setlease into lock/unlock cases Eventually we should probably do the same thing to the file operations as well. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-09-21 19:34:32 +07:00
return -EACCES;
if (!S_ISREG(inode->i_mode))
return -EINVAL;
error = security_file_lock(filp, arg);
if (error)
return error;
switch (arg) {
case F_UNLCK:
fs: track fl_owner for leases Just like for other lock types we should allow different owners to have a read lease on a file. Currently this can't happen, but with the addition of pNFS layout leases we'll need this feature. Signed-off-by: Christoph Hellwig <hch@lst.de>
2015-01-22 01:14:02 +07:00
return generic_delete_lease(filp, *priv);
leases: split up generic_setlease into lock/unlock cases Eventually we should probably do the same thing to the file operations as well. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-09-21 19:34:32 +07:00
case F_RDLCK:
case F_WRLCK:
locks: generic_delete_lease doesn't need a file_lock at all Ensure that it's OK to pass in a NULL file_lock double pointer on a F_UNLCK request and convert the vfs_setlease F_UNLCK callers to do just that. Finally, turn the BUG_ON in generic_setlease into a WARN_ON_ONCE with an error return. That's a problem we can handle without crashing the box if it occurs. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-08-22 21:18:45 +07:00
if (!(*flp)->fl_lmops->lm_break) {
WARN_ON_ONCE(1);
return -ENOLCK;
}
fs: add FL_LAYOUT lease type This (ab-)uses the file locking code to allow filesystems to recall outstanding pNFS layouts on a file. This new lease type is similar but not quite the same as FL_DELEG. A FL_LAYOUT lease can always be granted, an a per-filesystem lock (XFS iolock for the initial implementation) ensures not FL_LAYOUT leases granted when we would need to recall them. Also included are changes that allow multiple outstanding read leases of different types on the same file as long as they have a differnt owner. This wasn't a problem until now as nfsd never set FL_LEASE leases, and no one else used FL_DELEG leases, but given that nfsd will also issues FL_LAYOUT leases we will have to handle it now. Signed-off-by: Christoph Hellwig <hch@lst.de>
2015-01-22 01:17:03 +07:00
locks: plumb a "priv" pointer into the setlease routines In later patches, we're going to add a new lock_manager_operation to finish setting up the lease while still holding the i_lock. To do this, we'll need to pass a little bit of info in the fcntl setlease case (primarily an fasync structure). Plumb the extra pointer into there in advance of that. We declare this pointer as a void ** to make it clear that this is private info, and that the caller isn't required to set this unless the lm_setup specifically requires it. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-08-22 21:40:25 +07:00
return generic_add_lease(filp, arg, flp, priv);
leases: split up generic_setlease into lock/unlock cases Eventually we should probably do the same thing to the file operations as well. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-09-21 19:34:32 +07:00
default:
Remove easily user-triggerable BUG from generic_setlease This can be trivially triggered from userspace by passing in something unexpected. kernel BUG at fs/locks.c:1468! invalid opcode: 0000 [#1] SMP RIP: 0010:generic_setlease+0xc2/0x100 Call Trace: __vfs_setlease+0x35/0x40 fcntl_setlease+0x76/0x150 sys_fcntl+0x1c6/0x810 system_call_fastpath+0x1a/0x1f Signed-off-by: Dave Jones <davej@redhat.com> Cc: stable@kernel.org # 3.2+ Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-07-14 00:35:36 +07:00
return -EINVAL;
leases: split up generic_setlease into lock/unlock cases Eventually we should probably do the same thing to the file operations as well. Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-09-21 19:34:32 +07:00
}
}
rename setlease to generic_setlease Make it a little more clear that this is the default implementation for the setleast operation. Signed-off-by: Christoph Hellwig <hch@lst.de> Cc: Steven Whitehouse <swhiteho@redhat.com> Acked-by: "J. Bruce Fields" <bfields@fieldses.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-31 14:39:22 +07:00
EXPORT_SYMBOL(generic_setlease);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
fs/locks.c: prepare for BKL removal This prepares the removal of the big kernel lock from the file locking code. We still use the BKL as long as fs/lockd uses it and ceph might sleep, but we can flip the definition to a private spinlock as soon as that's done. All users outside of fs/lockd get converted to use lock_flocks() instead of lock_kernel() where appropriate. Based on an earlier patch to use a spinlock from Matthew Wilcox, who has attempted this a few times before, the earliest patch from over 10 years ago turned it into a semaphore, which ended up being slower than the BKL and was subsequently reverted. Someone should do some serious performance testing when this becomes a spinlock, since this has caused problems before. Using a spinlock should be at least as good as the BKL in theory, but who knows... Signed-off-by: Arnd Bergmann <arnd@arndb.de> Acked-by: Matthew Wilcox <willy@linux.intel.com> Cc: Christoph Hellwig <hch@lst.de> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Miklos Szeredi <mszeredi@suse.cz> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: John Kacur <jkacur@redhat.com> Cc: Sage Weil <sage@newdream.net> Cc: linux-kernel@vger.kernel.org Cc: linux-fsdevel@vger.kernel.org
2010-09-18 20:09:31 +07:00
/**
locks: clean up vfs_setlease kerneldoc comments Some of the latter paragraphs seem ambiguous and just plain wrong. In particular the break_lease comment makes no sense. We call break_lease (and break_deleg) from all sorts of vfs-layer functions, so there is clearly such a method. Also get rid of some of the other comments about what's needed for a full implementation. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-08-23 05:13:28 +07:00
* vfs_setlease - sets a lease on an open file
locks: define a lm_setup handler for leases ...and move the fasync setup into it for fcntl lease calls. At the same time, change the semantics of how the file_lock double-pointer is handled. Up until now, on a successful lease return you got a pointer to the lock on the list. This is bad, since that pointer can no longer be relied on as valid once the inode->i_lock has been released. Change the code to instead just zero out the pointer if the lease we passed in ended up being used. Then the callers can just check to see if it's NULL after the call and free it if it isn't. The priv argument has the same semantics. The lm_setup function can zero the pointer out to signal to the caller that it should not be freed after the function returns. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-08-22 21:55:47 +07:00
* @filp: file pointer
* @arg: type of lease to obtain
* @lease: file_lock to use when adding a lease
* @priv: private info for lm_setup when adding a lease (may be
* NULL if lm_setup doesn't require it)
locks: clean up vfs_setlease kerneldoc comments Some of the latter paragraphs seem ambiguous and just plain wrong. In particular the break_lease comment makes no sense. We call break_lease (and break_deleg) from all sorts of vfs-layer functions, so there is clearly such a method. Also get rid of some of the other comments about what's needed for a full implementation. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-08-23 05:13:28 +07:00
*
* Call this to establish a lease on the file. The "lease" argument is not
* used for F_UNLCK requests and may be NULL. For commands that set or alter
fs: locks: Fix some troubles at kernel-doc comments There are a few syntax violations that cause outputs of a few comments to not be properly parsed in ReST format. No functional changes. Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-05-27 17:07:18 +07:00
* an existing lease, the ``(*lease)->fl_lmops->lm_break`` operation must be
* set; if not, this function will return -ENOLCK (and generate a scary-looking
locks: clean up vfs_setlease kerneldoc comments Some of the latter paragraphs seem ambiguous and just plain wrong. In particular the break_lease comment makes no sense. We call break_lease (and break_deleg) from all sorts of vfs-layer functions, so there is clearly such a method. Also get rid of some of the other comments about what's needed for a full implementation. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-08-23 05:13:28 +07:00
* stack trace).
locks: define a lm_setup handler for leases ...and move the fasync setup into it for fcntl lease calls. At the same time, change the semantics of how the file_lock double-pointer is handled. Up until now, on a successful lease return you got a pointer to the lock on the list. This is bad, since that pointer can no longer be relied on as valid once the inode->i_lock has been released. Change the code to instead just zero out the pointer if the lease we passed in ended up being used. Then the callers can just check to see if it's NULL after the call and free it if it isn't. The priv argument has the same semantics. The lm_setup function can zero the pointer out to signal to the caller that it should not be freed after the function returns. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-08-22 21:55:47 +07:00
*
* The "priv" pointer is passed directly to the lm_setup function as-is. It
* may be NULL if the lm_setup operation doesn't require it.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
*/
locks: plumb a "priv" pointer into the setlease routines In later patches, we're going to add a new lock_manager_operation to finish setting up the lease while still holding the i_lock. To do this, we'll need to pass a little bit of info in the fcntl setlease case (primarily an fasync structure). Plumb the extra pointer into there in advance of that. We declare this pointer as a void ** to make it clear that this is private info, and that the caller isn't required to set this unless the lm_setup specifically requires it. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-08-22 21:40:25 +07:00
int
vfs_setlease(struct file *filp, long arg, struct file_lock **lease, void **priv)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
Partially revert "locks: fix file locking on overlayfs" This partially reverts commit c568d68341be7030f5647def68851e469b21ca11. Overlayfs files will now automatically get the correct locks, no need to hack overlay support in VFS. It is a partial revert, because it leaves the locks_inode() calls in place and defines locks_inode() to file_inode(). We could revert those as well, but it would be unnecessary code churn and it makes sense to document that we are getting the inode for locking purposes. Don't revert MS_NOREMOTELOCK yet since that has been part of the userspace API for some time (though not in a useful way). Will try to remove internal flags later when the dust around the new mount API settles. Signed-off-by: Miklos Szeredi <mszeredi@redhat.com> Acked-by: Jeff Layton <jlayton@kernel.org>
2018-07-18 20:44:43 +07:00
if (filp->f_op->setlease)
locks: move i_lock acquisition into generic_*_lease handlers Now that we have a saner internal API for managing leases, we no longer need to mandate that the inode->i_lock be held over most of the lease code. Push it down into generic_add_lease and generic_delete_lease. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-08-23 05:50:48 +07:00
return filp->f_op->setlease(filp, arg, lease, priv);
locks: define a lm_setup handler for leases ...and move the fasync setup into it for fcntl lease calls. At the same time, change the semantics of how the file_lock double-pointer is handled. Up until now, on a successful lease return you got a pointer to the lock on the list. This is bad, since that pointer can no longer be relied on as valid once the inode->i_lock has been released. Change the code to instead just zero out the pointer if the lease we passed in ended up being used. Then the callers can just check to see if it's NULL after the call and free it if it isn't. The priv argument has the same semantics. The lm_setup function can zero the pointer out to signal to the caller that it should not be freed after the function returns. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-08-22 21:55:47 +07:00
else
locks: move i_lock acquisition into generic_*_lease handlers Now that we have a saner internal API for managing leases, we no longer need to mandate that the inode->i_lock be held over most of the lease code. Push it down into generic_add_lease and generic_delete_lease. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-08-23 05:50:48 +07:00
return generic_setlease(filp, arg, lease, priv);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
locks: rename lease functions to reflect locks.c conventions We've been using the convention that vfs_foo is the function that calls a filesystem-specific foo method if it exists, or falls back on a generic method if it doesn't; thus vfs_foo is what is called when some other part of the kernel (normally lockd or nfsd) wants to get a lock, whereas foo is what filesystems call to use the underlying local functionality as part of their lock implementation. So rename setlease to vfs_setlease (which will call a filesystem-specific setlease after a later patch) and __setlease to setlease. Also, vfs_setlease need only be GPL-exported as long as it's only needed by lockd and nfsd. Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu>
2007-06-08 04:09:49 +07:00
EXPORT_SYMBOL_GPL(vfs_setlease);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
locks: prevent ENOMEM on lease unlock Removing a lock shouldn't require any allocations; a failure due to ENOMEM leaves the caller with a choice between retrying or giving up and leaking an unused lease. Next we should split the other lease calls into add and delete cases. I wanted to start with just the bugfix. Signed-off-by: J. Bruce Fields <bfields@redhat.com> Acked-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-10-31 04:31:13 +07:00
static int do_fcntl_add_lease(unsigned int fd, struct file *filp, long arg)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
locks: define a lm_setup handler for leases ...and move the fasync setup into it for fcntl lease calls. At the same time, change the semantics of how the file_lock double-pointer is handled. Up until now, on a successful lease return you got a pointer to the lock on the list. This is bad, since that pointer can no longer be relied on as valid once the inode->i_lock has been released. Change the code to instead just zero out the pointer if the lease we passed in ended up being used. Then the callers can just check to see if it's NULL after the call and free it if it isn't. The priv argument has the same semantics. The lm_setup function can zero the pointer out to signal to the caller that it should not be freed after the function returns. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-08-22 21:55:47 +07:00
struct file_lock *fl;
fasync: re-organize fasync entry insertion to allow it under a spinlock You currently cannot use "fasync_helper()" in an atomic environment to insert a new fasync entry, because it will need to allocate the new "struct fasync_struct". Yet fcntl_setlease() wants to call this under lock_flocks(), which is in the process of being converted from the BKL to a spinlock. In order to fix this, this abstracts out the actual fasync list insertion and the fasync allocations into functions of their own, and teaches fs/locks.c to pre-allocate the fasync_struct entry. That way the actual list insertion can happen while holding the required spinlock. Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> [bfields@redhat.com: rebase on top of my changes to Arnd's patch] Tested-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
2010-10-27 23:38:12 +07:00
struct fasync_struct *new;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
int error;
locks/nfsd: allocate file lock outside of spinlock As suggested by Christoph Hellwig, this moves allocation of new file locks out of generic_setlease into the callers, nfs4_open_delegation and fcntl_setlease in order to allow GFP_KERNEL allocations when lock_flocks has become a spinlock. Signed-off-by: Arnd Bergmann <arnd@arndb.de> Acked-by: J. Bruce Fields <bfields@redhat.com>
2010-10-27 20:46:08 +07:00
fl = lease_alloc(filp, arg);
if (IS_ERR(fl))
return PTR_ERR(fl);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
fasync: re-organize fasync entry insertion to allow it under a spinlock You currently cannot use "fasync_helper()" in an atomic environment to insert a new fasync entry, because it will need to allocate the new "struct fasync_struct". Yet fcntl_setlease() wants to call this under lock_flocks(), which is in the process of being converted from the BKL to a spinlock. In order to fix this, this abstracts out the actual fasync list insertion and the fasync allocations into functions of their own, and teaches fs/locks.c to pre-allocate the fasync_struct entry. That way the actual list insertion can happen while holding the required spinlock. Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> [bfields@redhat.com: rebase on top of my changes to Arnd's patch] Tested-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
2010-10-27 23:38:12 +07:00
new = fasync_alloc();
if (!new) {
locks_free_lock(fl);
return -ENOMEM;
}
locks: define a lm_setup handler for leases ...and move the fasync setup into it for fcntl lease calls. At the same time, change the semantics of how the file_lock double-pointer is handled. Up until now, on a successful lease return you got a pointer to the lock on the list. This is bad, since that pointer can no longer be relied on as valid once the inode->i_lock has been released. Change the code to instead just zero out the pointer if the lease we passed in ended up being used. Then the callers can just check to see if it's NULL after the call and free it if it isn't. The priv argument has the same semantics. The lm_setup function can zero the pointer out to signal to the caller that it should not be freed after the function returns. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-08-22 21:55:47 +07:00
new->fa_fd = fd;
fasync: re-organize fasync entry insertion to allow it under a spinlock You currently cannot use "fasync_helper()" in an atomic environment to insert a new fasync entry, because it will need to allocate the new "struct fasync_struct". Yet fcntl_setlease() wants to call this under lock_flocks(), which is in the process of being converted from the BKL to a spinlock. In order to fix this, this abstracts out the actual fasync list insertion and the fasync allocations into functions of their own, and teaches fs/locks.c to pre-allocate the fasync_struct entry. That way the actual list insertion can happen while holding the required spinlock. Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> [bfields@redhat.com: rebase on top of my changes to Arnd's patch] Tested-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
2010-10-27 23:38:12 +07:00
locks: define a lm_setup handler for leases ...and move the fasync setup into it for fcntl lease calls. At the same time, change the semantics of how the file_lock double-pointer is handled. Up until now, on a successful lease return you got a pointer to the lock on the list. This is bad, since that pointer can no longer be relied on as valid once the inode->i_lock has been released. Change the code to instead just zero out the pointer if the lease we passed in ended up being used. Then the callers can just check to see if it's NULL after the call and free it if it isn't. The priv argument has the same semantics. The lm_setup function can zero the pointer out to signal to the caller that it should not be freed after the function returns. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-08-22 21:55:47 +07:00
error = vfs_setlease(filp, arg, &fl, (void **)&new);
locks: move locks_free_lock calls in do_fcntl_add_lease outside spinlock There's no need to call locks_free_lock here while still holding the i_lock. Defer that until the lock has been dropped. Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-08-12 05:14:12 +07:00
if (fl)
locks_free_lock(fl);
fasync: re-organize fasync entry insertion to allow it under a spinlock You currently cannot use "fasync_helper()" in an atomic environment to insert a new fasync entry, because it will need to allocate the new "struct fasync_struct". Yet fcntl_setlease() wants to call this under lock_flocks(), which is in the process of being converted from the BKL to a spinlock. In order to fix this, this abstracts out the actual fasync list insertion and the fasync allocations into functions of their own, and teaches fs/locks.c to pre-allocate the fasync_struct entry. That way the actual list insertion can happen while holding the required spinlock. Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> [bfields@redhat.com: rebase on top of my changes to Arnd's patch] Tested-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
2010-10-27 23:38:12 +07:00
if (new)
fasync_free(new);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
return error;
}
locks: prevent ENOMEM on lease unlock Removing a lock shouldn't require any allocations; a failure due to ENOMEM leaves the caller with a choice between retrying or giving up and leaking an unused lease. Next we should split the other lease calls into add and delete cases. I wanted to start with just the bugfix. Signed-off-by: J. Bruce Fields <bfields@redhat.com> Acked-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-10-31 04:31:13 +07:00
/**
* fcntl_setlease - sets a lease on an open file
* @fd: open file descriptor
* @filp: file pointer
* @arg: type of lease to obtain
*
* Call this fcntl to establish a lease on the file.
* Note that you also need to call %F_SETSIG to
* receive a signal when the lease is broken.
*/
int fcntl_setlease(unsigned int fd, struct file *filp, long arg)
{
if (arg == F_UNLCK)
fs: track fl_owner for leases Just like for other lock types we should allow different owners to have a read lease on a file. Currently this can't happen, but with the addition of pNFS layout leases we'll need this feature. Signed-off-by: Christoph Hellwig <hch@lst.de>
2015-01-22 01:14:02 +07:00
return vfs_setlease(filp, F_UNLCK, NULL, (void **)&filp);
locks: prevent ENOMEM on lease unlock Removing a lock shouldn't require any allocations; a failure due to ENOMEM leaves the caller with a choice between retrying or giving up and leaking an unused lease. Next we should split the other lease calls into add and delete cases. I wanted to start with just the bugfix. Signed-off-by: J. Bruce Fields <bfields@redhat.com> Acked-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-10-31 04:31:13 +07:00
return do_fcntl_add_lease(fd, filp, arg);
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
/**
locks: new helpers - flock_lock_inode_wait and posix_lock_inode_wait Allow callers to pass in an inode instead of a filp. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Reviewed-by: "J. Bruce Fields" <bfields@fieldses.org> Tested-by: "J. Bruce Fields" <bfields@fieldses.org>
2015-07-11 17:43:02 +07:00
* flock_lock_inode_wait - Apply a FLOCK-style lock to a file
* @inode: inode of the file to apply to
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
* @fl: The lock to be applied
*
locks: new helpers - flock_lock_inode_wait and posix_lock_inode_wait Allow callers to pass in an inode instead of a filp. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Reviewed-by: "J. Bruce Fields" <bfields@fieldses.org> Tested-by: "J. Bruce Fields" <bfields@fieldses.org>
2015-07-11 17:43:02 +07:00
* Apply a FLOCK style lock request to an inode.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
*/
locks: cleanup posix_lock_inode_wait and flock_lock_inode_wait All callers use locks_lock_inode_wait() instead. Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-10-23 00:38:15 +07:00
static int flock_lock_inode_wait(struct inode *inode, struct file_lock *fl)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
int error;
might_sleep();
for (;;) {
locks: new helpers - flock_lock_inode_wait and posix_lock_inode_wait Allow callers to pass in an inode instead of a filp. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Reviewed-by: "J. Bruce Fields" <bfields@fieldses.org> Tested-by: "J. Bruce Fields" <bfields@fieldses.org>
2015-07-11 17:43:02 +07:00
error = flock_lock_inode(inode, fl);
locks: add special return value for asynchronous locks Use a special error value FILE_LOCK_DEFERRED to mean that a locking operation returned asynchronously. This is returned by posix_lock_file() for sleeping locks to mean that the lock has been queued on the block list, and will be woken up when it might become available and needs to be retried (either fl_lmops->fl_notify() is called or fl_wait is woken up). f_op->lock() to mean either the above, or that the filesystem will call back with fl_lmops->fl_grant() when the result of the locking operation is known. The filesystem can do this for sleeping as well as non-sleeping locks. This is to make sure, that return values of -EAGAIN and -EINPROGRESS by filesystems are not mistaken to mean an asynchronous locking. This also makes error handling in fs/locks.c and lockd/svclock.c slightly cleaner. Signed-off-by: Miklos Szeredi <mszeredi@suse.cz> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Matthew Wilcox <matthew@wil.cx> Cc: David Teigland <teigland@redhat.com> Cc: Christoph Hellwig <hch@lst.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-07-25 15:48:57 +07:00
if (error != FILE_LOCK_DEFERRED)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
break;
fs/locks: rename some lists and pointers. struct file lock contains an 'fl_next' pointer which is used to point to the lock that this request is blocked waiting for. So rename it to fl_blocker. The fl_blocked list_head in an active lock is the head of a list of blocked requests. In a request it is a node in that list. These are two distinct uses, so replace with two list_heads with different names. fl_blocked_requests is the head of a list of blocked requests fl_blocked_member is a node in a member of that list. The two different list_heads are never used at the same time, but that will change in a future patch. Note that a tracepoint is changed to report fl_blocker instead of fl_next. Signed-off-by: NeilBrown <neilb@suse.com> Reviewed-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Jeff Layton <jlayton@kernel.org>
2018-11-30 06:04:08 +07:00
error = wait_event_interruptible(fl->fl_wait, !fl->fl_blocker);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
if (!error)
continue;
locks_delete_block(fl);
break;
}
return error;
}
locks: introduce locks_lock_inode_wait() Users of the locks API commonly call either posix_lock_file_wait() or flock_lock_file_wait() depending upon the lock type. Add a new function locks_lock_inode_wait() which will check and call the correct function for the type of lock passed in. Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-10-23 00:38:13 +07:00
/**
* locks_lock_inode_wait - Apply a lock to an inode
* @inode: inode of the file to apply to
* @fl: The lock to be applied
*
* Apply a POSIX or FLOCK style lock request to an inode.
*/
int locks_lock_inode_wait(struct inode *inode, struct file_lock *fl)
{
int res = 0;
switch (fl->fl_flags & (FL_POSIX|FL_FLOCK)) {
case FL_POSIX:
res = posix_lock_inode_wait(inode, fl);
break;
case FL_FLOCK:
res = flock_lock_inode_wait(inode, fl);
break;
default:
BUG();
}
return res;
}
EXPORT_SYMBOL(locks_lock_inode_wait);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
/**
* sys_flock: - flock() system call.
* @fd: the file descriptor to lock.
* @cmd: the type of lock to apply.
*
* Apply a %FL_FLOCK style lock to an open file descriptor.
fs: locks: Fix some troubles at kernel-doc comments There are a few syntax violations that cause outputs of a few comments to not be properly parsed in ReST format. No functional changes. Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-05-27 17:07:18 +07:00
* The @cmd can be one of:
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
*
fs: locks: Fix some troubles at kernel-doc comments There are a few syntax violations that cause outputs of a few comments to not be properly parsed in ReST format. No functional changes. Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-05-27 17:07:18 +07:00
* - %LOCK_SH -- a shared lock.
* - %LOCK_EX -- an exclusive lock.
* - %LOCK_UN -- remove an existing lock.
* - %LOCK_MAND -- a 'mandatory' flock.
* This exists to emulate Windows Share Modes.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
*
* %LOCK_MAND can be combined with %LOCK_READ or %LOCK_WRITE to allow other
* processes read and write access respectively.
*/
[CVE-2009-0029] System call wrappers part 16 Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
2009-01-14 20:14:18 +07:00
SYSCALL_DEFINE2(flock, unsigned int, fd, unsigned int, cmd)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
switch simple cases of fget_light to fdget Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2012-08-28 23:52:22 +07:00
struct fd f = fdget(fd);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
struct file_lock *lock;
int can_sleep, unlock;
int error;
error = -EBADF;
switch simple cases of fget_light to fdget Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2012-08-28 23:52:22 +07:00
if (!f.file)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
goto out;
can_sleep = !(cmd & LOCK_NB);
cmd &= ~LOCK_NB;
unlock = (cmd == LOCK_UN);
[PATCH] introduce fmode_t, do annotations Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2008-09-03 02:28:45 +07:00
if (!unlock && !(cmd & LOCK_MAND) &&
switch simple cases of fget_light to fdget Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2012-08-28 23:52:22 +07:00
!(f.file->f_mode & (FMODE_READ|FMODE_WRITE)))
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
goto out_putf;
locks: flock_make_lock should return a struct file_lock (or PTR_ERR) Eliminate the need for a return pointer. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-09-04 21:25:06 +07:00
lock = flock_make_lock(f.file, cmd);
if (IS_ERR(lock)) {
error = PTR_ERR(lock);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
goto out_putf;
locks: flock_make_lock should return a struct file_lock (or PTR_ERR) Eliminate the need for a return pointer. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2014-09-04 21:25:06 +07:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
if (can_sleep)
lock->fl_flags |= FL_SLEEP;
switch simple cases of fget_light to fdget Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2012-08-28 23:52:22 +07:00
error = security_file_lock(f.file, lock->fl_type);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
if (error)
goto out_free;
Partially revert "locks: fix file locking on overlayfs" This partially reverts commit c568d68341be7030f5647def68851e469b21ca11. Overlayfs files will now automatically get the correct locks, no need to hack overlay support in VFS. It is a partial revert, because it leaves the locks_inode() calls in place and defines locks_inode() to file_inode(). We could revert those as well, but it would be unnecessary code churn and it makes sense to document that we are getting the inode for locking purposes. Don't revert MS_NOREMOTELOCK yet since that has been part of the userspace API for some time (though not in a useful way). Will try to remove internal flags later when the dust around the new mount API settles. Signed-off-by: Miklos Szeredi <mszeredi@redhat.com> Acked-by: Jeff Layton <jlayton@kernel.org>
2018-07-18 20:44:43 +07:00
if (f.file->f_op->flock)
switch simple cases of fget_light to fdget Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2012-08-28 23:52:22 +07:00
error = f.file->f_op->flock(f.file,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
(can_sleep) ? F_SETLKW : F_SETLK,
lock);
else
Move locks API users to locks_lock_inode_wait() Instead of having users check for FL_POSIX or FL_FLOCK to call the correct locks API function, use the check within locks_lock_inode_wait(). This allows for some later cleanup. Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-10-23 00:38:14 +07:00
error = locks_lock_file_wait(f.file, lock);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
out_free:
[PATCH] fs/locks.c: Fix sys_flock() race sys_flock() currently has a race which can result in a double free in the multi-thread case. Thread 1 Thread 2 sys_flock(file, LOCK_EX) sys_flock(file, LOCK_UN) If Thread 2 removes the lock from inode->i_lock before Thread 1 tests for list_empty(&lock->fl_link) at the end of sys_flock, then both threads will end up calling locks_free_lock for the same lock. Fix is to make flock_lock_file() do the same as posix_lock_file(), namely to make a copy of the request, so that the caller can always free the lock. This also has the side-effect of fixing up a reference problem in the lockd handling of flock. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-03-31 17:30:55 +07:00
locks_free_lock(lock);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
out_putf:
switch simple cases of fget_light to fdget Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2012-08-28 23:52:22 +07:00
fdput(f);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
out:
return error;
}
locks: factor out generic/filesystem switch from test_lock Factor out the code that switches between generic and filesystem-specific lock methods; eventually we want to call this from lock managers (lockd and nfsd) too; currently they only call the generic methods. This patch does that for test_lock. Note that this hasn't been necessary until recently, because the few filesystems that define ->lock() (nfs, cifs...) aren't exportable via NFS. However GFS (and, in the future, other cluster filesystems) need to implement their own locking to get cluster-coherent locking, and also want to be able to export locking to NFS (lockd and NFSv4). So we accomplish this by factoring out code such as this and exporting it for the use of lockd and nfsd. Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu>
2007-02-21 12:58:50 +07:00
/**
* vfs_test_lock - test file byte range lock
* @filp: The file to test lock for
locks: fix vfs_test_lock() comment Thanks to Doug Chapman for pointing out that the comment here is inconsistent with the function prototype. Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu>
2007-05-12 03:22:50 +07:00
* @fl: The lock to test; also used to hold result
locks: factor out generic/filesystem switch from test_lock Factor out the code that switches between generic and filesystem-specific lock methods; eventually we want to call this from lock managers (lockd and nfsd) too; currently they only call the generic methods. This patch does that for test_lock. Note that this hasn't been necessary until recently, because the few filesystems that define ->lock() (nfs, cifs...) aren't exportable via NFS. However GFS (and, in the future, other cluster filesystems) need to implement their own locking to get cluster-coherent locking, and also want to be able to export locking to NFS (lockd and NFSv4). So we accomplish this by factoring out code such as this and exporting it for the use of lockd and nfsd. Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu>
2007-02-21 12:58:50 +07:00
*
* Returns -ERRNO on failure. Indicates presence of conflicting lock by
* setting conf->fl_type to something other than F_UNLCK.
*/
int vfs_test_lock(struct file *filp, struct file_lock *fl)
{
Partially revert "locks: fix file locking on overlayfs" This partially reverts commit c568d68341be7030f5647def68851e469b21ca11. Overlayfs files will now automatically get the correct locks, no need to hack overlay support in VFS. It is a partial revert, because it leaves the locks_inode() calls in place and defines locks_inode() to file_inode(). We could revert those as well, but it would be unnecessary code churn and it makes sense to document that we are getting the inode for locking purposes. Don't revert MS_NOREMOTELOCK yet since that has been part of the userspace API for some time (though not in a useful way). Will try to remove internal flags later when the dust around the new mount API settles. Signed-off-by: Miklos Szeredi <mszeredi@redhat.com> Acked-by: Jeff Layton <jlayton@kernel.org>
2018-07-18 20:44:43 +07:00
if (filp->f_op->lock)
locks: factor out generic/filesystem switch from test_lock Factor out the code that switches between generic and filesystem-specific lock methods; eventually we want to call this from lock managers (lockd and nfsd) too; currently they only call the generic methods. This patch does that for test_lock. Note that this hasn't been necessary until recently, because the few filesystems that define ->lock() (nfs, cifs...) aren't exportable via NFS. However GFS (and, in the future, other cluster filesystems) need to implement their own locking to get cluster-coherent locking, and also want to be able to export locking to NFS (lockd and NFSv4). So we accomplish this by factoring out code such as this and exporting it for the use of lockd and nfsd. Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu>
2007-02-21 12:58:50 +07:00
return filp->f_op->lock(filp, F_GETLK, fl);
posix_test_lock(filp, fl);
return 0;
}
EXPORT_SYMBOL_GPL(vfs_test_lock);
fs/locks: Remove fl_nspid and use fs-specific l_pid for remote locks Since commit c69899a17ca4 "NFSv4: Update of VFS byte range lock must be atomic with the stateid update", NFSv4 has been inserting locks in rpciod worker context. The result is that the file_lock's fl_nspid is the kworker's pid instead of the original userspace pid. The fl_nspid is only used to represent the namespaced virtual pid number when displaying locks or returning from F_GETLK. There's no reason to set it for every inserted lock, since we can usually just look it up from fl_pid. So, instead of looking up and holding struct pid for every lock, let's just look up the virtual pid number from fl_pid when it is needed. That means we can remove fl_nspid entirely. The translaton and presentation of fl_pid should handle the following four cases: 1 - F_GETLK on a remote file with a remote lock: In this case, the filesystem should determine the l_pid to return here. Filesystems should indicate that the fl_pid represents a non-local pid value that should not be translated by returning an fl_pid <= 0. 2 - F_GETLK on a local file with a remote lock: This should be the l_pid of the lock manager process, and translated. 3 - F_GETLK on a remote file with a local lock, and 4 - F_GETLK on a local file with a local lock: These should be the translated l_pid of the local locking process. Fuse was already doing the correct thing by translating the pid into the caller's namespace. With this change we must update fuse to translate to init's pid namespace, so that the locks API can then translate from init's pid namespace into the pid namespace of the caller. With this change, the locks API will expect that if a filesystem returns a remote pid as opposed to a local pid for F_GETLK, that remote pid will be <= 0. This signifies that the pid is remote, and the locks API will forego translating that pid into the pid namespace of the local calling process. Finally, we convert remote filesystems to present remote pids using negative numbers. Have lustre, 9p, ceph, cifs, and dlm negate the remote pid returned for F_GETLK lock requests. Since local pids will never be larger than PID_MAX_LIMIT (which is currently defined as <= 4 million), but pid_t is an unsigned int, we should have plenty of room to represent remote pids with negative numbers if we assume that remote pid numbers are similarly limited. If this is not the case, then we run the risk of having a remote pid returned for which there is also a corresponding local pid. This is a problem we have now, but this patch should reduce the chances of that occurring, while also returning those remote pid numbers, for whatever that may be worth. Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-07-16 21:28:22 +07:00
/**
* locks_translate_pid - translate a file_lock's fl_pid number into a namespace
* @fl: The file_lock who's fl_pid should be translated
* @ns: The namespace into which the pid should be translated
*
* Used to tranlate a fl_pid into a namespace virtual pid number
*/
static pid_t locks_translate_pid(struct file_lock *fl, struct pid_namespace *ns)
{
pid_t vnr;
struct pid *pid;
if (IS_OFDLCK(fl))
return -1;
if (IS_REMOTELCK(fl))
return fl->fl_pid;
fs/lock: skip lock owner pid translation in case we are in init_pid_ns If the flock owner process is dead and its pid has been already freed, pid translation won't work, but we still want to show flock owner pid number when expecting /proc/$PID/fdinfo/$FD in init pidns. Reproducer: process A process A1 process A2 fork()---------> exit() open() flock() fork()---------> exit() sleep() Before the patch: ================ (root@vz7)/: cat /proc/${PID_A2}/fdinfo/3 pos: 4 flags: 02100002 mnt_id: 257 lock: (root@vz7)/: After the patch: =============== (root@vz7)/:cat /proc/${PID_A2}/fdinfo/3 pos: 4 flags: 02100002 mnt_id: 295 lock: 1: FLOCK ADVISORY WRITE ${PID_A1} b6:f8a61:529946 0 EOF Fixes: 9d5b86ac13c5 ("fs/locks: Remove fl_nspid and use fs-specific l_pid for remote locks") Signed-off-by: Konstantin Khorenko <khorenko@virtuozzo.com> Acked-by: Andrey Vagin <avagin@openvz.org> Reviewed-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2018-06-08 21:27:11 +07:00
/*
* If the flock owner process is dead and its pid has been already
* freed, the translation below won't work, but we still want to show
* flock owner pid number in init pidns.
*/
if (ns == &init_pid_ns)
return (pid_t)fl->fl_pid;
fs/locks: Remove fl_nspid and use fs-specific l_pid for remote locks Since commit c69899a17ca4 "NFSv4: Update of VFS byte range lock must be atomic with the stateid update", NFSv4 has been inserting locks in rpciod worker context. The result is that the file_lock's fl_nspid is the kworker's pid instead of the original userspace pid. The fl_nspid is only used to represent the namespaced virtual pid number when displaying locks or returning from F_GETLK. There's no reason to set it for every inserted lock, since we can usually just look it up from fl_pid. So, instead of looking up and holding struct pid for every lock, let's just look up the virtual pid number from fl_pid when it is needed. That means we can remove fl_nspid entirely. The translaton and presentation of fl_pid should handle the following four cases: 1 - F_GETLK on a remote file with a remote lock: In this case, the filesystem should determine the l_pid to return here. Filesystems should indicate that the fl_pid represents a non-local pid value that should not be translated by returning an fl_pid <= 0. 2 - F_GETLK on a local file with a remote lock: This should be the l_pid of the lock manager process, and translated. 3 - F_GETLK on a remote file with a local lock, and 4 - F_GETLK on a local file with a local lock: These should be the translated l_pid of the local locking process. Fuse was already doing the correct thing by translating the pid into the caller's namespace. With this change we must update fuse to translate to init's pid namespace, so that the locks API can then translate from init's pid namespace into the pid namespace of the caller. With this change, the locks API will expect that if a filesystem returns a remote pid as opposed to a local pid for F_GETLK, that remote pid will be <= 0. This signifies that the pid is remote, and the locks API will forego translating that pid into the pid namespace of the local calling process. Finally, we convert remote filesystems to present remote pids using negative numbers. Have lustre, 9p, ceph, cifs, and dlm negate the remote pid returned for F_GETLK lock requests. Since local pids will never be larger than PID_MAX_LIMIT (which is currently defined as <= 4 million), but pid_t is an unsigned int, we should have plenty of room to represent remote pids with negative numbers if we assume that remote pid numbers are similarly limited. If this is not the case, then we run the risk of having a remote pid returned for which there is also a corresponding local pid. This is a problem we have now, but this patch should reduce the chances of that occurring, while also returning those remote pid numbers, for whatever that may be worth. Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-07-16 21:28:22 +07:00
rcu_read_lock();
pid = find_pid_ns(fl->fl_pid, &init_pid_ns);
vnr = pid_nr_ns(pid, ns);
rcu_read_unlock();
return vnr;
}
locks: create posix-to-flock helper functions Factor out a bit of messy code by creating posix-to-flock counterparts to the existing flock-to-posix helper functions. Cc: Christoph Hellwig <hch@infradead.org> Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu>
2007-02-21 04:10:11 +07:00
static int posix_lock_to_flock(struct flock *flock, struct file_lock *fl)
{
fs/locks: Remove fl_nspid and use fs-specific l_pid for remote locks Since commit c69899a17ca4 "NFSv4: Update of VFS byte range lock must be atomic with the stateid update", NFSv4 has been inserting locks in rpciod worker context. The result is that the file_lock's fl_nspid is the kworker's pid instead of the original userspace pid. The fl_nspid is only used to represent the namespaced virtual pid number when displaying locks or returning from F_GETLK. There's no reason to set it for every inserted lock, since we can usually just look it up from fl_pid. So, instead of looking up and holding struct pid for every lock, let's just look up the virtual pid number from fl_pid when it is needed. That means we can remove fl_nspid entirely. The translaton and presentation of fl_pid should handle the following four cases: 1 - F_GETLK on a remote file with a remote lock: In this case, the filesystem should determine the l_pid to return here. Filesystems should indicate that the fl_pid represents a non-local pid value that should not be translated by returning an fl_pid <= 0. 2 - F_GETLK on a local file with a remote lock: This should be the l_pid of the lock manager process, and translated. 3 - F_GETLK on a remote file with a local lock, and 4 - F_GETLK on a local file with a local lock: These should be the translated l_pid of the local locking process. Fuse was already doing the correct thing by translating the pid into the caller's namespace. With this change we must update fuse to translate to init's pid namespace, so that the locks API can then translate from init's pid namespace into the pid namespace of the caller. With this change, the locks API will expect that if a filesystem returns a remote pid as opposed to a local pid for F_GETLK, that remote pid will be <= 0. This signifies that the pid is remote, and the locks API will forego translating that pid into the pid namespace of the local calling process. Finally, we convert remote filesystems to present remote pids using negative numbers. Have lustre, 9p, ceph, cifs, and dlm negate the remote pid returned for F_GETLK lock requests. Since local pids will never be larger than PID_MAX_LIMIT (which is currently defined as <= 4 million), but pid_t is an unsigned int, we should have plenty of room to represent remote pids with negative numbers if we assume that remote pid numbers are similarly limited. If this is not the case, then we run the risk of having a remote pid returned for which there is also a corresponding local pid. This is a problem we have now, but this patch should reduce the chances of that occurring, while also returning those remote pid numbers, for whatever that may be worth. Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-07-16 21:28:22 +07:00
flock->l_pid = locks_translate_pid(fl, task_active_pid_ns(current));
locks: create posix-to-flock helper functions Factor out a bit of messy code by creating posix-to-flock counterparts to the existing flock-to-posix helper functions. Cc: Christoph Hellwig <hch@infradead.org> Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu>
2007-02-21 04:10:11 +07:00
#if BITS_PER_LONG == 32
/*
* Make sure we can represent the posix lock via
* legacy 32bit flock.
*/
if (fl->fl_start > OFFT_OFFSET_MAX)
return -EOVERFLOW;
if (fl->fl_end != OFFSET_MAX && fl->fl_end > OFFT_OFFSET_MAX)
return -EOVERFLOW;
#endif
flock->l_start = fl->fl_start;
flock->l_len = fl->fl_end == OFFSET_MAX ? 0 :
fl->fl_end - fl->fl_start + 1;
flock->l_whence = 0;
locks: fix F_GETLK regression (failure to find conflicts) In 9d6a8c5c213e34c475e72b245a8eb709258e968c we changed posix_test_lock to modify its single file_lock argument instead of taking separate input and output arguments. This makes it no longer safe to set the output lock's fl_type to F_UNLCK before looking for a conflict, since that means searching for a conflict against a lock with type F_UNLCK. This fixes a regression which causes F_GETLK to incorrectly report no conflict on most filesystems (including any filesystem that doesn't do its own locking). Also fix posix_lock_to_flock() to copy the lock type. This isn't strictly necessary, since the caller already does this; but it seems less likely to cause confusion in the future. Thanks to Doug Chapman for the bug report. Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu> Acked-by: Doug Chapman <doug.chapman@hp.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-05-11 05:38:43 +07:00
flock->l_type = fl->fl_type;
locks: create posix-to-flock helper functions Factor out a bit of messy code by creating posix-to-flock counterparts to the existing flock-to-posix helper functions. Cc: Christoph Hellwig <hch@infradead.org> Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu>
2007-02-21 04:10:11 +07:00
return 0;
}
#if BITS_PER_LONG == 32
static void posix_lock_to_flock64(struct flock64 *flock, struct file_lock *fl)
{
fs/locks: Remove fl_nspid and use fs-specific l_pid for remote locks Since commit c69899a17ca4 "NFSv4: Update of VFS byte range lock must be atomic with the stateid update", NFSv4 has been inserting locks in rpciod worker context. The result is that the file_lock's fl_nspid is the kworker's pid instead of the original userspace pid. The fl_nspid is only used to represent the namespaced virtual pid number when displaying locks or returning from F_GETLK. There's no reason to set it for every inserted lock, since we can usually just look it up from fl_pid. So, instead of looking up and holding struct pid for every lock, let's just look up the virtual pid number from fl_pid when it is needed. That means we can remove fl_nspid entirely. The translaton and presentation of fl_pid should handle the following four cases: 1 - F_GETLK on a remote file with a remote lock: In this case, the filesystem should determine the l_pid to return here. Filesystems should indicate that the fl_pid represents a non-local pid value that should not be translated by returning an fl_pid <= 0. 2 - F_GETLK on a local file with a remote lock: This should be the l_pid of the lock manager process, and translated. 3 - F_GETLK on a remote file with a local lock, and 4 - F_GETLK on a local file with a local lock: These should be the translated l_pid of the local locking process. Fuse was already doing the correct thing by translating the pid into the caller's namespace. With this change we must update fuse to translate to init's pid namespace, so that the locks API can then translate from init's pid namespace into the pid namespace of the caller. With this change, the locks API will expect that if a filesystem returns a remote pid as opposed to a local pid for F_GETLK, that remote pid will be <= 0. This signifies that the pid is remote, and the locks API will forego translating that pid into the pid namespace of the local calling process. Finally, we convert remote filesystems to present remote pids using negative numbers. Have lustre, 9p, ceph, cifs, and dlm negate the remote pid returned for F_GETLK lock requests. Since local pids will never be larger than PID_MAX_LIMIT (which is currently defined as <= 4 million), but pid_t is an unsigned int, we should have plenty of room to represent remote pids with negative numbers if we assume that remote pid numbers are similarly limited. If this is not the case, then we run the risk of having a remote pid returned for which there is also a corresponding local pid. This is a problem we have now, but this patch should reduce the chances of that occurring, while also returning those remote pid numbers, for whatever that may be worth. Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-07-16 21:28:22 +07:00
flock->l_pid = locks_translate_pid(fl, task_active_pid_ns(current));
locks: create posix-to-flock helper functions Factor out a bit of messy code by creating posix-to-flock counterparts to the existing flock-to-posix helper functions. Cc: Christoph Hellwig <hch@infradead.org> Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu>
2007-02-21 04:10:11 +07:00
flock->l_start = fl->fl_start;
flock->l_len = fl->fl_end == OFFSET_MAX ? 0 :
fl->fl_end - fl->fl_start + 1;
flock->l_whence = 0;
flock->l_type = fl->fl_type;
}
#endif
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
/* Report the first existing lock that would conflict with l.
* This implements the F_GETLK command of fcntl().
*/
fs/locks: pass kernel struct flock to fcntl_getlk/setlk This will make it easier to implement a sane compat fcntl syscall. [ jlayton: fix undeclared identifiers in 32-bit fcntl64 syscall handler ] Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-05-27 17:07:19 +07:00
int fcntl_getlk(struct file *filp, unsigned int cmd, struct flock *flock)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
fs/locks: Use allocation rather than the stack in fcntl_getlk() Struct file_lock is fairly large, so let's save some space on the stack by using an allocation for struct file_lock in fcntl_getlk(), just as we do for fcntl_setlk(). Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-07-16 21:28:21 +07:00
struct file_lock *fl;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
int error;
fs/locks: Use allocation rather than the stack in fcntl_getlk() Struct file_lock is fairly large, so let's save some space on the stack by using an allocation for struct file_lock in fcntl_getlk(), just as we do for fcntl_setlk(). Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-07-16 21:28:21 +07:00
fl = locks_alloc_lock();
if (fl == NULL)
return -ENOMEM;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
error = -EINVAL;
fs/locks: pass kernel struct flock to fcntl_getlk/setlk This will make it easier to implement a sane compat fcntl syscall. [ jlayton: fix undeclared identifiers in 32-bit fcntl64 syscall handler ] Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-05-27 17:07:19 +07:00
if (flock->l_type != F_RDLCK && flock->l_type != F_WRLCK)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
goto out;
fs/locks: Use allocation rather than the stack in fcntl_getlk() Struct file_lock is fairly large, so let's save some space on the stack by using an allocation for struct file_lock in fcntl_getlk(), just as we do for fcntl_setlk(). Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-07-16 21:28:21 +07:00
error = flock_to_posix_lock(filp, fl, flock);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
if (error)
goto out;
locks: rename file-private locks to "open file description locks" File-private locks have been merged into Linux for v3.15, and *now* people are commenting that the name and macro definitions for the new file-private locks suck. ...and I can't even disagree. The names and command macros do suck. We're going to have to live with these for a long time, so it's important that we be happy with the names before we're stuck with them. The consensus on the lists so far is that they should be rechristened as "open file description locks". The name isn't a big deal for the kernel, but the command macros are not visually distinct enough from the traditional POSIX lock macros. The glibc and documentation folks are recommending that we change them to look like F_OFD_{GETLK|SETLK|SETLKW}. That lessens the chance that a programmer will typo one of the commands wrong, and also makes it easier to spot this difference when reading code. This patch makes the following changes that I think are necessary before v3.15 ships: 1) rename the command macros to their new names. These end up in the uapi headers and so are part of the external-facing API. It turns out that glibc doesn't actually use the fcntl.h uapi header, but it's hard to be sure that something else won't. Changing it now is safest. 2) make the the /proc/locks output display these as type "OFDLCK" Cc: Michael Kerrisk <mtk.manpages@gmail.com> Cc: Christoph Hellwig <hch@infradead.org> Cc: Carlos O'Donell <carlos@redhat.com> Cc: Stefan Metzmacher <metze@samba.org> Cc: Andy Lutomirski <luto@amacapital.net> Cc: Frank Filz <ffilzlnx@mindspring.com> Cc: Theodore Ts'o <tytso@mit.edu> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-04-22 19:23:58 +07:00
if (cmd == F_OFD_GETLK) {
locks: require that flock->l_pid be set to 0 for file-private locks Neil Brown suggested potentially overloading the l_pid value as a "lock context" field for file-private locks. While I don't think we will probably want to do that here, it's probably a good idea to ensure that in the future we could extend this API without breaking existing callers. Typically the l_pid value is ignored for incoming struct flock arguments, serving mainly as a place to return the pid of the owner if there is a conflicting lock. For file-private locks, require that it currently be set to 0 and return EINVAL if it isn't. If we eventually want to make a non-zero l_pid mean something, then this will help ensure that we don't break legacy programs that are using file-private locks. Cc: Neil Brown <neilb@suse.de> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-03-04 22:30:23 +07:00
error = -EINVAL;
fs/locks: pass kernel struct flock to fcntl_getlk/setlk This will make it easier to implement a sane compat fcntl syscall. [ jlayton: fix undeclared identifiers in 32-bit fcntl64 syscall handler ] Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-05-27 17:07:19 +07:00
if (flock->l_pid != 0)
locks: require that flock->l_pid be set to 0 for file-private locks Neil Brown suggested potentially overloading the l_pid value as a "lock context" field for file-private locks. While I don't think we will probably want to do that here, it's probably a good idea to ensure that in the future we could extend this API without breaking existing callers. Typically the l_pid value is ignored for incoming struct flock arguments, serving mainly as a place to return the pid of the owner if there is a conflicting lock. For file-private locks, require that it currently be set to 0 and return EINVAL if it isn't. If we eventually want to make a non-zero l_pid mean something, then this will help ensure that we don't break legacy programs that are using file-private locks. Cc: Neil Brown <neilb@suse.de> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-03-04 22:30:23 +07:00
goto out;
locks: add new fcntl cmd values for handling file private locks Due to some unfortunate history, POSIX locks have very strange and unhelpful semantics. The thing that usually catches people by surprise is that they are dropped whenever the process closes any file descriptor associated with the inode. This is extremely problematic for people developing file servers that need to implement byte-range locks. Developers often need a "lock management" facility to ensure that file descriptors are not closed until all of the locks associated with the inode are finished. Additionally, "classic" POSIX locks are owned by the process. Locks taken between threads within the same process won't conflict with one another, which renders them useless for synchronization between threads. This patchset adds a new type of lock that attempts to address these issues. These locks conflict with classic POSIX read/write locks, but have semantics that are more like BSD locks with respect to inheritance and behavior on close. This is implemented primarily by changing how fl_owner field is set for these locks. Instead of having them owned by the files_struct of the process, they are instead owned by the filp on which they were acquired. Thus, they are inherited across fork() and are only released when the last reference to a filp is put. These new semantics prevent them from being merged with classic POSIX locks, even if they are acquired by the same process. These locks will also conflict with classic POSIX locks even if they are acquired by the same process or on the same file descriptor. The new locks are managed using a new set of cmd values to the fcntl() syscall. The initial implementation of this converts these values to "classic" cmd values at a fairly high level, and the details are not exposed to the underlying filesystem. We may eventually want to push this handing out to the lower filesystem code but for now I don't see any need for it. Also, note that with this implementation the new cmd values are only available via fcntl64() on 32-bit arches. There's little need to add support for legacy apps on a new interface like this. Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:10 +07:00
cmd = F_GETLK;
fs/locks: Use allocation rather than the stack in fcntl_getlk() Struct file_lock is fairly large, so let's save some space on the stack by using an allocation for struct file_lock in fcntl_getlk(), just as we do for fcntl_setlk(). Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-07-16 21:28:21 +07:00
fl->fl_flags |= FL_OFDLCK;
fl->fl_owner = filp;
locks: add new fcntl cmd values for handling file private locks Due to some unfortunate history, POSIX locks have very strange and unhelpful semantics. The thing that usually catches people by surprise is that they are dropped whenever the process closes any file descriptor associated with the inode. This is extremely problematic for people developing file servers that need to implement byte-range locks. Developers often need a "lock management" facility to ensure that file descriptors are not closed until all of the locks associated with the inode are finished. Additionally, "classic" POSIX locks are owned by the process. Locks taken between threads within the same process won't conflict with one another, which renders them useless for synchronization between threads. This patchset adds a new type of lock that attempts to address these issues. These locks conflict with classic POSIX read/write locks, but have semantics that are more like BSD locks with respect to inheritance and behavior on close. This is implemented primarily by changing how fl_owner field is set for these locks. Instead of having them owned by the files_struct of the process, they are instead owned by the filp on which they were acquired. Thus, they are inherited across fork() and are only released when the last reference to a filp is put. These new semantics prevent them from being merged with classic POSIX locks, even if they are acquired by the same process. These locks will also conflict with classic POSIX locks even if they are acquired by the same process or on the same file descriptor. The new locks are managed using a new set of cmd values to the fcntl() syscall. The initial implementation of this converts these values to "classic" cmd values at a fairly high level, and the details are not exposed to the underlying filesystem. We may eventually want to push this handing out to the lower filesystem code but for now I don't see any need for it. Also, note that with this implementation the new cmd values are only available via fcntl64() on 32-bit arches. There's little need to add support for legacy apps on a new interface like this. Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:10 +07:00
}
fs/locks: Use allocation rather than the stack in fcntl_getlk() Struct file_lock is fairly large, so let's save some space on the stack by using an allocation for struct file_lock in fcntl_getlk(), just as we do for fcntl_setlk(). Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-07-16 21:28:21 +07:00
error = vfs_test_lock(filp, fl);
locks: factor out generic/filesystem switch from test_lock Factor out the code that switches between generic and filesystem-specific lock methods; eventually we want to call this from lock managers (lockd and nfsd) too; currently they only call the generic methods. This patch does that for test_lock. Note that this hasn't been necessary until recently, because the few filesystems that define ->lock() (nfs, cifs...) aren't exportable via NFS. However GFS (and, in the future, other cluster filesystems) need to implement their own locking to get cluster-coherent locking, and also want to be able to export locking to NFS (lockd and NFSv4). So we accomplish this by factoring out code such as this and exporting it for the use of lockd and nfsd. Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu>
2007-02-21 12:58:50 +07:00
if (error)
goto out;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
fs/locks: Use allocation rather than the stack in fcntl_getlk() Struct file_lock is fairly large, so let's save some space on the stack by using an allocation for struct file_lock in fcntl_getlk(), just as we do for fcntl_setlk(). Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-07-16 21:28:21 +07:00
flock->l_type = fl->fl_type;
if (fl->fl_type != F_UNLCK) {
error = posix_lock_to_flock(flock, fl);
locks: create posix-to-flock helper functions Factor out a bit of messy code by creating posix-to-flock counterparts to the existing flock-to-posix helper functions. Cc: Christoph Hellwig <hch@infradead.org> Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu>
2007-02-21 04:10:11 +07:00
if (error)
fs/locks: Use allocation rather than the stack in fcntl_getlk() Struct file_lock is fairly large, so let's save some space on the stack by using an allocation for struct file_lock in fcntl_getlk(), just as we do for fcntl_setlk(). Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-07-16 21:28:21 +07:00
goto out;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
out:
fs/locks: Use allocation rather than the stack in fcntl_getlk() Struct file_lock is fairly large, so let's save some space on the stack by using an allocation for struct file_lock in fcntl_getlk(), just as we do for fcntl_setlk(). Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-07-16 21:28:21 +07:00
locks_free_lock(fl);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
return error;
}
locks: factor out generic/filesystem switch from setlock code Factor out the code that switches between generic and filesystem-specific lock methods; eventually we want to call this from lock managers (lockd and nfsd) too; currently they only call the generic methods. This patch does that for all the setlk code. Signed-off-by: Marc Eshel <eshel@almaden.ibm.com> Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu>
2007-01-19 03:08:55 +07:00
/**
* vfs_lock_file - file byte range lock
* @filp: The file to apply the lock to
* @cmd: type of locking operation (F_SETLK, F_GETLK, etc.)
* @fl: The lock to be applied
locks: allow {vfs,posix}_lock_file to return conflicting lock The nfsv4 protocol's lock operation, in the case of a conflict, returns information about the conflicting lock. It's unclear how clients can use this, so for now we're not going so far as to add a filesystem method that can return a conflicting lock, but we may as well return something in the local case when it's easy to. Signed-off-by: Marc Eshel <eshel@almaden.ibm.com> Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu>
2007-01-19 04:15:35 +07:00
* @conf: Place to return a copy of the conflicting lock, if found.
*
* A caller that doesn't care about the conflicting lock may pass NULL
* as the final argument.
*
* If the filesystem defines a private ->lock() method, then @conf will
* be left unchanged; so a caller that cares should initialize it to
* some acceptable default.
locks: add fl_grant callback for asynchronous lock return Acquiring a lock on a cluster filesystem may require communication with remote hosts, and to avoid blocking lockd or nfsd threads during such communication, we allow the results to be returned asynchronously. When a ->lock() call needs to block, the file system will return -EINPROGRESS, and then later return the results with a call to the routine in the fl_grant field of the lock_manager_operations struct. This differs from the case when ->lock returns -EAGAIN to a blocking lock request; in that case, the filesystem calls fl_notify when the lock is granted, and the caller retries the original lock. So while fl_notify is merely a hint to the caller that it should retry, fl_grant actually communicates the final result of the lock operation (with the lock already acquired in the succesful case). Therefore fl_grant takes a lock, a status and, for the test lock case, a conflicting lock. We also allow fl_grant to return an error to the filesystem, to handle the case where the fl_grant requests arrives after the lock manager has already given up waiting for it. Signed-off-by: Marc Eshel <eshel@almaden.ibm.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2006-12-06 11:31:28 +07:00
*
* To avoid blocking kernel daemons, such as lockd, that need to acquire POSIX
* locks, the ->lock() interface may return asynchronously, before the lock has
* been granted or denied by the underlying filesystem, if (and only if)
locks: rename lock-manager ops 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: J. Bruce Fields <bfields@redhat.com>
2011-07-21 07:21:59 +07:00
* lm_grant is set. Callers expecting ->lock() to return asynchronously
locks: add fl_grant callback for asynchronous lock return Acquiring a lock on a cluster filesystem may require communication with remote hosts, and to avoid blocking lockd or nfsd threads during such communication, we allow the results to be returned asynchronously. When a ->lock() call needs to block, the file system will return -EINPROGRESS, and then later return the results with a call to the routine in the fl_grant field of the lock_manager_operations struct. This differs from the case when ->lock returns -EAGAIN to a blocking lock request; in that case, the filesystem calls fl_notify when the lock is granted, and the caller retries the original lock. So while fl_notify is merely a hint to the caller that it should retry, fl_grant actually communicates the final result of the lock operation (with the lock already acquired in the succesful case). Therefore fl_grant takes a lock, a status and, for the test lock case, a conflicting lock. We also allow fl_grant to return an error to the filesystem, to handle the case where the fl_grant requests arrives after the lock manager has already given up waiting for it. Signed-off-by: Marc Eshel <eshel@almaden.ibm.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2006-12-06 11:31:28 +07:00
* will only use F_SETLK, not F_SETLKW; they will set FL_SLEEP if (and only if)
* the request is for a blocking lock. When ->lock() does return asynchronously,
locks: rename lock-manager ops 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: J. Bruce Fields <bfields@redhat.com>
2011-07-21 07:21:59 +07:00
* it must return FILE_LOCK_DEFERRED, and call ->lm_grant() when the lock
locks: add fl_grant callback for asynchronous lock return Acquiring a lock on a cluster filesystem may require communication with remote hosts, and to avoid blocking lockd or nfsd threads during such communication, we allow the results to be returned asynchronously. When a ->lock() call needs to block, the file system will return -EINPROGRESS, and then later return the results with a call to the routine in the fl_grant field of the lock_manager_operations struct. This differs from the case when ->lock returns -EAGAIN to a blocking lock request; in that case, the filesystem calls fl_notify when the lock is granted, and the caller retries the original lock. So while fl_notify is merely a hint to the caller that it should retry, fl_grant actually communicates the final result of the lock operation (with the lock already acquired in the succesful case). Therefore fl_grant takes a lock, a status and, for the test lock case, a conflicting lock. We also allow fl_grant to return an error to the filesystem, to handle the case where the fl_grant requests arrives after the lock manager has already given up waiting for it. Signed-off-by: Marc Eshel <eshel@almaden.ibm.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2006-12-06 11:31:28 +07:00
* request completes.
* If the request is for non-blocking lock the file system should return
locks: add special return value for asynchronous locks Use a special error value FILE_LOCK_DEFERRED to mean that a locking operation returned asynchronously. This is returned by posix_lock_file() for sleeping locks to mean that the lock has been queued on the block list, and will be woken up when it might become available and needs to be retried (either fl_lmops->fl_notify() is called or fl_wait is woken up). f_op->lock() to mean either the above, or that the filesystem will call back with fl_lmops->fl_grant() when the result of the locking operation is known. The filesystem can do this for sleeping as well as non-sleeping locks. This is to make sure, that return values of -EAGAIN and -EINPROGRESS by filesystems are not mistaken to mean an asynchronous locking. This also makes error handling in fs/locks.c and lockd/svclock.c slightly cleaner. Signed-off-by: Miklos Szeredi <mszeredi@suse.cz> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Matthew Wilcox <matthew@wil.cx> Cc: David Teigland <teigland@redhat.com> Cc: Christoph Hellwig <hch@lst.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-07-25 15:48:57 +07:00
* FILE_LOCK_DEFERRED then try to get the lock and call the callback routine
* with the result. If the request timed out the callback routine will return a
locks: add fl_grant callback for asynchronous lock return Acquiring a lock on a cluster filesystem may require communication with remote hosts, and to avoid blocking lockd or nfsd threads during such communication, we allow the results to be returned asynchronously. When a ->lock() call needs to block, the file system will return -EINPROGRESS, and then later return the results with a call to the routine in the fl_grant field of the lock_manager_operations struct. This differs from the case when ->lock returns -EAGAIN to a blocking lock request; in that case, the filesystem calls fl_notify when the lock is granted, and the caller retries the original lock. So while fl_notify is merely a hint to the caller that it should retry, fl_grant actually communicates the final result of the lock operation (with the lock already acquired in the succesful case). Therefore fl_grant takes a lock, a status and, for the test lock case, a conflicting lock. We also allow fl_grant to return an error to the filesystem, to handle the case where the fl_grant requests arrives after the lock manager has already given up waiting for it. Signed-off-by: Marc Eshel <eshel@almaden.ibm.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2006-12-06 11:31:28 +07:00
* nonzero return code and the file system should release the lock. The file
* system is also responsible to keep a corresponding posix lock when it
* grants a lock so the VFS can find out which locks are locally held and do
* the correct lock cleanup when required.
* The underlying filesystem must not drop the kernel lock or call
locks: rename lock-manager ops 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: J. Bruce Fields <bfields@redhat.com>
2011-07-21 07:21:59 +07:00
* ->lm_grant() before returning to the caller with a FILE_LOCK_DEFERRED
locks: add fl_grant callback for asynchronous lock return Acquiring a lock on a cluster filesystem may require communication with remote hosts, and to avoid blocking lockd or nfsd threads during such communication, we allow the results to be returned asynchronously. When a ->lock() call needs to block, the file system will return -EINPROGRESS, and then later return the results with a call to the routine in the fl_grant field of the lock_manager_operations struct. This differs from the case when ->lock returns -EAGAIN to a blocking lock request; in that case, the filesystem calls fl_notify when the lock is granted, and the caller retries the original lock. So while fl_notify is merely a hint to the caller that it should retry, fl_grant actually communicates the final result of the lock operation (with the lock already acquired in the succesful case). Therefore fl_grant takes a lock, a status and, for the test lock case, a conflicting lock. We also allow fl_grant to return an error to the filesystem, to handle the case where the fl_grant requests arrives after the lock manager has already given up waiting for it. Signed-off-by: Marc Eshel <eshel@almaden.ibm.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2006-12-06 11:31:28 +07:00
* return code.
locks: factor out generic/filesystem switch from setlock code Factor out the code that switches between generic and filesystem-specific lock methods; eventually we want to call this from lock managers (lockd and nfsd) too; currently they only call the generic methods. This patch does that for all the setlk code. Signed-off-by: Marc Eshel <eshel@almaden.ibm.com> Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu>
2007-01-19 03:08:55 +07:00
*/
locks: allow {vfs,posix}_lock_file to return conflicting lock The nfsv4 protocol's lock operation, in the case of a conflict, returns information about the conflicting lock. It's unclear how clients can use this, so for now we're not going so far as to add a filesystem method that can return a conflicting lock, but we may as well return something in the local case when it's easy to. Signed-off-by: Marc Eshel <eshel@almaden.ibm.com> Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu>
2007-01-19 04:15:35 +07:00
int vfs_lock_file(struct file *filp, unsigned int cmd, struct file_lock *fl, struct file_lock *conf)
locks: factor out generic/filesystem switch from setlock code Factor out the code that switches between generic and filesystem-specific lock methods; eventually we want to call this from lock managers (lockd and nfsd) too; currently they only call the generic methods. This patch does that for all the setlk code. Signed-off-by: Marc Eshel <eshel@almaden.ibm.com> Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu>
2007-01-19 03:08:55 +07:00
{
Partially revert "locks: fix file locking on overlayfs" This partially reverts commit c568d68341be7030f5647def68851e469b21ca11. Overlayfs files will now automatically get the correct locks, no need to hack overlay support in VFS. It is a partial revert, because it leaves the locks_inode() calls in place and defines locks_inode() to file_inode(). We could revert those as well, but it would be unnecessary code churn and it makes sense to document that we are getting the inode for locking purposes. Don't revert MS_NOREMOTELOCK yet since that has been part of the userspace API for some time (though not in a useful way). Will try to remove internal flags later when the dust around the new mount API settles. Signed-off-by: Miklos Szeredi <mszeredi@redhat.com> Acked-by: Jeff Layton <jlayton@kernel.org>
2018-07-18 20:44:43 +07:00
if (filp->f_op->lock)
locks: factor out generic/filesystem switch from setlock code Factor out the code that switches between generic and filesystem-specific lock methods; eventually we want to call this from lock managers (lockd and nfsd) too; currently they only call the generic methods. This patch does that for all the setlk code. Signed-off-by: Marc Eshel <eshel@almaden.ibm.com> Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu>
2007-01-19 03:08:55 +07:00
return filp->f_op->lock(filp, cmd, fl);
else
locks: allow {vfs,posix}_lock_file to return conflicting lock The nfsv4 protocol's lock operation, in the case of a conflict, returns information about the conflicting lock. It's unclear how clients can use this, so for now we're not going so far as to add a filesystem method that can return a conflicting lock, but we may as well return something in the local case when it's easy to. Signed-off-by: Marc Eshel <eshel@almaden.ibm.com> Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu>
2007-01-19 04:15:35 +07:00
return posix_lock_file(filp, fl, conf);
locks: factor out generic/filesystem switch from setlock code Factor out the code that switches between generic and filesystem-specific lock methods; eventually we want to call this from lock managers (lockd and nfsd) too; currently they only call the generic methods. This patch does that for all the setlk code. Signed-off-by: Marc Eshel <eshel@almaden.ibm.com> Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu>
2007-01-19 03:08:55 +07:00
}
EXPORT_SYMBOL_GPL(vfs_lock_file);
locks: cleanup code duplication Extract common code into a function. Signed-off-by: Miklos Szeredi <mszeredi@suse.cz> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Cc: Matthew Wilcox <matthew@wil.cx> Cc: David Teigland <teigland@redhat.com> Cc: Christoph Hellwig <hch@lst.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-07-25 15:48:57 +07:00
static int do_lock_file_wait(struct file *filp, unsigned int cmd,
struct file_lock *fl)
{
int error;
error = security_file_lock(filp, fl->fl_type);
if (error)
return error;
locks: allow ->lock() to return FILE_LOCK_DEFERRED Allow filesystem's ->lock() method to call posix_lock_file() instead of posix_lock_file_wait(), and return FILE_LOCK_DEFERRED. This makes it possible to implement a such a ->lock() function, that works with the lock manager, which needs the call to be asynchronous. Now the vfs_lock_file() helper can be used, so this is a cleanup as well. Signed-off-by: Miklos Szeredi <mszeredi@suse.cz> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Cc: Matthew Wilcox <matthew@wil.cx> Cc: David Teigland <teigland@redhat.com> Cc: Christoph Hellwig <hch@lst.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-07-25 15:48:58 +07:00
for (;;) {
error = vfs_lock_file(filp, cmd, fl, NULL);
if (error != FILE_LOCK_DEFERRED)
locks: cleanup code duplication Extract common code into a function. Signed-off-by: Miklos Szeredi <mszeredi@suse.cz> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Cc: Matthew Wilcox <matthew@wil.cx> Cc: David Teigland <teigland@redhat.com> Cc: Christoph Hellwig <hch@lst.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-07-25 15:48:57 +07:00
break;
fs/locks: rename some lists and pointers. struct file lock contains an 'fl_next' pointer which is used to point to the lock that this request is blocked waiting for. So rename it to fl_blocker. The fl_blocked list_head in an active lock is the head of a list of blocked requests. In a request it is a node in that list. These are two distinct uses, so replace with two list_heads with different names. fl_blocked_requests is the head of a list of blocked requests fl_blocked_member is a node in a member of that list. The two different list_heads are never used at the same time, but that will change in a future patch. Note that a tracepoint is changed to report fl_blocker instead of fl_next. Signed-off-by: NeilBrown <neilb@suse.com> Reviewed-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Jeff Layton <jlayton@kernel.org>
2018-11-30 06:04:08 +07:00
error = wait_event_interruptible(fl->fl_wait, !fl->fl_blocker);
locks: allow ->lock() to return FILE_LOCK_DEFERRED Allow filesystem's ->lock() method to call posix_lock_file() instead of posix_lock_file_wait(), and return FILE_LOCK_DEFERRED. This makes it possible to implement a such a ->lock() function, that works with the lock manager, which needs the call to be asynchronous. Now the vfs_lock_file() helper can be used, so this is a cleanup as well. Signed-off-by: Miklos Szeredi <mszeredi@suse.cz> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Cc: Matthew Wilcox <matthew@wil.cx> Cc: David Teigland <teigland@redhat.com> Cc: Christoph Hellwig <hch@lst.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-07-25 15:48:58 +07:00
if (!error)
continue;
locks_delete_block(fl);
break;
locks: cleanup code duplication Extract common code into a function. Signed-off-by: Miklos Szeredi <mszeredi@suse.cz> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Cc: Matthew Wilcox <matthew@wil.cx> Cc: David Teigland <teigland@redhat.com> Cc: Christoph Hellwig <hch@lst.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-07-25 15:48:57 +07:00
}
return error;
}
locks: Use more file_inode and fix a comment Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-10-15 20:07:07 +07:00
/* Ensure that fl->fl_file has compatible f_mode for F_SETLK calls */
locks: only validate the lock vs. f_mode in F_SETLK codepaths v2: replace missing break in switch statement (as pointed out by Dave Jones) commit bce7560d4946 (locks: consolidate checks for compatible filp->f_mode values in setlk handlers) introduced a regression in the F_GETLK handler. flock64_to_posix_lock is a shared codepath between F_GETLK and F_SETLK, but the f_mode checks should only be applicable to the F_SETLK codepaths according to POSIX. Instead of just reverting the patch, add a new function to do this checking and have the F_SETLK handlers call it. Cc: Dave Jones <davej@redhat.com> Reported-and-Tested-by: Reuben Farrelly <reuben@reub.net> Signed-off-by: Jeff Layton <jlayton@poochiereds.net>
2014-05-09 22:41:54 +07:00
static int
check_fmode_for_setlk(struct file_lock *fl)
{
switch (fl->fl_type) {
case F_RDLCK:
if (!(fl->fl_file->f_mode & FMODE_READ))
return -EBADF;
break;
case F_WRLCK:
if (!(fl->fl_file->f_mode & FMODE_WRITE))
return -EBADF;
}
return 0;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
/* Apply the lock described by l to an open file descriptor.
* This implements both the F_SETLK and F_SETLKW commands of fcntl().
*/
[PATCH] stale POSIX lock handling 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>
2005-07-28 01:45:09 +07:00
int fcntl_setlk(unsigned int fd, struct file *filp, unsigned int cmd,
fs/locks: pass kernel struct flock to fcntl_getlk/setlk This will make it easier to implement a sane compat fcntl syscall. [ jlayton: fix undeclared identifiers in 32-bit fcntl64 syscall handler ] Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-05-27 17:07:19 +07:00
struct flock *flock)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
struct file_lock *file_lock = locks_alloc_lock();
fs/locks: pass kernel struct flock to fcntl_getlk/setlk This will make it easier to implement a sane compat fcntl syscall. [ jlayton: fix undeclared identifiers in 32-bit fcntl64 syscall handler ] Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-05-27 17:07:19 +07:00
struct inode *inode = locks_inode(filp);
[PATCH] fix SMP ordering hole in fcntl_setlk() fcntl_setlk()/close() race prevention has a subtle hole - we need to make sure that if we *do* have an fcntl/close race on SMP box, the access to descriptor table and inode->i_flock won't get reordered. As it is, we get STORE inode->i_flock, LOAD descriptor table entry vs. STORE descriptor table entry, LOAD inode->i_flock with not a single lock in common on both sides. We do have BKL around the first STORE, but check in locks_remove_posix() is outside of BKL and for a good reason - we don't want BKL on common path of close(2). Solution is to hold ->file_lock around fcheck() in there; that orders us wrt removal from descriptor table that preceded locks_remove_posix() on close path and we either come first (in which case eviction will be handled by the close side) or we'll see the effect of close and do eviction ourselves. Note that even though it's read-only access, we do need ->file_lock here - rcu_read_lock() won't be enough to order the things. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2008-05-07 00:58:34 +07:00
struct file *f;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
int error;
if (file_lock == NULL)
return -ENOLCK;
/* Don't allow mandatory locks on files that may be memory mapped
* and shared.
*/
Cleanup macros for distinguishing mandatory locks The combination of S_ISGID bit set and S_IXGRP bit unset is used to mark the inode as "mandatory lockable" and there's a macro for this check called MANDATORY_LOCK(inode). However, fs/locks.c and some filesystems still perform the explicit i_mode checking. Besides, Andrew pointed out, that this macro is buggy itself, as it dereferences the inode arg twice. Convert this macro into static inline function and switch its users to it, making the code shorter and more readable. The __mandatory_lock() helper is to be used in places where the IS_MANDLOCK() for superblock is already known to be true. Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: David Howells <dhowells@redhat.com> Cc: Eric Van Hensbergen <ericvh@gmail.com> Cc: Ron Minnich <rminnich@sandia.gov> Cc: Latchesar Ionkov <lucho@ionkov.net> Cc: Steven Whitehouse <swhiteho@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2007-10-02 04:41:11 +07:00
if (mandatory_lock(inode) && mapping_writably_mapped(filp->f_mapping)) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
error = -EAGAIN;
goto out;
}
fs/locks: pass kernel struct flock to fcntl_getlk/setlk This will make it easier to implement a sane compat fcntl syscall. [ jlayton: fix undeclared identifiers in 32-bit fcntl64 syscall handler ] Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-05-27 17:07:19 +07:00
error = flock_to_posix_lock(filp, file_lock, flock);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
if (error)
goto out;
locks: add new fcntl cmd values for handling file private locks Due to some unfortunate history, POSIX locks have very strange and unhelpful semantics. The thing that usually catches people by surprise is that they are dropped whenever the process closes any file descriptor associated with the inode. This is extremely problematic for people developing file servers that need to implement byte-range locks. Developers often need a "lock management" facility to ensure that file descriptors are not closed until all of the locks associated with the inode are finished. Additionally, "classic" POSIX locks are owned by the process. Locks taken between threads within the same process won't conflict with one another, which renders them useless for synchronization between threads. This patchset adds a new type of lock that attempts to address these issues. These locks conflict with classic POSIX read/write locks, but have semantics that are more like BSD locks with respect to inheritance and behavior on close. This is implemented primarily by changing how fl_owner field is set for these locks. Instead of having them owned by the files_struct of the process, they are instead owned by the filp on which they were acquired. Thus, they are inherited across fork() and are only released when the last reference to a filp is put. These new semantics prevent them from being merged with classic POSIX locks, even if they are acquired by the same process. These locks will also conflict with classic POSIX locks even if they are acquired by the same process or on the same file descriptor. The new locks are managed using a new set of cmd values to the fcntl() syscall. The initial implementation of this converts these values to "classic" cmd values at a fairly high level, and the details are not exposed to the underlying filesystem. We may eventually want to push this handing out to the lower filesystem code but for now I don't see any need for it. Also, note that with this implementation the new cmd values are only available via fcntl64() on 32-bit arches. There's little need to add support for legacy apps on a new interface like this. Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:10 +07:00
locks: only validate the lock vs. f_mode in F_SETLK codepaths v2: replace missing break in switch statement (as pointed out by Dave Jones) commit bce7560d4946 (locks: consolidate checks for compatible filp->f_mode values in setlk handlers) introduced a regression in the F_GETLK handler. flock64_to_posix_lock is a shared codepath between F_GETLK and F_SETLK, but the f_mode checks should only be applicable to the F_SETLK codepaths according to POSIX. Instead of just reverting the patch, add a new function to do this checking and have the F_SETLK handlers call it. Cc: Dave Jones <davej@redhat.com> Reported-and-Tested-by: Reuben Farrelly <reuben@reub.net> Signed-off-by: Jeff Layton <jlayton@poochiereds.net>
2014-05-09 22:41:54 +07:00
error = check_fmode_for_setlk(file_lock);
if (error)
goto out;
locks: add new fcntl cmd values for handling file private locks Due to some unfortunate history, POSIX locks have very strange and unhelpful semantics. The thing that usually catches people by surprise is that they are dropped whenever the process closes any file descriptor associated with the inode. This is extremely problematic for people developing file servers that need to implement byte-range locks. Developers often need a "lock management" facility to ensure that file descriptors are not closed until all of the locks associated with the inode are finished. Additionally, "classic" POSIX locks are owned by the process. Locks taken between threads within the same process won't conflict with one another, which renders them useless for synchronization between threads. This patchset adds a new type of lock that attempts to address these issues. These locks conflict with classic POSIX read/write locks, but have semantics that are more like BSD locks with respect to inheritance and behavior on close. This is implemented primarily by changing how fl_owner field is set for these locks. Instead of having them owned by the files_struct of the process, they are instead owned by the filp on which they were acquired. Thus, they are inherited across fork() and are only released when the last reference to a filp is put. These new semantics prevent them from being merged with classic POSIX locks, even if they are acquired by the same process. These locks will also conflict with classic POSIX locks even if they are acquired by the same process or on the same file descriptor. The new locks are managed using a new set of cmd values to the fcntl() syscall. The initial implementation of this converts these values to "classic" cmd values at a fairly high level, and the details are not exposed to the underlying filesystem. We may eventually want to push this handing out to the lower filesystem code but for now I don't see any need for it. Also, note that with this implementation the new cmd values are only available via fcntl64() on 32-bit arches. There's little need to add support for legacy apps on a new interface like this. Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:10 +07:00
/*
* If the cmd is requesting file-private locks, then set the
locks: rename FL_FILE_PVT and IS_FILE_PVT to use "*_OFDLCK" instead File-private locks have been re-christened as "open file description" locks. Finish the symbol name cleanup in the internal implementation. Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-04-22 19:24:32 +07:00
* FL_OFDLCK flag and override the owner.
locks: add new fcntl cmd values for handling file private locks Due to some unfortunate history, POSIX locks have very strange and unhelpful semantics. The thing that usually catches people by surprise is that they are dropped whenever the process closes any file descriptor associated with the inode. This is extremely problematic for people developing file servers that need to implement byte-range locks. Developers often need a "lock management" facility to ensure that file descriptors are not closed until all of the locks associated with the inode are finished. Additionally, "classic" POSIX locks are owned by the process. Locks taken between threads within the same process won't conflict with one another, which renders them useless for synchronization between threads. This patchset adds a new type of lock that attempts to address these issues. These locks conflict with classic POSIX read/write locks, but have semantics that are more like BSD locks with respect to inheritance and behavior on close. This is implemented primarily by changing how fl_owner field is set for these locks. Instead of having them owned by the files_struct of the process, they are instead owned by the filp on which they were acquired. Thus, they are inherited across fork() and are only released when the last reference to a filp is put. These new semantics prevent them from being merged with classic POSIX locks, even if they are acquired by the same process. These locks will also conflict with classic POSIX locks even if they are acquired by the same process or on the same file descriptor. The new locks are managed using a new set of cmd values to the fcntl() syscall. The initial implementation of this converts these values to "classic" cmd values at a fairly high level, and the details are not exposed to the underlying filesystem. We may eventually want to push this handing out to the lower filesystem code but for now I don't see any need for it. Also, note that with this implementation the new cmd values are only available via fcntl64() on 32-bit arches. There's little need to add support for legacy apps on a new interface like this. Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:10 +07:00
*/
switch (cmd) {
locks: rename file-private locks to "open file description locks" File-private locks have been merged into Linux for v3.15, and *now* people are commenting that the name and macro definitions for the new file-private locks suck. ...and I can't even disagree. The names and command macros do suck. We're going to have to live with these for a long time, so it's important that we be happy with the names before we're stuck with them. The consensus on the lists so far is that they should be rechristened as "open file description locks". The name isn't a big deal for the kernel, but the command macros are not visually distinct enough from the traditional POSIX lock macros. The glibc and documentation folks are recommending that we change them to look like F_OFD_{GETLK|SETLK|SETLKW}. That lessens the chance that a programmer will typo one of the commands wrong, and also makes it easier to spot this difference when reading code. This patch makes the following changes that I think are necessary before v3.15 ships: 1) rename the command macros to their new names. These end up in the uapi headers and so are part of the external-facing API. It turns out that glibc doesn't actually use the fcntl.h uapi header, but it's hard to be sure that something else won't. Changing it now is safest. 2) make the the /proc/locks output display these as type "OFDLCK" Cc: Michael Kerrisk <mtk.manpages@gmail.com> Cc: Christoph Hellwig <hch@infradead.org> Cc: Carlos O'Donell <carlos@redhat.com> Cc: Stefan Metzmacher <metze@samba.org> Cc: Andy Lutomirski <luto@amacapital.net> Cc: Frank Filz <ffilzlnx@mindspring.com> Cc: Theodore Ts'o <tytso@mit.edu> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-04-22 19:23:58 +07:00
case F_OFD_SETLK:
locks: require that flock->l_pid be set to 0 for file-private locks Neil Brown suggested potentially overloading the l_pid value as a "lock context" field for file-private locks. While I don't think we will probably want to do that here, it's probably a good idea to ensure that in the future we could extend this API without breaking existing callers. Typically the l_pid value is ignored for incoming struct flock arguments, serving mainly as a place to return the pid of the owner if there is a conflicting lock. For file-private locks, require that it currently be set to 0 and return EINVAL if it isn't. If we eventually want to make a non-zero l_pid mean something, then this will help ensure that we don't break legacy programs that are using file-private locks. Cc: Neil Brown <neilb@suse.de> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-03-04 22:30:23 +07:00
error = -EINVAL;
fs/locks: pass kernel struct flock to fcntl_getlk/setlk This will make it easier to implement a sane compat fcntl syscall. [ jlayton: fix undeclared identifiers in 32-bit fcntl64 syscall handler ] Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-05-27 17:07:19 +07:00
if (flock->l_pid != 0)
locks: require that flock->l_pid be set to 0 for file-private locks Neil Brown suggested potentially overloading the l_pid value as a "lock context" field for file-private locks. While I don't think we will probably want to do that here, it's probably a good idea to ensure that in the future we could extend this API without breaking existing callers. Typically the l_pid value is ignored for incoming struct flock arguments, serving mainly as a place to return the pid of the owner if there is a conflicting lock. For file-private locks, require that it currently be set to 0 and return EINVAL if it isn't. If we eventually want to make a non-zero l_pid mean something, then this will help ensure that we don't break legacy programs that are using file-private locks. Cc: Neil Brown <neilb@suse.de> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-03-04 22:30:23 +07:00
goto out;
locks: add new fcntl cmd values for handling file private locks Due to some unfortunate history, POSIX locks have very strange and unhelpful semantics. The thing that usually catches people by surprise is that they are dropped whenever the process closes any file descriptor associated with the inode. This is extremely problematic for people developing file servers that need to implement byte-range locks. Developers often need a "lock management" facility to ensure that file descriptors are not closed until all of the locks associated with the inode are finished. Additionally, "classic" POSIX locks are owned by the process. Locks taken between threads within the same process won't conflict with one another, which renders them useless for synchronization between threads. This patchset adds a new type of lock that attempts to address these issues. These locks conflict with classic POSIX read/write locks, but have semantics that are more like BSD locks with respect to inheritance and behavior on close. This is implemented primarily by changing how fl_owner field is set for these locks. Instead of having them owned by the files_struct of the process, they are instead owned by the filp on which they were acquired. Thus, they are inherited across fork() and are only released when the last reference to a filp is put. These new semantics prevent them from being merged with classic POSIX locks, even if they are acquired by the same process. These locks will also conflict with classic POSIX locks even if they are acquired by the same process or on the same file descriptor. The new locks are managed using a new set of cmd values to the fcntl() syscall. The initial implementation of this converts these values to "classic" cmd values at a fairly high level, and the details are not exposed to the underlying filesystem. We may eventually want to push this handing out to the lower filesystem code but for now I don't see any need for it. Also, note that with this implementation the new cmd values are only available via fcntl64() on 32-bit arches. There's little need to add support for legacy apps on a new interface like this. Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:10 +07:00
cmd = F_SETLK;
locks: rename FL_FILE_PVT and IS_FILE_PVT to use "*_OFDLCK" instead File-private locks have been re-christened as "open file description" locks. Finish the symbol name cleanup in the internal implementation. Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-04-22 19:24:32 +07:00
file_lock->fl_flags |= FL_OFDLCK;
locks: purge fl_owner_t from fs/locks.c Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-07-13 22:00:38 +07:00
file_lock->fl_owner = filp;
locks: add new fcntl cmd values for handling file private locks Due to some unfortunate history, POSIX locks have very strange and unhelpful semantics. The thing that usually catches people by surprise is that they are dropped whenever the process closes any file descriptor associated with the inode. This is extremely problematic for people developing file servers that need to implement byte-range locks. Developers often need a "lock management" facility to ensure that file descriptors are not closed until all of the locks associated with the inode are finished. Additionally, "classic" POSIX locks are owned by the process. Locks taken between threads within the same process won't conflict with one another, which renders them useless for synchronization between threads. This patchset adds a new type of lock that attempts to address these issues. These locks conflict with classic POSIX read/write locks, but have semantics that are more like BSD locks with respect to inheritance and behavior on close. This is implemented primarily by changing how fl_owner field is set for these locks. Instead of having them owned by the files_struct of the process, they are instead owned by the filp on which they were acquired. Thus, they are inherited across fork() and are only released when the last reference to a filp is put. These new semantics prevent them from being merged with classic POSIX locks, even if they are acquired by the same process. These locks will also conflict with classic POSIX locks even if they are acquired by the same process or on the same file descriptor. The new locks are managed using a new set of cmd values to the fcntl() syscall. The initial implementation of this converts these values to "classic" cmd values at a fairly high level, and the details are not exposed to the underlying filesystem. We may eventually want to push this handing out to the lower filesystem code but for now I don't see any need for it. Also, note that with this implementation the new cmd values are only available via fcntl64() on 32-bit arches. There's little need to add support for legacy apps on a new interface like this. Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:10 +07:00
break;
locks: rename file-private locks to "open file description locks" File-private locks have been merged into Linux for v3.15, and *now* people are commenting that the name and macro definitions for the new file-private locks suck. ...and I can't even disagree. The names and command macros do suck. We're going to have to live with these for a long time, so it's important that we be happy with the names before we're stuck with them. The consensus on the lists so far is that they should be rechristened as "open file description locks". The name isn't a big deal for the kernel, but the command macros are not visually distinct enough from the traditional POSIX lock macros. The glibc and documentation folks are recommending that we change them to look like F_OFD_{GETLK|SETLK|SETLKW}. That lessens the chance that a programmer will typo one of the commands wrong, and also makes it easier to spot this difference when reading code. This patch makes the following changes that I think are necessary before v3.15 ships: 1) rename the command macros to their new names. These end up in the uapi headers and so are part of the external-facing API. It turns out that glibc doesn't actually use the fcntl.h uapi header, but it's hard to be sure that something else won't. Changing it now is safest. 2) make the the /proc/locks output display these as type "OFDLCK" Cc: Michael Kerrisk <mtk.manpages@gmail.com> Cc: Christoph Hellwig <hch@infradead.org> Cc: Carlos O'Donell <carlos@redhat.com> Cc: Stefan Metzmacher <metze@samba.org> Cc: Andy Lutomirski <luto@amacapital.net> Cc: Frank Filz <ffilzlnx@mindspring.com> Cc: Theodore Ts'o <tytso@mit.edu> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-04-22 19:23:58 +07:00
case F_OFD_SETLKW:
locks: require that flock->l_pid be set to 0 for file-private locks Neil Brown suggested potentially overloading the l_pid value as a "lock context" field for file-private locks. While I don't think we will probably want to do that here, it's probably a good idea to ensure that in the future we could extend this API without breaking existing callers. Typically the l_pid value is ignored for incoming struct flock arguments, serving mainly as a place to return the pid of the owner if there is a conflicting lock. For file-private locks, require that it currently be set to 0 and return EINVAL if it isn't. If we eventually want to make a non-zero l_pid mean something, then this will help ensure that we don't break legacy programs that are using file-private locks. Cc: Neil Brown <neilb@suse.de> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-03-04 22:30:23 +07:00
error = -EINVAL;
fs/locks: pass kernel struct flock to fcntl_getlk/setlk This will make it easier to implement a sane compat fcntl syscall. [ jlayton: fix undeclared identifiers in 32-bit fcntl64 syscall handler ] Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-05-27 17:07:19 +07:00
if (flock->l_pid != 0)
locks: require that flock->l_pid be set to 0 for file-private locks Neil Brown suggested potentially overloading the l_pid value as a "lock context" field for file-private locks. While I don't think we will probably want to do that here, it's probably a good idea to ensure that in the future we could extend this API without breaking existing callers. Typically the l_pid value is ignored for incoming struct flock arguments, serving mainly as a place to return the pid of the owner if there is a conflicting lock. For file-private locks, require that it currently be set to 0 and return EINVAL if it isn't. If we eventually want to make a non-zero l_pid mean something, then this will help ensure that we don't break legacy programs that are using file-private locks. Cc: Neil Brown <neilb@suse.de> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-03-04 22:30:23 +07:00
goto out;
locks: add new fcntl cmd values for handling file private locks Due to some unfortunate history, POSIX locks have very strange and unhelpful semantics. The thing that usually catches people by surprise is that they are dropped whenever the process closes any file descriptor associated with the inode. This is extremely problematic for people developing file servers that need to implement byte-range locks. Developers often need a "lock management" facility to ensure that file descriptors are not closed until all of the locks associated with the inode are finished. Additionally, "classic" POSIX locks are owned by the process. Locks taken between threads within the same process won't conflict with one another, which renders them useless for synchronization between threads. This patchset adds a new type of lock that attempts to address these issues. These locks conflict with classic POSIX read/write locks, but have semantics that are more like BSD locks with respect to inheritance and behavior on close. This is implemented primarily by changing how fl_owner field is set for these locks. Instead of having them owned by the files_struct of the process, they are instead owned by the filp on which they were acquired. Thus, they are inherited across fork() and are only released when the last reference to a filp is put. These new semantics prevent them from being merged with classic POSIX locks, even if they are acquired by the same process. These locks will also conflict with classic POSIX locks even if they are acquired by the same process or on the same file descriptor. The new locks are managed using a new set of cmd values to the fcntl() syscall. The initial implementation of this converts these values to "classic" cmd values at a fairly high level, and the details are not exposed to the underlying filesystem. We may eventually want to push this handing out to the lower filesystem code but for now I don't see any need for it. Also, note that with this implementation the new cmd values are only available via fcntl64() on 32-bit arches. There's little need to add support for legacy apps on a new interface like this. Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:10 +07:00
cmd = F_SETLKW;
locks: rename FL_FILE_PVT and IS_FILE_PVT to use "*_OFDLCK" instead File-private locks have been re-christened as "open file description" locks. Finish the symbol name cleanup in the internal implementation. Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-04-22 19:24:32 +07:00
file_lock->fl_flags |= FL_OFDLCK;
locks: purge fl_owner_t from fs/locks.c Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-07-13 22:00:38 +07:00
file_lock->fl_owner = filp;
locks: add new fcntl cmd values for handling file private locks Due to some unfortunate history, POSIX locks have very strange and unhelpful semantics. The thing that usually catches people by surprise is that they are dropped whenever the process closes any file descriptor associated with the inode. This is extremely problematic for people developing file servers that need to implement byte-range locks. Developers often need a "lock management" facility to ensure that file descriptors are not closed until all of the locks associated with the inode are finished. Additionally, "classic" POSIX locks are owned by the process. Locks taken between threads within the same process won't conflict with one another, which renders them useless for synchronization between threads. This patchset adds a new type of lock that attempts to address these issues. These locks conflict with classic POSIX read/write locks, but have semantics that are more like BSD locks with respect to inheritance and behavior on close. This is implemented primarily by changing how fl_owner field is set for these locks. Instead of having them owned by the files_struct of the process, they are instead owned by the filp on which they were acquired. Thus, they are inherited across fork() and are only released when the last reference to a filp is put. These new semantics prevent them from being merged with classic POSIX locks, even if they are acquired by the same process. These locks will also conflict with classic POSIX locks even if they are acquired by the same process or on the same file descriptor. The new locks are managed using a new set of cmd values to the fcntl() syscall. The initial implementation of this converts these values to "classic" cmd values at a fairly high level, and the details are not exposed to the underlying filesystem. We may eventually want to push this handing out to the lower filesystem code but for now I don't see any need for it. Also, note that with this implementation the new cmd values are only available via fcntl64() on 32-bit arches. There's little need to add support for legacy apps on a new interface like this. Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:10 +07:00
/* Fallthrough */
case F_SETLKW:
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
file_lock->fl_flags |= FL_SLEEP;
}
locks: add new fcntl cmd values for handling file private locks Due to some unfortunate history, POSIX locks have very strange and unhelpful semantics. The thing that usually catches people by surprise is that they are dropped whenever the process closes any file descriptor associated with the inode. This is extremely problematic for people developing file servers that need to implement byte-range locks. Developers often need a "lock management" facility to ensure that file descriptors are not closed until all of the locks associated with the inode are finished. Additionally, "classic" POSIX locks are owned by the process. Locks taken between threads within the same process won't conflict with one another, which renders them useless for synchronization between threads. This patchset adds a new type of lock that attempts to address these issues. These locks conflict with classic POSIX read/write locks, but have semantics that are more like BSD locks with respect to inheritance and behavior on close. This is implemented primarily by changing how fl_owner field is set for these locks. Instead of having them owned by the files_struct of the process, they are instead owned by the filp on which they were acquired. Thus, they are inherited across fork() and are only released when the last reference to a filp is put. These new semantics prevent them from being merged with classic POSIX locks, even if they are acquired by the same process. These locks will also conflict with classic POSIX locks even if they are acquired by the same process or on the same file descriptor. The new locks are managed using a new set of cmd values to the fcntl() syscall. The initial implementation of this converts these values to "classic" cmd values at a fairly high level, and the details are not exposed to the underlying filesystem. We may eventually want to push this handing out to the lower filesystem code but for now I don't see any need for it. Also, note that with this implementation the new cmd values are only available via fcntl64() on 32-bit arches. There's little need to add support for legacy apps on a new interface like this. Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:10 +07:00
locks: cleanup code duplication Extract common code into a function. Signed-off-by: Miklos Szeredi <mszeredi@suse.cz> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Cc: Matthew Wilcox <matthew@wil.cx> Cc: David Teigland <teigland@redhat.com> Cc: Christoph Hellwig <hch@lst.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-07-25 15:48:57 +07:00
error = do_lock_file_wait(filp, cmd, file_lock);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
[PATCH] stale POSIX lock handling 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>
2005-07-28 01:45:09 +07:00
/*
locks: don't check for race with close when setting OFD lock We don't clean out OFD locks on close(), so there's no need to check for a race with them here. They'll get cleaned out at the same time that flock locks are. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Acked-by: "J. Bruce Fields" <bfields@fieldses.org>
2016-01-08 19:30:43 +07:00
* Attempt to detect a close/fcntl race and recover by releasing the
* lock that was just acquired. There is no need to do that when we're
* unlocking though, or for OFD locks.
[PATCH] stale POSIX lock handling 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>
2005-07-28 01:45:09 +07:00
*/
locks: don't check for race with close when setting OFD lock We don't clean out OFD locks on close(), so there's no need to check for a race with them here. They'll get cleaned out at the same time that flock locks are. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Acked-by: "J. Bruce Fields" <bfields@fieldses.org>
2016-01-08 19:30:43 +07:00
if (!error && file_lock->fl_type != F_UNLCK &&
!(file_lock->fl_flags & FL_OFDLCK)) {
locks: fix unlock when fcntl_setlk races with a close Dmitry reported that he was able to reproduce the WARN_ON_ONCE that fires in locks_free_lock_context when the flc_posix list isn't empty. The problem turns out to be that we're basically rebuilding the file_lock from scratch in fcntl_setlk when we discover that the setlk has raced with a close. If the l_whence field is SEEK_CUR or SEEK_END, then we may end up with fl_start and fl_end values that differ from when the lock was initially set, if the file position or length of the file has changed in the interim. Fix this by just reusing the same lock request structure, and simply override fl_type value with F_UNLCK as appropriate. That ensures that we really are unlocking the lock that was initially set. While we're there, make sure that we do pop a WARN_ON_ONCE if the removal ever fails. Also return -EBADF in this event, since that's what we would have returned if the close had happened earlier. Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: <stable@vger.kernel.org> Fixes: c293621bbf67 (stale POSIX lock handling) Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Acked-by: "J. Bruce Fields" <bfields@fieldses.org>
2016-01-08 04:38:10 +07:00
/*
* We need that spin_lock here - it prevents reordering between
* update of i_flctx->flc_posix and check for it done in
* close(). rcu_read_lock() wouldn't do.
*/
spin_lock(&current->files->file_lock);
f = fcheck(fd);
spin_unlock(&current->files->file_lock);
if (f != filp) {
file_lock->fl_type = F_UNLCK;
error = do_lock_file_wait(filp, cmd, file_lock);
WARN_ON_ONCE(error);
error = -EBADF;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
[PATCH] stale POSIX lock handling 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>
2005-07-28 01:45:09 +07:00
out:
locks: sprinkle some tracepoints around the file locking code Add some tracepoints around the POSIX locking code. These were useful when tracking down problems when handling the race between setlk and close. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Acked-by: "J. Bruce Fields" <bfields@fieldses.org>
2016-01-07 09:26:10 +07:00
trace_fcntl_setlk(inode, file_lock, error);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
locks_free_lock(file_lock);
return error;
}
#if BITS_PER_LONG == 32
/* Report the first existing lock that would conflict with l.
* This implements the F_GETLK command of fcntl().
*/
fs/locks: pass kernel struct flock to fcntl_getlk/setlk This will make it easier to implement a sane compat fcntl syscall. [ jlayton: fix undeclared identifiers in 32-bit fcntl64 syscall handler ] Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-05-27 17:07:19 +07:00
int fcntl_getlk64(struct file *filp, unsigned int cmd, struct flock64 *flock)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
fs/locks: Use allocation rather than the stack in fcntl_getlk() Struct file_lock is fairly large, so let's save some space on the stack by using an allocation for struct file_lock in fcntl_getlk(), just as we do for fcntl_setlk(). Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-07-16 21:28:21 +07:00
struct file_lock *fl;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
int error;
fs/locks: Use allocation rather than the stack in fcntl_getlk() Struct file_lock is fairly large, so let's save some space on the stack by using an allocation for struct file_lock in fcntl_getlk(), just as we do for fcntl_setlk(). Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-07-16 21:28:21 +07:00
fl = locks_alloc_lock();
if (fl == NULL)
return -ENOMEM;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
error = -EINVAL;
fs/locks: pass kernel struct flock to fcntl_getlk/setlk This will make it easier to implement a sane compat fcntl syscall. [ jlayton: fix undeclared identifiers in 32-bit fcntl64 syscall handler ] Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-05-27 17:07:19 +07:00
if (flock->l_type != F_RDLCK && flock->l_type != F_WRLCK)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
goto out;
fs/locks: Use allocation rather than the stack in fcntl_getlk() Struct file_lock is fairly large, so let's save some space on the stack by using an allocation for struct file_lock in fcntl_getlk(), just as we do for fcntl_setlk(). Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-07-16 21:28:21 +07:00
error = flock64_to_posix_lock(filp, fl, flock);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
if (error)
goto out;
locks: rename file-private locks to "open file description locks" File-private locks have been merged into Linux for v3.15, and *now* people are commenting that the name and macro definitions for the new file-private locks suck. ...and I can't even disagree. The names and command macros do suck. We're going to have to live with these for a long time, so it's important that we be happy with the names before we're stuck with them. The consensus on the lists so far is that they should be rechristened as "open file description locks". The name isn't a big deal for the kernel, but the command macros are not visually distinct enough from the traditional POSIX lock macros. The glibc and documentation folks are recommending that we change them to look like F_OFD_{GETLK|SETLK|SETLKW}. That lessens the chance that a programmer will typo one of the commands wrong, and also makes it easier to spot this difference when reading code. This patch makes the following changes that I think are necessary before v3.15 ships: 1) rename the command macros to their new names. These end up in the uapi headers and so are part of the external-facing API. It turns out that glibc doesn't actually use the fcntl.h uapi header, but it's hard to be sure that something else won't. Changing it now is safest. 2) make the the /proc/locks output display these as type "OFDLCK" Cc: Michael Kerrisk <mtk.manpages@gmail.com> Cc: Christoph Hellwig <hch@infradead.org> Cc: Carlos O'Donell <carlos@redhat.com> Cc: Stefan Metzmacher <metze@samba.org> Cc: Andy Lutomirski <luto@amacapital.net> Cc: Frank Filz <ffilzlnx@mindspring.com> Cc: Theodore Ts'o <tytso@mit.edu> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-04-22 19:23:58 +07:00
if (cmd == F_OFD_GETLK) {
locks: require that flock->l_pid be set to 0 for file-private locks Neil Brown suggested potentially overloading the l_pid value as a "lock context" field for file-private locks. While I don't think we will probably want to do that here, it's probably a good idea to ensure that in the future we could extend this API without breaking existing callers. Typically the l_pid value is ignored for incoming struct flock arguments, serving mainly as a place to return the pid of the owner if there is a conflicting lock. For file-private locks, require that it currently be set to 0 and return EINVAL if it isn't. If we eventually want to make a non-zero l_pid mean something, then this will help ensure that we don't break legacy programs that are using file-private locks. Cc: Neil Brown <neilb@suse.de> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-03-04 22:30:23 +07:00
error = -EINVAL;
fs/locks: pass kernel struct flock to fcntl_getlk/setlk This will make it easier to implement a sane compat fcntl syscall. [ jlayton: fix undeclared identifiers in 32-bit fcntl64 syscall handler ] Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-05-27 17:07:19 +07:00
if (flock->l_pid != 0)
locks: require that flock->l_pid be set to 0 for file-private locks Neil Brown suggested potentially overloading the l_pid value as a "lock context" field for file-private locks. While I don't think we will probably want to do that here, it's probably a good idea to ensure that in the future we could extend this API without breaking existing callers. Typically the l_pid value is ignored for incoming struct flock arguments, serving mainly as a place to return the pid of the owner if there is a conflicting lock. For file-private locks, require that it currently be set to 0 and return EINVAL if it isn't. If we eventually want to make a non-zero l_pid mean something, then this will help ensure that we don't break legacy programs that are using file-private locks. Cc: Neil Brown <neilb@suse.de> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-03-04 22:30:23 +07:00
goto out;
locks: add new fcntl cmd values for handling file private locks Due to some unfortunate history, POSIX locks have very strange and unhelpful semantics. The thing that usually catches people by surprise is that they are dropped whenever the process closes any file descriptor associated with the inode. This is extremely problematic for people developing file servers that need to implement byte-range locks. Developers often need a "lock management" facility to ensure that file descriptors are not closed until all of the locks associated with the inode are finished. Additionally, "classic" POSIX locks are owned by the process. Locks taken between threads within the same process won't conflict with one another, which renders them useless for synchronization between threads. This patchset adds a new type of lock that attempts to address these issues. These locks conflict with classic POSIX read/write locks, but have semantics that are more like BSD locks with respect to inheritance and behavior on close. This is implemented primarily by changing how fl_owner field is set for these locks. Instead of having them owned by the files_struct of the process, they are instead owned by the filp on which they were acquired. Thus, they are inherited across fork() and are only released when the last reference to a filp is put. These new semantics prevent them from being merged with classic POSIX locks, even if they are acquired by the same process. These locks will also conflict with classic POSIX locks even if they are acquired by the same process or on the same file descriptor. The new locks are managed using a new set of cmd values to the fcntl() syscall. The initial implementation of this converts these values to "classic" cmd values at a fairly high level, and the details are not exposed to the underlying filesystem. We may eventually want to push this handing out to the lower filesystem code but for now I don't see any need for it. Also, note that with this implementation the new cmd values are only available via fcntl64() on 32-bit arches. There's little need to add support for legacy apps on a new interface like this. Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:10 +07:00
cmd = F_GETLK64;
fs/locks: Use allocation rather than the stack in fcntl_getlk() Struct file_lock is fairly large, so let's save some space on the stack by using an allocation for struct file_lock in fcntl_getlk(), just as we do for fcntl_setlk(). Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-07-16 21:28:21 +07:00
fl->fl_flags |= FL_OFDLCK;
fl->fl_owner = filp;
locks: add new fcntl cmd values for handling file private locks Due to some unfortunate history, POSIX locks have very strange and unhelpful semantics. The thing that usually catches people by surprise is that they are dropped whenever the process closes any file descriptor associated with the inode. This is extremely problematic for people developing file servers that need to implement byte-range locks. Developers often need a "lock management" facility to ensure that file descriptors are not closed until all of the locks associated with the inode are finished. Additionally, "classic" POSIX locks are owned by the process. Locks taken between threads within the same process won't conflict with one another, which renders them useless for synchronization between threads. This patchset adds a new type of lock that attempts to address these issues. These locks conflict with classic POSIX read/write locks, but have semantics that are more like BSD locks with respect to inheritance and behavior on close. This is implemented primarily by changing how fl_owner field is set for these locks. Instead of having them owned by the files_struct of the process, they are instead owned by the filp on which they were acquired. Thus, they are inherited across fork() and are only released when the last reference to a filp is put. These new semantics prevent them from being merged with classic POSIX locks, even if they are acquired by the same process. These locks will also conflict with classic POSIX locks even if they are acquired by the same process or on the same file descriptor. The new locks are managed using a new set of cmd values to the fcntl() syscall. The initial implementation of this converts these values to "classic" cmd values at a fairly high level, and the details are not exposed to the underlying filesystem. We may eventually want to push this handing out to the lower filesystem code but for now I don't see any need for it. Also, note that with this implementation the new cmd values are only available via fcntl64() on 32-bit arches. There's little need to add support for legacy apps on a new interface like this. Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:10 +07:00
}
fs/locks: Use allocation rather than the stack in fcntl_getlk() Struct file_lock is fairly large, so let's save some space on the stack by using an allocation for struct file_lock in fcntl_getlk(), just as we do for fcntl_setlk(). Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-07-16 21:28:21 +07:00
error = vfs_test_lock(filp, fl);
locks: factor out generic/filesystem switch from test_lock Factor out the code that switches between generic and filesystem-specific lock methods; eventually we want to call this from lock managers (lockd and nfsd) too; currently they only call the generic methods. This patch does that for test_lock. Note that this hasn't been necessary until recently, because the few filesystems that define ->lock() (nfs, cifs...) aren't exportable via NFS. However GFS (and, in the future, other cluster filesystems) need to implement their own locking to get cluster-coherent locking, and also want to be able to export locking to NFS (lockd and NFSv4). So we accomplish this by factoring out code such as this and exporting it for the use of lockd and nfsd. Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu>
2007-02-21 12:58:50 +07:00
if (error)
goto out;
fs/locks: Use allocation rather than the stack in fcntl_getlk() Struct file_lock is fairly large, so let's save some space on the stack by using an allocation for struct file_lock in fcntl_getlk(), just as we do for fcntl_setlk(). Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-07-16 21:28:21 +07:00
flock->l_type = fl->fl_type;
if (fl->fl_type != F_UNLCK)
posix_lock_to_flock64(flock, fl);
locks: Copy fl_lmops information for conflock in locks_copy_conflock() Commit d5b9026a67 ([PATCH] knfsd: locks: flag NFSv4-owned locks) using fl_lmops field in file_lock for checking nfsd4 lockowner. But, commit 1a747ee0cc (locks: don't call ->copy_lock methods on return of conflicting locks) causes the fl_lmops of conflock always be NULL. Also, commit 0996905f93 (lockd: posix_test_lock() should not call locks_copy_lock()) caused the fl_lmops of conflock always be NULL too. Make sure copy the private information by fl_copy_lock() in struct file_lock_operations, merge __locks_copy_lock() to fl_copy_lock(). Jeff advice, "Set fl_lmops on conflocks, but don't set fl_ops. fl_ops are superfluous, since they are callbacks into the filesystem. There should be no need to bother the filesystem at all with info in a conflock. But, lock _ownership_ matters for conflocks and that's indicated by the fl_lmops. So you really do want to copy the fl_lmops for conflocks I think." v5: add missing calling of locks_release_private() in nlmsvc_testlock() v4: only copy fl_lmops for conflock, don't copy fl_ops Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-08-22 21:18:43 +07:00
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
out:
fs/locks: Use allocation rather than the stack in fcntl_getlk() Struct file_lock is fairly large, so let's save some space on the stack by using an allocation for struct file_lock in fcntl_getlk(), just as we do for fcntl_setlk(). Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-07-16 21:28:21 +07:00
locks_free_lock(fl);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
return error;
}
/* Apply the lock described by l to an open file descriptor.
* This implements both the F_SETLK and F_SETLKW commands of fcntl().
*/
[PATCH] stale POSIX lock handling 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>
2005-07-28 01:45:09 +07:00
int fcntl_setlk64(unsigned int fd, struct file *filp, unsigned int cmd,
fs/locks: pass kernel struct flock to fcntl_getlk/setlk This will make it easier to implement a sane compat fcntl syscall. [ jlayton: fix undeclared identifiers in 32-bit fcntl64 syscall handler ] Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-05-27 17:07:19 +07:00
struct flock64 *flock)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
struct file_lock *file_lock = locks_alloc_lock();
fs/locks: pass kernel struct flock to fcntl_getlk/setlk This will make it easier to implement a sane compat fcntl syscall. [ jlayton: fix undeclared identifiers in 32-bit fcntl64 syscall handler ] Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-05-27 17:07:19 +07:00
struct inode *inode = locks_inode(filp);
[PATCH] fix SMP ordering hole in fcntl_setlk() fcntl_setlk()/close() race prevention has a subtle hole - we need to make sure that if we *do* have an fcntl/close race on SMP box, the access to descriptor table and inode->i_flock won't get reordered. As it is, we get STORE inode->i_flock, LOAD descriptor table entry vs. STORE descriptor table entry, LOAD inode->i_flock with not a single lock in common on both sides. We do have BKL around the first STORE, but check in locks_remove_posix() is outside of BKL and for a good reason - we don't want BKL on common path of close(2). Solution is to hold ->file_lock around fcheck() in there; that orders us wrt removal from descriptor table that preceded locks_remove_posix() on close path and we either come first (in which case eviction will be handled by the close side) or we'll see the effect of close and do eviction ourselves. Note that even though it's read-only access, we do need ->file_lock here - rcu_read_lock() won't be enough to order the things. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2008-05-07 00:58:34 +07:00
struct file *f;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
int error;
if (file_lock == NULL)
return -ENOLCK;
/* Don't allow mandatory locks on files that may be memory mapped
* and shared.
*/
Cleanup macros for distinguishing mandatory locks The combination of S_ISGID bit set and S_IXGRP bit unset is used to mark the inode as "mandatory lockable" and there's a macro for this check called MANDATORY_LOCK(inode). However, fs/locks.c and some filesystems still perform the explicit i_mode checking. Besides, Andrew pointed out, that this macro is buggy itself, as it dereferences the inode arg twice. Convert this macro into static inline function and switch its users to it, making the code shorter and more readable. The __mandatory_lock() helper is to be used in places where the IS_MANDLOCK() for superblock is already known to be true. Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: David Howells <dhowells@redhat.com> Cc: Eric Van Hensbergen <ericvh@gmail.com> Cc: Ron Minnich <rminnich@sandia.gov> Cc: Latchesar Ionkov <lucho@ionkov.net> Cc: Steven Whitehouse <swhiteho@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2007-10-02 04:41:11 +07:00
if (mandatory_lock(inode) && mapping_writably_mapped(filp->f_mapping)) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
error = -EAGAIN;
goto out;
}
fs/locks: pass kernel struct flock to fcntl_getlk/setlk This will make it easier to implement a sane compat fcntl syscall. [ jlayton: fix undeclared identifiers in 32-bit fcntl64 syscall handler ] Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-05-27 17:07:19 +07:00
error = flock64_to_posix_lock(filp, file_lock, flock);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
if (error)
goto out;
locks: add new fcntl cmd values for handling file private locks Due to some unfortunate history, POSIX locks have very strange and unhelpful semantics. The thing that usually catches people by surprise is that they are dropped whenever the process closes any file descriptor associated with the inode. This is extremely problematic for people developing file servers that need to implement byte-range locks. Developers often need a "lock management" facility to ensure that file descriptors are not closed until all of the locks associated with the inode are finished. Additionally, "classic" POSIX locks are owned by the process. Locks taken between threads within the same process won't conflict with one another, which renders them useless for synchronization between threads. This patchset adds a new type of lock that attempts to address these issues. These locks conflict with classic POSIX read/write locks, but have semantics that are more like BSD locks with respect to inheritance and behavior on close. This is implemented primarily by changing how fl_owner field is set for these locks. Instead of having them owned by the files_struct of the process, they are instead owned by the filp on which they were acquired. Thus, they are inherited across fork() and are only released when the last reference to a filp is put. These new semantics prevent them from being merged with classic POSIX locks, even if they are acquired by the same process. These locks will also conflict with classic POSIX locks even if they are acquired by the same process or on the same file descriptor. The new locks are managed using a new set of cmd values to the fcntl() syscall. The initial implementation of this converts these values to "classic" cmd values at a fairly high level, and the details are not exposed to the underlying filesystem. We may eventually want to push this handing out to the lower filesystem code but for now I don't see any need for it. Also, note that with this implementation the new cmd values are only available via fcntl64() on 32-bit arches. There's little need to add support for legacy apps on a new interface like this. Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:10 +07:00
locks: only validate the lock vs. f_mode in F_SETLK codepaths v2: replace missing break in switch statement (as pointed out by Dave Jones) commit bce7560d4946 (locks: consolidate checks for compatible filp->f_mode values in setlk handlers) introduced a regression in the F_GETLK handler. flock64_to_posix_lock is a shared codepath between F_GETLK and F_SETLK, but the f_mode checks should only be applicable to the F_SETLK codepaths according to POSIX. Instead of just reverting the patch, add a new function to do this checking and have the F_SETLK handlers call it. Cc: Dave Jones <davej@redhat.com> Reported-and-Tested-by: Reuben Farrelly <reuben@reub.net> Signed-off-by: Jeff Layton <jlayton@poochiereds.net>
2014-05-09 22:41:54 +07:00
error = check_fmode_for_setlk(file_lock);
if (error)
goto out;
locks: add new fcntl cmd values for handling file private locks Due to some unfortunate history, POSIX locks have very strange and unhelpful semantics. The thing that usually catches people by surprise is that they are dropped whenever the process closes any file descriptor associated with the inode. This is extremely problematic for people developing file servers that need to implement byte-range locks. Developers often need a "lock management" facility to ensure that file descriptors are not closed until all of the locks associated with the inode are finished. Additionally, "classic" POSIX locks are owned by the process. Locks taken between threads within the same process won't conflict with one another, which renders them useless for synchronization between threads. This patchset adds a new type of lock that attempts to address these issues. These locks conflict with classic POSIX read/write locks, but have semantics that are more like BSD locks with respect to inheritance and behavior on close. This is implemented primarily by changing how fl_owner field is set for these locks. Instead of having them owned by the files_struct of the process, they are instead owned by the filp on which they were acquired. Thus, they are inherited across fork() and are only released when the last reference to a filp is put. These new semantics prevent them from being merged with classic POSIX locks, even if they are acquired by the same process. These locks will also conflict with classic POSIX locks even if they are acquired by the same process or on the same file descriptor. The new locks are managed using a new set of cmd values to the fcntl() syscall. The initial implementation of this converts these values to "classic" cmd values at a fairly high level, and the details are not exposed to the underlying filesystem. We may eventually want to push this handing out to the lower filesystem code but for now I don't see any need for it. Also, note that with this implementation the new cmd values are only available via fcntl64() on 32-bit arches. There's little need to add support for legacy apps on a new interface like this. Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:10 +07:00
/*
* If the cmd is requesting file-private locks, then set the
locks: rename FL_FILE_PVT and IS_FILE_PVT to use "*_OFDLCK" instead File-private locks have been re-christened as "open file description" locks. Finish the symbol name cleanup in the internal implementation. Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-04-22 19:24:32 +07:00
* FL_OFDLCK flag and override the owner.
locks: add new fcntl cmd values for handling file private locks Due to some unfortunate history, POSIX locks have very strange and unhelpful semantics. The thing that usually catches people by surprise is that they are dropped whenever the process closes any file descriptor associated with the inode. This is extremely problematic for people developing file servers that need to implement byte-range locks. Developers often need a "lock management" facility to ensure that file descriptors are not closed until all of the locks associated with the inode are finished. Additionally, "classic" POSIX locks are owned by the process. Locks taken between threads within the same process won't conflict with one another, which renders them useless for synchronization between threads. This patchset adds a new type of lock that attempts to address these issues. These locks conflict with classic POSIX read/write locks, but have semantics that are more like BSD locks with respect to inheritance and behavior on close. This is implemented primarily by changing how fl_owner field is set for these locks. Instead of having them owned by the files_struct of the process, they are instead owned by the filp on which they were acquired. Thus, they are inherited across fork() and are only released when the last reference to a filp is put. These new semantics prevent them from being merged with classic POSIX locks, even if they are acquired by the same process. These locks will also conflict with classic POSIX locks even if they are acquired by the same process or on the same file descriptor. The new locks are managed using a new set of cmd values to the fcntl() syscall. The initial implementation of this converts these values to "classic" cmd values at a fairly high level, and the details are not exposed to the underlying filesystem. We may eventually want to push this handing out to the lower filesystem code but for now I don't see any need for it. Also, note that with this implementation the new cmd values are only available via fcntl64() on 32-bit arches. There's little need to add support for legacy apps on a new interface like this. Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:10 +07:00
*/
switch (cmd) {
locks: rename file-private locks to "open file description locks" File-private locks have been merged into Linux for v3.15, and *now* people are commenting that the name and macro definitions for the new file-private locks suck. ...and I can't even disagree. The names and command macros do suck. We're going to have to live with these for a long time, so it's important that we be happy with the names before we're stuck with them. The consensus on the lists so far is that they should be rechristened as "open file description locks". The name isn't a big deal for the kernel, but the command macros are not visually distinct enough from the traditional POSIX lock macros. The glibc and documentation folks are recommending that we change them to look like F_OFD_{GETLK|SETLK|SETLKW}. That lessens the chance that a programmer will typo one of the commands wrong, and also makes it easier to spot this difference when reading code. This patch makes the following changes that I think are necessary before v3.15 ships: 1) rename the command macros to their new names. These end up in the uapi headers and so are part of the external-facing API. It turns out that glibc doesn't actually use the fcntl.h uapi header, but it's hard to be sure that something else won't. Changing it now is safest. 2) make the the /proc/locks output display these as type "OFDLCK" Cc: Michael Kerrisk <mtk.manpages@gmail.com> Cc: Christoph Hellwig <hch@infradead.org> Cc: Carlos O'Donell <carlos@redhat.com> Cc: Stefan Metzmacher <metze@samba.org> Cc: Andy Lutomirski <luto@amacapital.net> Cc: Frank Filz <ffilzlnx@mindspring.com> Cc: Theodore Ts'o <tytso@mit.edu> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-04-22 19:23:58 +07:00
case F_OFD_SETLK:
locks: require that flock->l_pid be set to 0 for file-private locks Neil Brown suggested potentially overloading the l_pid value as a "lock context" field for file-private locks. While I don't think we will probably want to do that here, it's probably a good idea to ensure that in the future we could extend this API without breaking existing callers. Typically the l_pid value is ignored for incoming struct flock arguments, serving mainly as a place to return the pid of the owner if there is a conflicting lock. For file-private locks, require that it currently be set to 0 and return EINVAL if it isn't. If we eventually want to make a non-zero l_pid mean something, then this will help ensure that we don't break legacy programs that are using file-private locks. Cc: Neil Brown <neilb@suse.de> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-03-04 22:30:23 +07:00
error = -EINVAL;
fs/locks: pass kernel struct flock to fcntl_getlk/setlk This will make it easier to implement a sane compat fcntl syscall. [ jlayton: fix undeclared identifiers in 32-bit fcntl64 syscall handler ] Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-05-27 17:07:19 +07:00
if (flock->l_pid != 0)
locks: require that flock->l_pid be set to 0 for file-private locks Neil Brown suggested potentially overloading the l_pid value as a "lock context" field for file-private locks. While I don't think we will probably want to do that here, it's probably a good idea to ensure that in the future we could extend this API without breaking existing callers. Typically the l_pid value is ignored for incoming struct flock arguments, serving mainly as a place to return the pid of the owner if there is a conflicting lock. For file-private locks, require that it currently be set to 0 and return EINVAL if it isn't. If we eventually want to make a non-zero l_pid mean something, then this will help ensure that we don't break legacy programs that are using file-private locks. Cc: Neil Brown <neilb@suse.de> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-03-04 22:30:23 +07:00
goto out;
locks: add new fcntl cmd values for handling file private locks Due to some unfortunate history, POSIX locks have very strange and unhelpful semantics. The thing that usually catches people by surprise is that they are dropped whenever the process closes any file descriptor associated with the inode. This is extremely problematic for people developing file servers that need to implement byte-range locks. Developers often need a "lock management" facility to ensure that file descriptors are not closed until all of the locks associated with the inode are finished. Additionally, "classic" POSIX locks are owned by the process. Locks taken between threads within the same process won't conflict with one another, which renders them useless for synchronization between threads. This patchset adds a new type of lock that attempts to address these issues. These locks conflict with classic POSIX read/write locks, but have semantics that are more like BSD locks with respect to inheritance and behavior on close. This is implemented primarily by changing how fl_owner field is set for these locks. Instead of having them owned by the files_struct of the process, they are instead owned by the filp on which they were acquired. Thus, they are inherited across fork() and are only released when the last reference to a filp is put. These new semantics prevent them from being merged with classic POSIX locks, even if they are acquired by the same process. These locks will also conflict with classic POSIX locks even if they are acquired by the same process or on the same file descriptor. The new locks are managed using a new set of cmd values to the fcntl() syscall. The initial implementation of this converts these values to "classic" cmd values at a fairly high level, and the details are not exposed to the underlying filesystem. We may eventually want to push this handing out to the lower filesystem code but for now I don't see any need for it. Also, note that with this implementation the new cmd values are only available via fcntl64() on 32-bit arches. There's little need to add support for legacy apps on a new interface like this. Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:10 +07:00
cmd = F_SETLK64;
locks: rename FL_FILE_PVT and IS_FILE_PVT to use "*_OFDLCK" instead File-private locks have been re-christened as "open file description" locks. Finish the symbol name cleanup in the internal implementation. Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-04-22 19:24:32 +07:00
file_lock->fl_flags |= FL_OFDLCK;
locks: purge fl_owner_t from fs/locks.c Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-07-13 22:00:38 +07:00
file_lock->fl_owner = filp;
locks: add new fcntl cmd values for handling file private locks Due to some unfortunate history, POSIX locks have very strange and unhelpful semantics. The thing that usually catches people by surprise is that they are dropped whenever the process closes any file descriptor associated with the inode. This is extremely problematic for people developing file servers that need to implement byte-range locks. Developers often need a "lock management" facility to ensure that file descriptors are not closed until all of the locks associated with the inode are finished. Additionally, "classic" POSIX locks are owned by the process. Locks taken between threads within the same process won't conflict with one another, which renders them useless for synchronization between threads. This patchset adds a new type of lock that attempts to address these issues. These locks conflict with classic POSIX read/write locks, but have semantics that are more like BSD locks with respect to inheritance and behavior on close. This is implemented primarily by changing how fl_owner field is set for these locks. Instead of having them owned by the files_struct of the process, they are instead owned by the filp on which they were acquired. Thus, they are inherited across fork() and are only released when the last reference to a filp is put. These new semantics prevent them from being merged with classic POSIX locks, even if they are acquired by the same process. These locks will also conflict with classic POSIX locks even if they are acquired by the same process or on the same file descriptor. The new locks are managed using a new set of cmd values to the fcntl() syscall. The initial implementation of this converts these values to "classic" cmd values at a fairly high level, and the details are not exposed to the underlying filesystem. We may eventually want to push this handing out to the lower filesystem code but for now I don't see any need for it. Also, note that with this implementation the new cmd values are only available via fcntl64() on 32-bit arches. There's little need to add support for legacy apps on a new interface like this. Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:10 +07:00
break;
locks: rename file-private locks to "open file description locks" File-private locks have been merged into Linux for v3.15, and *now* people are commenting that the name and macro definitions for the new file-private locks suck. ...and I can't even disagree. The names and command macros do suck. We're going to have to live with these for a long time, so it's important that we be happy with the names before we're stuck with them. The consensus on the lists so far is that they should be rechristened as "open file description locks". The name isn't a big deal for the kernel, but the command macros are not visually distinct enough from the traditional POSIX lock macros. The glibc and documentation folks are recommending that we change them to look like F_OFD_{GETLK|SETLK|SETLKW}. That lessens the chance that a programmer will typo one of the commands wrong, and also makes it easier to spot this difference when reading code. This patch makes the following changes that I think are necessary before v3.15 ships: 1) rename the command macros to their new names. These end up in the uapi headers and so are part of the external-facing API. It turns out that glibc doesn't actually use the fcntl.h uapi header, but it's hard to be sure that something else won't. Changing it now is safest. 2) make the the /proc/locks output display these as type "OFDLCK" Cc: Michael Kerrisk <mtk.manpages@gmail.com> Cc: Christoph Hellwig <hch@infradead.org> Cc: Carlos O'Donell <carlos@redhat.com> Cc: Stefan Metzmacher <metze@samba.org> Cc: Andy Lutomirski <luto@amacapital.net> Cc: Frank Filz <ffilzlnx@mindspring.com> Cc: Theodore Ts'o <tytso@mit.edu> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-04-22 19:23:58 +07:00
case F_OFD_SETLKW:
locks: require that flock->l_pid be set to 0 for file-private locks Neil Brown suggested potentially overloading the l_pid value as a "lock context" field for file-private locks. While I don't think we will probably want to do that here, it's probably a good idea to ensure that in the future we could extend this API without breaking existing callers. Typically the l_pid value is ignored for incoming struct flock arguments, serving mainly as a place to return the pid of the owner if there is a conflicting lock. For file-private locks, require that it currently be set to 0 and return EINVAL if it isn't. If we eventually want to make a non-zero l_pid mean something, then this will help ensure that we don't break legacy programs that are using file-private locks. Cc: Neil Brown <neilb@suse.de> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-03-04 22:30:23 +07:00
error = -EINVAL;
fs/locks: pass kernel struct flock to fcntl_getlk/setlk This will make it easier to implement a sane compat fcntl syscall. [ jlayton: fix undeclared identifiers in 32-bit fcntl64 syscall handler ] Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-05-27 17:07:19 +07:00
if (flock->l_pid != 0)
locks: require that flock->l_pid be set to 0 for file-private locks Neil Brown suggested potentially overloading the l_pid value as a "lock context" field for file-private locks. While I don't think we will probably want to do that here, it's probably a good idea to ensure that in the future we could extend this API without breaking existing callers. Typically the l_pid value is ignored for incoming struct flock arguments, serving mainly as a place to return the pid of the owner if there is a conflicting lock. For file-private locks, require that it currently be set to 0 and return EINVAL if it isn't. If we eventually want to make a non-zero l_pid mean something, then this will help ensure that we don't break legacy programs that are using file-private locks. Cc: Neil Brown <neilb@suse.de> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-03-04 22:30:23 +07:00
goto out;
locks: add new fcntl cmd values for handling file private locks Due to some unfortunate history, POSIX locks have very strange and unhelpful semantics. The thing that usually catches people by surprise is that they are dropped whenever the process closes any file descriptor associated with the inode. This is extremely problematic for people developing file servers that need to implement byte-range locks. Developers often need a "lock management" facility to ensure that file descriptors are not closed until all of the locks associated with the inode are finished. Additionally, "classic" POSIX locks are owned by the process. Locks taken between threads within the same process won't conflict with one another, which renders them useless for synchronization between threads. This patchset adds a new type of lock that attempts to address these issues. These locks conflict with classic POSIX read/write locks, but have semantics that are more like BSD locks with respect to inheritance and behavior on close. This is implemented primarily by changing how fl_owner field is set for these locks. Instead of having them owned by the files_struct of the process, they are instead owned by the filp on which they were acquired. Thus, they are inherited across fork() and are only released when the last reference to a filp is put. These new semantics prevent them from being merged with classic POSIX locks, even if they are acquired by the same process. These locks will also conflict with classic POSIX locks even if they are acquired by the same process or on the same file descriptor. The new locks are managed using a new set of cmd values to the fcntl() syscall. The initial implementation of this converts these values to "classic" cmd values at a fairly high level, and the details are not exposed to the underlying filesystem. We may eventually want to push this handing out to the lower filesystem code but for now I don't see any need for it. Also, note that with this implementation the new cmd values are only available via fcntl64() on 32-bit arches. There's little need to add support for legacy apps on a new interface like this. Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:10 +07:00
cmd = F_SETLKW64;
locks: rename FL_FILE_PVT and IS_FILE_PVT to use "*_OFDLCK" instead File-private locks have been re-christened as "open file description" locks. Finish the symbol name cleanup in the internal implementation. Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-04-22 19:24:32 +07:00
file_lock->fl_flags |= FL_OFDLCK;
locks: purge fl_owner_t from fs/locks.c Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-07-13 22:00:38 +07:00
file_lock->fl_owner = filp;
locks: add new fcntl cmd values for handling file private locks Due to some unfortunate history, POSIX locks have very strange and unhelpful semantics. The thing that usually catches people by surprise is that they are dropped whenever the process closes any file descriptor associated with the inode. This is extremely problematic for people developing file servers that need to implement byte-range locks. Developers often need a "lock management" facility to ensure that file descriptors are not closed until all of the locks associated with the inode are finished. Additionally, "classic" POSIX locks are owned by the process. Locks taken between threads within the same process won't conflict with one another, which renders them useless for synchronization between threads. This patchset adds a new type of lock that attempts to address these issues. These locks conflict with classic POSIX read/write locks, but have semantics that are more like BSD locks with respect to inheritance and behavior on close. This is implemented primarily by changing how fl_owner field is set for these locks. Instead of having them owned by the files_struct of the process, they are instead owned by the filp on which they were acquired. Thus, they are inherited across fork() and are only released when the last reference to a filp is put. These new semantics prevent them from being merged with classic POSIX locks, even if they are acquired by the same process. These locks will also conflict with classic POSIX locks even if they are acquired by the same process or on the same file descriptor. The new locks are managed using a new set of cmd values to the fcntl() syscall. The initial implementation of this converts these values to "classic" cmd values at a fairly high level, and the details are not exposed to the underlying filesystem. We may eventually want to push this handing out to the lower filesystem code but for now I don't see any need for it. Also, note that with this implementation the new cmd values are only available via fcntl64() on 32-bit arches. There's little need to add support for legacy apps on a new interface like this. Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:10 +07:00
/* Fallthrough */
case F_SETLKW64:
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
file_lock->fl_flags |= FL_SLEEP;
}
locks: add new fcntl cmd values for handling file private locks Due to some unfortunate history, POSIX locks have very strange and unhelpful semantics. The thing that usually catches people by surprise is that they are dropped whenever the process closes any file descriptor associated with the inode. This is extremely problematic for people developing file servers that need to implement byte-range locks. Developers often need a "lock management" facility to ensure that file descriptors are not closed until all of the locks associated with the inode are finished. Additionally, "classic" POSIX locks are owned by the process. Locks taken between threads within the same process won't conflict with one another, which renders them useless for synchronization between threads. This patchset adds a new type of lock that attempts to address these issues. These locks conflict with classic POSIX read/write locks, but have semantics that are more like BSD locks with respect to inheritance and behavior on close. This is implemented primarily by changing how fl_owner field is set for these locks. Instead of having them owned by the files_struct of the process, they are instead owned by the filp on which they were acquired. Thus, they are inherited across fork() and are only released when the last reference to a filp is put. These new semantics prevent them from being merged with classic POSIX locks, even if they are acquired by the same process. These locks will also conflict with classic POSIX locks even if they are acquired by the same process or on the same file descriptor. The new locks are managed using a new set of cmd values to the fcntl() syscall. The initial implementation of this converts these values to "classic" cmd values at a fairly high level, and the details are not exposed to the underlying filesystem. We may eventually want to push this handing out to the lower filesystem code but for now I don't see any need for it. Also, note that with this implementation the new cmd values are only available via fcntl64() on 32-bit arches. There's little need to add support for legacy apps on a new interface like this. Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:10 +07:00
locks: cleanup code duplication Extract common code into a function. Signed-off-by: Miklos Szeredi <mszeredi@suse.cz> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Cc: Matthew Wilcox <matthew@wil.cx> Cc: David Teigland <teigland@redhat.com> Cc: Christoph Hellwig <hch@lst.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-07-25 15:48:57 +07:00
error = do_lock_file_wait(filp, cmd, file_lock);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
[PATCH] stale POSIX lock handling 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>
2005-07-28 01:45:09 +07:00
/*
locks: don't check for race with close when setting OFD lock We don't clean out OFD locks on close(), so there's no need to check for a race with them here. They'll get cleaned out at the same time that flock locks are. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Acked-by: "J. Bruce Fields" <bfields@fieldses.org>
2016-01-08 19:30:43 +07:00
* Attempt to detect a close/fcntl race and recover by releasing the
* lock that was just acquired. There is no need to do that when we're
* unlocking though, or for OFD locks.
[PATCH] stale POSIX lock handling 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>
2005-07-28 01:45:09 +07:00
*/
locks: don't check for race with close when setting OFD lock We don't clean out OFD locks on close(), so there's no need to check for a race with them here. They'll get cleaned out at the same time that flock locks are. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Acked-by: "J. Bruce Fields" <bfields@fieldses.org>
2016-01-08 19:30:43 +07:00
if (!error && file_lock->fl_type != F_UNLCK &&
!(file_lock->fl_flags & FL_OFDLCK)) {
locks: fix unlock when fcntl_setlk races with a close Dmitry reported that he was able to reproduce the WARN_ON_ONCE that fires in locks_free_lock_context when the flc_posix list isn't empty. The problem turns out to be that we're basically rebuilding the file_lock from scratch in fcntl_setlk when we discover that the setlk has raced with a close. If the l_whence field is SEEK_CUR or SEEK_END, then we may end up with fl_start and fl_end values that differ from when the lock was initially set, if the file position or length of the file has changed in the interim. Fix this by just reusing the same lock request structure, and simply override fl_type value with F_UNLCK as appropriate. That ensures that we really are unlocking the lock that was initially set. While we're there, make sure that we do pop a WARN_ON_ONCE if the removal ever fails. Also return -EBADF in this event, since that's what we would have returned if the close had happened earlier. Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: <stable@vger.kernel.org> Fixes: c293621bbf67 (stale POSIX lock handling) Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Acked-by: "J. Bruce Fields" <bfields@fieldses.org>
2016-01-08 04:38:10 +07:00
/*
* We need that spin_lock here - it prevents reordering between
* update of i_flctx->flc_posix and check for it done in
* close(). rcu_read_lock() wouldn't do.
*/
spin_lock(&current->files->file_lock);
f = fcheck(fd);
spin_unlock(&current->files->file_lock);
if (f != filp) {
file_lock->fl_type = F_UNLCK;
error = do_lock_file_wait(filp, cmd, file_lock);
WARN_ON_ONCE(error);
error = -EBADF;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
out:
locks_free_lock(file_lock);
return error;
}
#endif /* BITS_PER_LONG == 32 */
/*
* This function is called when the file is being removed
* from the task's fd array. POSIX locks belonging to this task
* are deleted at this time.
*/
void locks_remove_posix(struct file *filp, fl_owner_t owner)
{
locks: sprinkle some tracepoints around the file locking code Add some tracepoints around the POSIX locking code. These were useful when tracking down problems when handling the race between setlk and close. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Acked-by: "J. Bruce Fields" <bfields@fieldses.org>
2016-01-07 09:26:10 +07:00
int error;
locks: fix file locking on overlayfs This patch allows flock, posix locks, ofd locks and leases to work correctly on overlayfs. Instead of using the underlying inode for storing lock context use the overlay inode. This allows locks to be persistent across copy-up. This is done by introducing locks_inode() helper and using it instead of file_inode() to get the inode in locking code. For non-overlayfs the two are equivalent, except for an extra pointer dereference in locks_inode(). Since lock operations are in "struct file_operations" we must also make sure not to call underlying filesystem's lock operations. Introcude a super block flag MS_NOREMOTELOCK to this effect. Signed-off-by: Miklos Szeredi <mszeredi@redhat.com> Acked-by: Jeff Layton <jlayton@poochiereds.net> Cc: "J. Bruce Fields" <bfields@fieldses.org>
2016-09-16 17:44:20 +07:00
struct inode *inode = locks_inode(filp);
[PATCH] locks: clean up locks_remove_posix() locks_remove_posix() can use posix_lock_file() instead of doing the lock removal by hand. posix_lock_file() now does exacly the same. The comment about pids no longer applies, posix_lock_file() takes only the owner into account. Signed-off-by: Miklos Szeredi <miklos@szeredi.hu> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-06-23 16:05:11 +07:00
struct file_lock lock;
fs: fix data races on inode->i_flctx locks_get_lock_context() uses cmpxchg() to install i_flctx. cmpxchg() is a release operation which is correct. But it uses a plain load to load i_flctx. This is incorrect. Subsequent loads from i_flctx can hoist above the load of i_flctx pointer itself and observe uninitialized garbage there. This in turn can lead to corruption of ctx->flc_lock and other members. Documentation/memory-barriers.txt explicitly requires to use a barrier in such context: "A load-load control dependency requires a full read memory barrier". Use smp_load_acquire() in locks_get_lock_context() and in bunch of other functions that can proceed concurrently with locks_get_lock_context(). The data race was found with KernelThreadSanitizer (KTSAN). Signed-off-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-09-21 14:43:06 +07:00
struct file_lock_context *ctx;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
/*
* If there are no locks held on this file, we don't need to call
* posix_lock_file(). Another process could be setting a lock on this
* file at the same time, but we wouldn't remove that lock anyway.
*/
locks: fix file locking on overlayfs This patch allows flock, posix locks, ofd locks and leases to work correctly on overlayfs. Instead of using the underlying inode for storing lock context use the overlay inode. This allows locks to be persistent across copy-up. This is done by introducing locks_inode() helper and using it instead of file_inode() to get the inode in locking code. For non-overlayfs the two are equivalent, except for an extra pointer dereference in locks_inode(). Since lock operations are in "struct file_operations" we must also make sure not to call underlying filesystem's lock operations. Introcude a super block flag MS_NOREMOTELOCK to this effect. Signed-off-by: Miklos Szeredi <mszeredi@redhat.com> Acked-by: Jeff Layton <jlayton@poochiereds.net> Cc: "J. Bruce Fields" <bfields@fieldses.org>
2016-09-16 17:44:20 +07:00
ctx = smp_load_acquire(&inode->i_flctx);
locks: convert posix locks to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
if (!ctx || list_empty(&ctx->flc_posix))
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
return;
lock.fl_type = F_UNLCK;
[PATCH] vfs: add lock owner argument to flush operation Pass the POSIX lock owner ID to the flush operation. This is useful for filesystems which don't want to store any locking state in inode->i_flock but want to handle locking/unlocking POSIX locks internally. FUSE is one such filesystem but I think it possible that some network filesystems would need this also. Also add a flag to indicate that a POSIX locking request was generated by close(), so filesystems using the above feature won't send an extra locking request in this case. Signed-off-by: Miklos Szeredi <miklos@szeredi.hu> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Cc: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-06-23 16:05:12 +07:00
lock.fl_flags = FL_POSIX | FL_CLOSE;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
lock.fl_start = 0;
lock.fl_end = OFFSET_MAX;
lock.fl_owner = owner;
lock.fl_pid = current->tgid;
lock.fl_file = filp;
lock.fl_ops = NULL;
lock.fl_lmops = NULL;
locks: sprinkle some tracepoints around the file locking code Add some tracepoints around the POSIX locking code. These were useful when tracking down problems when handling the race between setlk and close. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Acked-by: "J. Bruce Fields" <bfields@fieldses.org>
2016-01-07 09:26:10 +07:00
error = vfs_lock_file(filp, F_SETLK, &lock, NULL);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
if (lock.fl_ops && lock.fl_ops->fl_release_private)
lock.fl_ops->fl_release_private(&lock);
locks: fix file locking on overlayfs This patch allows flock, posix locks, ofd locks and leases to work correctly on overlayfs. Instead of using the underlying inode for storing lock context use the overlay inode. This allows locks to be persistent across copy-up. This is done by introducing locks_inode() helper and using it instead of file_inode() to get the inode in locking code. For non-overlayfs the two are equivalent, except for an extra pointer dereference in locks_inode(). Since lock operations are in "struct file_operations" we must also make sure not to call underlying filesystem's lock operations. Introcude a super block flag MS_NOREMOTELOCK to this effect. Signed-off-by: Miklos Szeredi <mszeredi@redhat.com> Acked-by: Jeff Layton <jlayton@poochiereds.net> Cc: "J. Bruce Fields" <bfields@fieldses.org>
2016-09-16 17:44:20 +07:00
trace_locks_remove_posix(inode, &lock, error);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
EXPORT_SYMBOL(locks_remove_posix);
locks: consolidate NULL i_flctx checks in locks_remove_file We have each of the locks_remove_* variants doing this individually. Have the caller do it instead, and have locks_remove_flock and locks_remove_lease just assume that it's a valid pointer. Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2015-01-17 03:05:58 +07:00
/* The i_flctx must be valid when calling into here */
locks: have locks_release_file use flock_lock_file to release generic flock locks ...instead of open-coding it and removing flock locks directly. This helps consolidate the flock lock removal logic into a single spot. Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2015-01-17 03:05:54 +07:00
static void
fs: fix data races on inode->i_flctx locks_get_lock_context() uses cmpxchg() to install i_flctx. cmpxchg() is a release operation which is correct. But it uses a plain load to load i_flctx. This is incorrect. Subsequent loads from i_flctx can hoist above the load of i_flctx pointer itself and observe uninitialized garbage there. This in turn can lead to corruption of ctx->flc_lock and other members. Documentation/memory-barriers.txt explicitly requires to use a barrier in such context: "A load-load control dependency requires a full read memory barrier". Use smp_load_acquire() in locks_get_lock_context() and in bunch of other functions that can proceed concurrently with locks_get_lock_context(). The data race was found with KernelThreadSanitizer (KTSAN). Signed-off-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-09-21 14:43:06 +07:00
locks_remove_flock(struct file *filp, struct file_lock_context *flctx)
locks: have locks_release_file use flock_lock_file to release generic flock locks ...instead of open-coding it and removing flock locks directly. This helps consolidate the flock lock removal logic into a single spot. Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2015-01-17 03:05:54 +07:00
{
struct file_lock fl = {
.fl_owner = filp,
.fl_pid = current->tgid,
.fl_file = filp,
locks: Set FL_CLOSE when removing flock locks on close() Set FL_CLOSE in fl_flags as in locks_remove_posix() when clearing locks. NFS will check for this flag to ensure an unlock is sent in a following patch. Fuse handles flock and posix locks differently for FL_CLOSE, and so requires a fixup to retain the existing behavior for flock. Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Reviewed-by: Jeff Layton <jlayton@redhat.com> Acked-by: Miklos Szeredi <miklos@szeredi.hu> Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
2017-04-11 23:50:09 +07:00
.fl_flags = FL_FLOCK | FL_CLOSE,
locks: have locks_release_file use flock_lock_file to release generic flock locks ...instead of open-coding it and removing flock locks directly. This helps consolidate the flock lock removal logic into a single spot. Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2015-01-17 03:05:54 +07:00
.fl_type = F_UNLCK,
.fl_end = OFFSET_MAX,
};
locks: fix file locking on overlayfs This patch allows flock, posix locks, ofd locks and leases to work correctly on overlayfs. Instead of using the underlying inode for storing lock context use the overlay inode. This allows locks to be persistent across copy-up. This is done by introducing locks_inode() helper and using it instead of file_inode() to get the inode in locking code. For non-overlayfs the two are equivalent, except for an extra pointer dereference in locks_inode(). Since lock operations are in "struct file_operations" we must also make sure not to call underlying filesystem's lock operations. Introcude a super block flag MS_NOREMOTELOCK to this effect. Signed-off-by: Miklos Szeredi <mszeredi@redhat.com> Acked-by: Jeff Layton <jlayton@poochiereds.net> Cc: "J. Bruce Fields" <bfields@fieldses.org>
2016-09-16 17:44:20 +07:00
struct inode *inode = locks_inode(filp);
locks: have locks_release_file use flock_lock_file to release generic flock locks ...instead of open-coding it and removing flock locks directly. This helps consolidate the flock lock removal logic into a single spot. Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2015-01-17 03:05:54 +07:00
locks: consolidate NULL i_flctx checks in locks_remove_file We have each of the locks_remove_* variants doing this individually. Have the caller do it instead, and have locks_remove_flock and locks_remove_lease just assume that it's a valid pointer. Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2015-01-17 03:05:58 +07:00
if (list_empty(&flctx->flc_flock))
locks: have locks_release_file use flock_lock_file to release generic flock locks ...instead of open-coding it and removing flock locks directly. This helps consolidate the flock lock removal logic into a single spot. Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2015-01-17 03:05:54 +07:00
return;
Partially revert "locks: fix file locking on overlayfs" This partially reverts commit c568d68341be7030f5647def68851e469b21ca11. Overlayfs files will now automatically get the correct locks, no need to hack overlay support in VFS. It is a partial revert, because it leaves the locks_inode() calls in place and defines locks_inode() to file_inode(). We could revert those as well, but it would be unnecessary code churn and it makes sense to document that we are getting the inode for locking purposes. Don't revert MS_NOREMOTELOCK yet since that has been part of the userspace API for some time (though not in a useful way). Will try to remove internal flags later when the dust around the new mount API settles. Signed-off-by: Miklos Szeredi <mszeredi@redhat.com> Acked-by: Jeff Layton <jlayton@kernel.org>
2018-07-18 20:44:43 +07:00
if (filp->f_op->flock)
locks: have locks_release_file use flock_lock_file to release generic flock locks ...instead of open-coding it and removing flock locks directly. This helps consolidate the flock lock removal logic into a single spot. Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2015-01-17 03:05:54 +07:00
filp->f_op->flock(filp, F_SETLKW, &fl);
else
locks: have flock_lock_file take an inode pointer instead of a filp ...and rename it to better describe how it works. In order to fix a use-after-free in NFS, we need to be able to remove locks from an inode after the filp associated with them may have already been freed. flock_lock_file already only dereferences the filp to get to the inode, so just change it so the callers do that. All of the callers already pass in a lock request that has the fl_file set properly, so we don't need to pass it in individually. With that change it now only dereferences the filp to get to the inode, so just push that out to the callers. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Reviewed-by: "J. Bruce Fields" <bfields@fieldses.org> Tested-by: "J. Bruce Fields" <bfields@fieldses.org>
2015-07-11 17:43:02 +07:00
flock_lock_inode(inode, &fl);
locks: have locks_release_file use flock_lock_file to release generic flock locks ...instead of open-coding it and removing flock locks directly. This helps consolidate the flock lock removal logic into a single spot. Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2015-01-17 03:05:54 +07:00
if (fl.fl_ops && fl.fl_ops->fl_release_private)
fl.fl_ops->fl_release_private(&fl);
}
locks: consolidate NULL i_flctx checks in locks_remove_file We have each of the locks_remove_* variants doing this individually. Have the caller do it instead, and have locks_remove_flock and locks_remove_lease just assume that it's a valid pointer. Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2015-01-17 03:05:58 +07:00
/* The i_flctx must be valid when calling into here */
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
static void
fs: fix data races on inode->i_flctx locks_get_lock_context() uses cmpxchg() to install i_flctx. cmpxchg() is a release operation which is correct. But it uses a plain load to load i_flctx. This is incorrect. Subsequent loads from i_flctx can hoist above the load of i_flctx pointer itself and observe uninitialized garbage there. This in turn can lead to corruption of ctx->flc_lock and other members. Documentation/memory-barriers.txt explicitly requires to use a barrier in such context: "A load-load control dependency requires a full read memory barrier". Use smp_load_acquire() in locks_get_lock_context() and in bunch of other functions that can proceed concurrently with locks_get_lock_context(). The data race was found with KernelThreadSanitizer (KTSAN). Signed-off-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-09-21 14:43:06 +07:00
locks_remove_lease(struct file *filp, struct file_lock_context *ctx)
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
{
struct file_lock *fl, *tmp;
LIST_HEAD(dispose);
locks: consolidate NULL i_flctx checks in locks_remove_file We have each of the locks_remove_* variants doing this individually. Have the caller do it instead, and have locks_remove_flock and locks_remove_lease just assume that it's a valid pointer. Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2015-01-17 03:05:58 +07:00
if (list_empty(&ctx->flc_lease))
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
return;
locking, fs/locks: Add missing file_sem locks I overlooked a few code-paths that can lead to locks_delete_global_locks(). Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Reviewed-by: Jeff Layton <jlayton@poochiereds.net> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Bruce Fields <bfields@fieldses.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: linux-fsdevel@vger.kernel.org Cc: syzkaller <syzkaller@googlegroups.com> Link: http://lkml.kernel.org/r/20161008081228.GF3142@twins.programming.kicks-ass.net Signed-off-by: Ingo Molnar <mingo@kernel.org>
2016-10-08 15:12:28 +07:00
percpu_down_read_preempt_disable(&file_rwsem);
locks: add a dedicated spinlock to protect i_flctx lists We can now add a dedicated spinlock without expanding struct inode. Change to using that to protect the various i_flctx lists. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
spin_lock(&ctx->flc_lock);
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
list_for_each_entry_safe(fl, tmp, &ctx->flc_lease, fl_list)
locks: only remove leases associated with the file being closed We don't want to remove all leases just because one filp was closed. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-02-17 07:37:42 +07:00
if (filp == fl->fl_file)
lease_modify(fl, F_UNLCK, &dispose);
locks: add a dedicated spinlock to protect i_flctx lists We can now add a dedicated spinlock without expanding struct inode. Change to using that to protect the various i_flctx lists. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:57 +07:00
spin_unlock(&ctx->flc_lock);
locking, fs/locks: Add missing file_sem locks I overlooked a few code-paths that can lead to locks_delete_global_locks(). Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Reviewed-by: Jeff Layton <jlayton@poochiereds.net> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Bruce Fields <bfields@fieldses.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: linux-fsdevel@vger.kernel.org Cc: syzkaller <syzkaller@googlegroups.com> Link: http://lkml.kernel.org/r/20161008081228.GF3142@twins.programming.kicks-ass.net Signed-off-by: Ingo Molnar <mingo@kernel.org>
2016-10-08 15:12:28 +07:00
percpu_up_read_preempt_enable(&file_rwsem);
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
locks_dispose_list(&dispose);
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
/*
* This function is called on the last close of an open file.
*/
locks: rename locks_remove_flock to locks_remove_file 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>
2014-02-04 00:13:08 +07:00
void locks_remove_file(struct file *filp)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
fs: fix data races on inode->i_flctx locks_get_lock_context() uses cmpxchg() to install i_flctx. cmpxchg() is a release operation which is correct. But it uses a plain load to load i_flctx. This is incorrect. Subsequent loads from i_flctx can hoist above the load of i_flctx pointer itself and observe uninitialized garbage there. This in turn can lead to corruption of ctx->flc_lock and other members. Documentation/memory-barriers.txt explicitly requires to use a barrier in such context: "A load-load control dependency requires a full read memory barrier". Use smp_load_acquire() in locks_get_lock_context() and in bunch of other functions that can proceed concurrently with locks_get_lock_context(). The data race was found with KernelThreadSanitizer (KTSAN). Signed-off-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-09-21 14:43:06 +07:00
struct file_lock_context *ctx;
locks: fix file locking on overlayfs This patch allows flock, posix locks, ofd locks and leases to work correctly on overlayfs. Instead of using the underlying inode for storing lock context use the overlay inode. This allows locks to be persistent across copy-up. This is done by introducing locks_inode() helper and using it instead of file_inode() to get the inode in locking code. For non-overlayfs the two are equivalent, except for an extra pointer dereference in locks_inode(). Since lock operations are in "struct file_operations" we must also make sure not to call underlying filesystem's lock operations. Introcude a super block flag MS_NOREMOTELOCK to this effect. Signed-off-by: Miklos Szeredi <mszeredi@redhat.com> Acked-by: Jeff Layton <jlayton@poochiereds.net> Cc: "J. Bruce Fields" <bfields@fieldses.org>
2016-09-16 17:44:20 +07:00
ctx = smp_load_acquire(&locks_inode(filp)->i_flctx);
fs: fix data races on inode->i_flctx locks_get_lock_context() uses cmpxchg() to install i_flctx. cmpxchg() is a release operation which is correct. But it uses a plain load to load i_flctx. This is incorrect. Subsequent loads from i_flctx can hoist above the load of i_flctx pointer itself and observe uninitialized garbage there. This in turn can lead to corruption of ctx->flc_lock and other members. Documentation/memory-barriers.txt explicitly requires to use a barrier in such context: "A load-load control dependency requires a full read memory barrier". Use smp_load_acquire() in locks_get_lock_context() and in bunch of other functions that can proceed concurrently with locks_get_lock_context(). The data race was found with KernelThreadSanitizer (KTSAN). Signed-off-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-09-21 14:43:06 +07:00
if (!ctx)
locks: consolidate NULL i_flctx checks in locks_remove_file We have each of the locks_remove_* variants doing this individually. Have the caller do it instead, and have locks_remove_flock and locks_remove_lease just assume that it's a valid pointer. Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2015-01-17 03:05:58 +07:00
return;
locks: have locks_release_file use flock_lock_file to release generic flock locks ...instead of open-coding it and removing flock locks directly. This helps consolidate the flock lock removal logic into a single spot. Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2015-01-17 03:05:54 +07:00
/* remove any OFD locks */
locks: purge fl_owner_t from fs/locks.c Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-07-13 22:00:38 +07:00
locks_remove_posix(filp, filp);
locks: add new fcntl cmd values for handling file private locks Due to some unfortunate history, POSIX locks have very strange and unhelpful semantics. The thing that usually catches people by surprise is that they are dropped whenever the process closes any file descriptor associated with the inode. This is extremely problematic for people developing file servers that need to implement byte-range locks. Developers often need a "lock management" facility to ensure that file descriptors are not closed until all of the locks associated with the inode are finished. Additionally, "classic" POSIX locks are owned by the process. Locks taken between threads within the same process won't conflict with one another, which renders them useless for synchronization between threads. This patchset adds a new type of lock that attempts to address these issues. These locks conflict with classic POSIX read/write locks, but have semantics that are more like BSD locks with respect to inheritance and behavior on close. This is implemented primarily by changing how fl_owner field is set for these locks. Instead of having them owned by the files_struct of the process, they are instead owned by the filp on which they were acquired. Thus, they are inherited across fork() and are only released when the last reference to a filp is put. These new semantics prevent them from being merged with classic POSIX locks, even if they are acquired by the same process. These locks will also conflict with classic POSIX locks even if they are acquired by the same process or on the same file descriptor. The new locks are managed using a new set of cmd values to the fcntl() syscall. The initial implementation of this converts these values to "classic" cmd values at a fairly high level, and the details are not exposed to the underlying filesystem. We may eventually want to push this handing out to the lower filesystem code but for now I don't see any need for it. Also, note that with this implementation the new cmd values are only available via fcntl64() on 32-bit arches. There's little need to add support for legacy apps on a new interface like this. Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:10 +07:00
locks: have locks_release_file use flock_lock_file to release generic flock locks ...instead of open-coding it and removing flock locks directly. This helps consolidate the flock lock removal logic into a single spot. Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2015-01-17 03:05:54 +07:00
/* remove flock locks */
fs: fix data races on inode->i_flctx locks_get_lock_context() uses cmpxchg() to install i_flctx. cmpxchg() is a release operation which is correct. But it uses a plain load to load i_flctx. This is incorrect. Subsequent loads from i_flctx can hoist above the load of i_flctx pointer itself and observe uninitialized garbage there. This in turn can lead to corruption of ctx->flc_lock and other members. Documentation/memory-barriers.txt explicitly requires to use a barrier in such context: "A load-load control dependency requires a full read memory barrier". Use smp_load_acquire() in locks_get_lock_context() and in bunch of other functions that can proceed concurrently with locks_get_lock_context(). The data race was found with KernelThreadSanitizer (KTSAN). Signed-off-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-09-21 14:43:06 +07:00
locks_remove_flock(filp, ctx);
locks: have locks_release_file use flock_lock_file to release generic flock locks ...instead of open-coding it and removing flock locks directly. This helps consolidate the flock lock removal logic into a single spot. Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2015-01-17 03:05:54 +07:00
locks: convert lease handling to file_lock_context Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:55 +07:00
/* remove any leases */
fs: fix data races on inode->i_flctx locks_get_lock_context() uses cmpxchg() to install i_flctx. cmpxchg() is a release operation which is correct. But it uses a plain load to load i_flctx. This is incorrect. Subsequent loads from i_flctx can hoist above the load of i_flctx pointer itself and observe uninitialized garbage there. This in turn can lead to corruption of ctx->flc_lock and other members. Documentation/memory-barriers.txt explicitly requires to use a barrier in such context: "A load-load control dependency requires a full read memory barrier". Use smp_load_acquire() in locks_get_lock_context() and in bunch of other functions that can proceed concurrently with locks_get_lock_context(). The data race was found with KernelThreadSanitizer (KTSAN). Signed-off-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-09-21 14:43:06 +07:00
locks_remove_lease(filp, ctx);
locks: restore a warn for leaked locks on close When locks.c moved to using file_lock_context, the check for any locks that were not released was moved from the __fput() to destroy_inode() path in commit 8634b51f6ca2 ("locks: convert lease handling to file_lock_context"). This warning has been quite useful for catching bugs, particularly in NFS where lock handling still sees some churn. Let's bring back the warning for leaked locks on __fput, as this warning is much more likely to be seen and reported by users. Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-07-22 00:36:25 +07:00
spin_lock(&ctx->flc_lock);
locks_check_ctx_file_list(filp, &ctx->flc_posix, "POSIX");
locks_check_ctx_file_list(filp, &ctx->flc_flock, "FLOCK");
locks_check_ctx_file_list(filp, &ctx->flc_lease, "LEASE");
spin_unlock(&ctx->flc_lock);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
/**
* posix_unblock_lock - stop waiting for a file lock
* @waiter: the lock which was waiting
*
* lockd needs to block waiting for locks.
*/
NLM: Further cancel fixes 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>
2006-01-03 15:55:46 +07:00
int
locks: drop the unused filp argument to posix_unblock_lock Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:09 +07:00
posix_unblock_lock(struct file_lock *waiter)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
NLM: Further cancel fixes 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>
2006-01-03 15:55:46 +07:00
int status = 0;
locks: give the blocked_hash its own spinlock There's no reason we have to protect the blocked_hash and file_lock_list with the same spinlock. With the tests I have, breaking it in two gives a barely measurable performance benefit, but it seems reasonable to make this locking as granular as possible. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:20 +07:00
spin_lock(&blocked_lock_lock);
fs/locks: rename some lists and pointers. struct file lock contains an 'fl_next' pointer which is used to point to the lock that this request is blocked waiting for. So rename it to fl_blocker. The fl_blocked list_head in an active lock is the head of a list of blocked requests. In a request it is a node in that list. These are two distinct uses, so replace with two list_heads with different names. fl_blocked_requests is the head of a list of blocked requests fl_blocked_member is a node in a member of that list. The two different list_heads are never used at the same time, but that will change in a future patch. Note that a tracepoint is changed to report fl_blocker instead of fl_next. Signed-off-by: NeilBrown <neilb@suse.com> Reviewed-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Jeff Layton <jlayton@kernel.org>
2018-11-30 06:04:08 +07:00
if (waiter->fl_blocker)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
__locks_delete_block(waiter);
NLM: Further cancel fixes 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>
2006-01-03 15:55:46 +07:00
else
status = -ENOENT;
locks: give the blocked_hash its own spinlock There's no reason we have to protect the blocked_hash and file_lock_list with the same spinlock. With the tests I have, breaking it in two gives a barely measurable performance benefit, but it seems reasonable to make this locking as granular as possible. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:20 +07:00
spin_unlock(&blocked_lock_lock);
NLM: Further cancel fixes 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>
2006-01-03 15:55:46 +07:00
return status;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
EXPORT_SYMBOL(posix_unblock_lock);
locks: add lock cancel command Lock managers need to be able to cancel pending lock requests. In the case where the exported filesystem manages its own locks, it's not sufficient just to call posix_unblock_lock(); we need to let the filesystem know what's happening too. We do this by adding a new fcntl lock command: FL_CANCELLK. Some day this might also be made available to userspace applications that could benefit from an asynchronous locking api. Signed-off-by: Marc Eshel <eshel@almaden.ibm.com> Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu>
2007-01-19 05:52:58 +07:00
/**
* vfs_cancel_lock - file byte range unblock lock
* @filp: The file to apply the unblock to
* @fl: The lock to be unblocked
*
* Used by lock managers to cancel blocked requests
*/
int vfs_cancel_lock(struct file *filp, struct file_lock *fl)
{
Partially revert "locks: fix file locking on overlayfs" This partially reverts commit c568d68341be7030f5647def68851e469b21ca11. Overlayfs files will now automatically get the correct locks, no need to hack overlay support in VFS. It is a partial revert, because it leaves the locks_inode() calls in place and defines locks_inode() to file_inode(). We could revert those as well, but it would be unnecessary code churn and it makes sense to document that we are getting the inode for locking purposes. Don't revert MS_NOREMOTELOCK yet since that has been part of the userspace API for some time (though not in a useful way). Will try to remove internal flags later when the dust around the new mount API settles. Signed-off-by: Miklos Szeredi <mszeredi@redhat.com> Acked-by: Jeff Layton <jlayton@kernel.org>
2018-07-18 20:44:43 +07:00
if (filp->f_op->lock)
locks: add lock cancel command Lock managers need to be able to cancel pending lock requests. In the case where the exported filesystem manages its own locks, it's not sufficient just to call posix_unblock_lock(); we need to let the filesystem know what's happening too. We do this by adding a new fcntl lock command: FL_CANCELLK. Some day this might also be made available to userspace applications that could benefit from an asynchronous locking api. Signed-off-by: Marc Eshel <eshel@almaden.ibm.com> Signed-off-by: "J. Bruce Fields" <bfields@citi.umich.edu>
2007-01-19 05:52:58 +07:00
return filp->f_op->lock(filp, F_CANCELLK, fl);
return 0;
}
EXPORT_SYMBOL_GPL(vfs_cancel_lock);
Rework /proc/locks via seq_files and seq_list helpers Currently /proc/locks is shown with a proc_read function, but its behavior is rather complex as it has to manually handle current offset and buffer length. On the other hand, files that show objects from lists can be easily reimplemented using the sequential files and the seq_list_XXX() helpers. This saves (as usually) 16 lines of code and more than 200 from the .text section. [akpm@linux-foundation.org: no externs in C] [akpm@linux-foundation.org: warning fixes] Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2007-10-02 04:41:15 +07:00
#ifdef CONFIG_PROC_FS
proc: move rest of /proc/locks to fs/locks.c Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
2008-10-05 01:34:18 +07:00
#include <linux/proc_fs.h>
Rework /proc/locks via seq_files and seq_list helpers Currently /proc/locks is shown with a proc_read function, but its behavior is rather complex as it has to manually handle current offset and buffer length. On the other hand, files that show objects from lists can be easily reimplemented using the sequential files and the seq_list_XXX() helpers. This saves (as usually) 16 lines of code and more than 200 from the .text section. [akpm@linux-foundation.org: no externs in C] [akpm@linux-foundation.org: warning fixes] Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2007-10-02 04:41:15 +07:00
#include <linux/seq_file.h>
locks: move file_lock_list to a set of percpu hlist_heads and convert file_lock_lock to an lglock The file_lock_list is only used for /proc/locks. The vastly common case is for locks to be put onto the list and come off again, without ever being traversed. Help optimize for this use-case by moving to percpu hlist_head-s. At the same time, we can make the locking less contentious by moving to an lglock. When iterating over the lists for /proc/locks, we must take the global lock and then iterate over each CPU's list in turn. This change necessitates a new fl_link_cpu field to keep track of which CPU the entry is on. On x86_64 at least, this field is placed within an existing hole in the struct to avoid growing the size. Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:22 +07:00
struct locks_iterator {
int li_cpu;
loff_t li_pos;
};
Rework /proc/locks via seq_files and seq_list helpers Currently /proc/locks is shown with a proc_read function, but its behavior is rather complex as it has to manually handle current offset and buffer length. On the other hand, files that show objects from lists can be easily reimplemented using the sequential files and the seq_list_XXX() helpers. This saves (as usually) 16 lines of code and more than 200 from the .text section. [akpm@linux-foundation.org: no externs in C] [akpm@linux-foundation.org: warning fixes] Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2007-10-02 04:41:15 +07:00
static void lock_get_status(struct seq_file *f, struct file_lock *fl,
procfs: fix numbering in /proc/locks The lock number in /proc/locks (first field) is implemented by a counter (private field of struct seq_file) which is incremented at each call of locks_show() and reset to 1 in locks_start() whatever the offset is. It should be reset according to the actual position in the list. Because of this, the numbering erratically restarts at 1 several times when reading a long /proc/locks file. Moreover, locks_show() can be called twice to print a single line thus skipping a number. The counter should be incremented in locks_next(). And last, pos is a loff_t, which can be bigger than a pointer, so we don't use the pointer as an integer anymore, and allocate a loff_t instead. Signed-off-by: Jerome Marchand <jmarchan@redhat.com> Cc: Pavel Emelyanov <xemul@parallels.com> Cc: Matthew Wilcox <matthew@wil.cx> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-10-27 04:22:33 +07:00
loff_t id, char *pfx)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
struct inode *inode = NULL;
pid-namespaces-vs-locks-interaction fcntl(F_GETLK,..) can return pid of process for not current pid namespace (if process is belonged to the several namespaces). It is true also for pids in /proc/locks. So correct behavior is saving pointer to the struct pid of the process lock owner. Signed-off-by: Vitaliy Gusev <vgusev@openvz.org> Acked-by: Serge Hallyn <serue@us.ibm.com> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-01-17 07:07:08 +07:00
unsigned int fl_pid;
fs/locks: Remove fl_nspid and use fs-specific l_pid for remote locks Since commit c69899a17ca4 "NFSv4: Update of VFS byte range lock must be atomic with the stateid update", NFSv4 has been inserting locks in rpciod worker context. The result is that the file_lock's fl_nspid is the kworker's pid instead of the original userspace pid. The fl_nspid is only used to represent the namespaced virtual pid number when displaying locks or returning from F_GETLK. There's no reason to set it for every inserted lock, since we can usually just look it up from fl_pid. So, instead of looking up and holding struct pid for every lock, let's just look up the virtual pid number from fl_pid when it is needed. That means we can remove fl_nspid entirely. The translaton and presentation of fl_pid should handle the following four cases: 1 - F_GETLK on a remote file with a remote lock: In this case, the filesystem should determine the l_pid to return here. Filesystems should indicate that the fl_pid represents a non-local pid value that should not be translated by returning an fl_pid <= 0. 2 - F_GETLK on a local file with a remote lock: This should be the l_pid of the lock manager process, and translated. 3 - F_GETLK on a remote file with a local lock, and 4 - F_GETLK on a local file with a local lock: These should be the translated l_pid of the local locking process. Fuse was already doing the correct thing by translating the pid into the caller's namespace. With this change we must update fuse to translate to init's pid namespace, so that the locks API can then translate from init's pid namespace into the pid namespace of the caller. With this change, the locks API will expect that if a filesystem returns a remote pid as opposed to a local pid for F_GETLK, that remote pid will be <= 0. This signifies that the pid is remote, and the locks API will forego translating that pid into the pid namespace of the local calling process. Finally, we convert remote filesystems to present remote pids using negative numbers. Have lustre, 9p, ceph, cifs, and dlm negate the remote pid returned for F_GETLK lock requests. Since local pids will never be larger than PID_MAX_LIMIT (which is currently defined as <= 4 million), but pid_t is an unsigned int, we should have plenty of room to represent remote pids with negative numbers if we assume that remote pid numbers are similarly limited. If this is not the case, then we run the risk of having a remote pid returned for which there is also a corresponding local pid. This is a problem we have now, but this patch should reduce the chances of that occurring, while also returning those remote pid numbers, for whatever that may be worth. Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-07-16 21:28:22 +07:00
struct pid_namespace *proc_pidns = file_inode(f->file)->i_sb->s_fs_info;
pid-namespaces-vs-locks-interaction fcntl(F_GETLK,..) can return pid of process for not current pid namespace (if process is belonged to the several namespaces). It is true also for pids in /proc/locks. So correct behavior is saving pointer to the struct pid of the process lock owner. Signed-off-by: Vitaliy Gusev <vgusev@openvz.org> Acked-by: Serge Hallyn <serue@us.ibm.com> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-01-17 07:07:08 +07:00
fs/locks: Remove fl_nspid and use fs-specific l_pid for remote locks Since commit c69899a17ca4 "NFSv4: Update of VFS byte range lock must be atomic with the stateid update", NFSv4 has been inserting locks in rpciod worker context. The result is that the file_lock's fl_nspid is the kworker's pid instead of the original userspace pid. The fl_nspid is only used to represent the namespaced virtual pid number when displaying locks or returning from F_GETLK. There's no reason to set it for every inserted lock, since we can usually just look it up from fl_pid. So, instead of looking up and holding struct pid for every lock, let's just look up the virtual pid number from fl_pid when it is needed. That means we can remove fl_nspid entirely. The translaton and presentation of fl_pid should handle the following four cases: 1 - F_GETLK on a remote file with a remote lock: In this case, the filesystem should determine the l_pid to return here. Filesystems should indicate that the fl_pid represents a non-local pid value that should not be translated by returning an fl_pid <= 0. 2 - F_GETLK on a local file with a remote lock: This should be the l_pid of the lock manager process, and translated. 3 - F_GETLK on a remote file with a local lock, and 4 - F_GETLK on a local file with a local lock: These should be the translated l_pid of the local locking process. Fuse was already doing the correct thing by translating the pid into the caller's namespace. With this change we must update fuse to translate to init's pid namespace, so that the locks API can then translate from init's pid namespace into the pid namespace of the caller. With this change, the locks API will expect that if a filesystem returns a remote pid as opposed to a local pid for F_GETLK, that remote pid will be <= 0. This signifies that the pid is remote, and the locks API will forego translating that pid into the pid namespace of the local calling process. Finally, we convert remote filesystems to present remote pids using negative numbers. Have lustre, 9p, ceph, cifs, and dlm negate the remote pid returned for F_GETLK lock requests. Since local pids will never be larger than PID_MAX_LIMIT (which is currently defined as <= 4 million), but pid_t is an unsigned int, we should have plenty of room to represent remote pids with negative numbers if we assume that remote pid numbers are similarly limited. If this is not the case, then we run the risk of having a remote pid returned for which there is also a corresponding local pid. This is a problem we have now, but this patch should reduce the chances of that occurring, while also returning those remote pid numbers, for whatever that may be worth. Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-07-16 21:28:22 +07:00
fl_pid = locks_translate_pid(fl, proc_pidns);
/*
fs/lock: show locks taken by processes from another pidns Currently if we face a lock taken by a process invisible in the current pidns we skip the lock completely, but this 1) makes the output not that nice (root@vz7)/: cat /proc/${PID_A2}/fdinfo/3 pos: 4 flags: 02100002 mnt_id: 257 lock: (root@vz7)/: 2) makes it more difficult to debug issues with leaked flocks if you get error on lock, but don't see any locks in /proc/$id/fdinfo/$file Let's show information about such locks again as previously, but show zero in the owner pid field. After the patch: =============== (root@vz7)/:cat /proc/${PID_A2}/fdinfo/3 pos: 4 flags: 02100002 mnt_id: 295 lock: 1: FLOCK ADVISORY WRITE 0 b6:f8a61:529946 0 EOF Fixes: 9d5b86ac13c5 ("fs/locks: Remove fl_nspid and use fs-specific l_pid for remote locks") Signed-off-by: Konstantin Khorenko <khorenko@virtuozzo.com> Acked-by: Andrey Vagin <avagin@openvz.org> Reviewed-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2018-06-08 21:27:12 +07:00
* If lock owner is dead (and pid is freed) or not visible in current
* pidns, zero is shown as a pid value. Check lock info from
* init_pid_ns to get saved lock pid value.
fs/locks: Remove fl_nspid and use fs-specific l_pid for remote locks Since commit c69899a17ca4 "NFSv4: Update of VFS byte range lock must be atomic with the stateid update", NFSv4 has been inserting locks in rpciod worker context. The result is that the file_lock's fl_nspid is the kworker's pid instead of the original userspace pid. The fl_nspid is only used to represent the namespaced virtual pid number when displaying locks or returning from F_GETLK. There's no reason to set it for every inserted lock, since we can usually just look it up from fl_pid. So, instead of looking up and holding struct pid for every lock, let's just look up the virtual pid number from fl_pid when it is needed. That means we can remove fl_nspid entirely. The translaton and presentation of fl_pid should handle the following four cases: 1 - F_GETLK on a remote file with a remote lock: In this case, the filesystem should determine the l_pid to return here. Filesystems should indicate that the fl_pid represents a non-local pid value that should not be translated by returning an fl_pid <= 0. 2 - F_GETLK on a local file with a remote lock: This should be the l_pid of the lock manager process, and translated. 3 - F_GETLK on a remote file with a local lock, and 4 - F_GETLK on a local file with a local lock: These should be the translated l_pid of the local locking process. Fuse was already doing the correct thing by translating the pid into the caller's namespace. With this change we must update fuse to translate to init's pid namespace, so that the locks API can then translate from init's pid namespace into the pid namespace of the caller. With this change, the locks API will expect that if a filesystem returns a remote pid as opposed to a local pid for F_GETLK, that remote pid will be <= 0. This signifies that the pid is remote, and the locks API will forego translating that pid into the pid namespace of the local calling process. Finally, we convert remote filesystems to present remote pids using negative numbers. Have lustre, 9p, ceph, cifs, and dlm negate the remote pid returned for F_GETLK lock requests. Since local pids will never be larger than PID_MAX_LIMIT (which is currently defined as <= 4 million), but pid_t is an unsigned int, we should have plenty of room to represent remote pids with negative numbers if we assume that remote pid numbers are similarly limited. If this is not the case, then we run the risk of having a remote pid returned for which there is also a corresponding local pid. This is a problem we have now, but this patch should reduce the chances of that occurring, while also returning those remote pid numbers, for whatever that may be worth. Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-07-16 21:28:22 +07:00
*/
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
if (fl->fl_file != NULL)
locks: fix file locking on overlayfs This patch allows flock, posix locks, ofd locks and leases to work correctly on overlayfs. Instead of using the underlying inode for storing lock context use the overlay inode. This allows locks to be persistent across copy-up. This is done by introducing locks_inode() helper and using it instead of file_inode() to get the inode in locking code. For non-overlayfs the two are equivalent, except for an extra pointer dereference in locks_inode(). Since lock operations are in "struct file_operations" we must also make sure not to call underlying filesystem's lock operations. Introcude a super block flag MS_NOREMOTELOCK to this effect. Signed-off-by: Miklos Szeredi <mszeredi@redhat.com> Acked-by: Jeff Layton <jlayton@poochiereds.net> Cc: "J. Bruce Fields" <bfields@fieldses.org>
2016-09-16 17:44:20 +07:00
inode = locks_inode(fl->fl_file);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
procfs: fix numbering in /proc/locks The lock number in /proc/locks (first field) is implemented by a counter (private field of struct seq_file) which is incremented at each call of locks_show() and reset to 1 in locks_start() whatever the offset is. It should be reset according to the actual position in the list. Because of this, the numbering erratically restarts at 1 several times when reading a long /proc/locks file. Moreover, locks_show() can be called twice to print a single line thus skipping a number. The counter should be incremented in locks_next(). And last, pos is a loff_t, which can be bigger than a pointer, so we don't use the pointer as an integer anymore, and allocate a loff_t instead. Signed-off-by: Jerome Marchand <jmarchan@redhat.com> Cc: Pavel Emelyanov <xemul@parallels.com> Cc: Matthew Wilcox <matthew@wil.cx> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-10-27 04:22:33 +07:00
seq_printf(f, "%lld:%s ", id, pfx);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
if (IS_POSIX(fl)) {
locks: make /proc/locks show IS_FILE_PVT locks as type "FLPVT" In a later patch, we'll be adding a new type of lock that's owned by the struct file instead of the files_struct. Those sorts of locks will be flagged with a new FL_FILE_PVT flag. Report these types of locks as "FLPVT" in /proc/locks to distinguish them from "classic" POSIX locks. Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:09 +07:00
if (fl->fl_flags & FL_ACCESS)
fs/locks.c: replace seq_printf by seq_puts Replace seq_printf where possible Cc: Jeff Layton <jlayton@redhat.com> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Fabian Frederick <fabf@skynet.be> Signed-off-by: Jeff Layton <jlayton@poochiereds.net>
2014-05-10 01:13:05 +07:00
seq_puts(f, "ACCESS");
locks: rename FL_FILE_PVT and IS_FILE_PVT to use "*_OFDLCK" instead File-private locks have been re-christened as "open file description" locks. Finish the symbol name cleanup in the internal implementation. Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-04-22 19:24:32 +07:00
else if (IS_OFDLCK(fl))
fs/locks.c: replace seq_printf by seq_puts Replace seq_printf where possible Cc: Jeff Layton <jlayton@redhat.com> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Fabian Frederick <fabf@skynet.be> Signed-off-by: Jeff Layton <jlayton@poochiereds.net>
2014-05-10 01:13:05 +07:00
seq_puts(f, "OFDLCK");
locks: make /proc/locks show IS_FILE_PVT locks as type "FLPVT" In a later patch, we'll be adding a new type of lock that's owned by the struct file instead of the files_struct. Those sorts of locks will be flagged with a new FL_FILE_PVT flag. Report these types of locks as "FLPVT" in /proc/locks to distinguish them from "classic" POSIX locks. Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:09 +07:00
else
fs/locks.c: replace seq_printf by seq_puts Replace seq_printf where possible Cc: Jeff Layton <jlayton@redhat.com> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Fabian Frederick <fabf@skynet.be> Signed-off-by: Jeff Layton <jlayton@poochiereds.net>
2014-05-10 01:13:05 +07:00
seq_puts(f, "POSIX ");
locks: make /proc/locks show IS_FILE_PVT locks as type "FLPVT" In a later patch, we'll be adding a new type of lock that's owned by the struct file instead of the files_struct. Those sorts of locks will be flagged with a new FL_FILE_PVT flag. Report these types of locks as "FLPVT" in /proc/locks to distinguish them from "classic" POSIX locks. Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:09 +07:00
seq_printf(f, " %s ",
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
(inode == NULL) ? "*NOINODE*" :
Cleanup macros for distinguishing mandatory locks The combination of S_ISGID bit set and S_IXGRP bit unset is used to mark the inode as "mandatory lockable" and there's a macro for this check called MANDATORY_LOCK(inode). However, fs/locks.c and some filesystems still perform the explicit i_mode checking. Besides, Andrew pointed out, that this macro is buggy itself, as it dereferences the inode arg twice. Convert this macro into static inline function and switch its users to it, making the code shorter and more readable. The __mandatory_lock() helper is to be used in places where the IS_MANDLOCK() for superblock is already known to be true. Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: David Howells <dhowells@redhat.com> Cc: Eric Van Hensbergen <ericvh@gmail.com> Cc: Ron Minnich <rminnich@sandia.gov> Cc: Latchesar Ionkov <lucho@ionkov.net> Cc: Steven Whitehouse <swhiteho@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2007-10-02 04:41:11 +07:00
mandatory_lock(inode) ? "MANDATORY" : "ADVISORY ");
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
} else if (IS_FLOCK(fl)) {
if (fl->fl_type & LOCK_MAND) {
fs/locks.c: replace seq_printf by seq_puts Replace seq_printf where possible Cc: Jeff Layton <jlayton@redhat.com> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Fabian Frederick <fabf@skynet.be> Signed-off-by: Jeff Layton <jlayton@poochiereds.net>
2014-05-10 01:13:05 +07:00
seq_puts(f, "FLOCK MSNFS ");
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
} else {
fs/locks.c: replace seq_printf by seq_puts Replace seq_printf where possible Cc: Jeff Layton <jlayton@redhat.com> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Fabian Frederick <fabf@skynet.be> Signed-off-by: Jeff Layton <jlayton@poochiereds.net>
2014-05-10 01:13:05 +07:00
seq_puts(f, "FLOCK ADVISORY ");
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
} else if (IS_LEASE(fl)) {
locks: show delegations as "DELEG" in /proc/locks Now that they are a distinct lease type, show them as such. Cc: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@primarydata.com>
2014-08-12 00:36:54 +07:00
if (fl->fl_flags & FL_DELEG)
seq_puts(f, "DELEG ");
else
seq_puts(f, "LEASE ");
locks: minor lease cleanup Use a helper function, to simplify upcoming changes. Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-27 07:10:51 +07:00
if (lease_breaking(fl))
fs/locks.c: replace seq_printf by seq_puts Replace seq_printf where possible Cc: Jeff Layton <jlayton@redhat.com> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Fabian Frederick <fabf@skynet.be> Signed-off-by: Jeff Layton <jlayton@poochiereds.net>
2014-05-10 01:13:05 +07:00
seq_puts(f, "BREAKING ");
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
else if (fl->fl_file)
fs/locks.c: replace seq_printf by seq_puts Replace seq_printf where possible Cc: Jeff Layton <jlayton@redhat.com> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Fabian Frederick <fabf@skynet.be> Signed-off-by: Jeff Layton <jlayton@poochiereds.net>
2014-05-10 01:13:05 +07:00
seq_puts(f, "ACTIVE ");
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
else
fs/locks.c: replace seq_printf by seq_puts Replace seq_printf where possible Cc: Jeff Layton <jlayton@redhat.com> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Fabian Frederick <fabf@skynet.be> Signed-off-by: Jeff Layton <jlayton@poochiereds.net>
2014-05-10 01:13:05 +07:00
seq_puts(f, "BREAKER ");
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
} else {
fs/locks.c: replace seq_printf by seq_puts Replace seq_printf where possible Cc: Jeff Layton <jlayton@redhat.com> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Fabian Frederick <fabf@skynet.be> Signed-off-by: Jeff Layton <jlayton@poochiereds.net>
2014-05-10 01:13:05 +07:00
seq_puts(f, "UNKNOWN UNKNOWN ");
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
if (fl->fl_type & LOCK_MAND) {
Rework /proc/locks via seq_files and seq_list helpers Currently /proc/locks is shown with a proc_read function, but its behavior is rather complex as it has to manually handle current offset and buffer length. On the other hand, files that show objects from lists can be easily reimplemented using the sequential files and the seq_list_XXX() helpers. This saves (as usually) 16 lines of code and more than 200 from the .text section. [akpm@linux-foundation.org: no externs in C] [akpm@linux-foundation.org: warning fixes] Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2007-10-02 04:41:15 +07:00
seq_printf(f, "%s ",
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
(fl->fl_type & LOCK_READ)
? (fl->fl_type & LOCK_WRITE) ? "RW " : "READ "
: (fl->fl_type & LOCK_WRITE) ? "WRITE" : "NONE ");
} else {
Rework /proc/locks via seq_files and seq_list helpers Currently /proc/locks is shown with a proc_read function, but its behavior is rather complex as it has to manually handle current offset and buffer length. On the other hand, files that show objects from lists can be easily reimplemented using the sequential files and the seq_list_XXX() helpers. This saves (as usually) 16 lines of code and more than 200 from the .text section. [akpm@linux-foundation.org: no externs in C] [akpm@linux-foundation.org: warning fixes] Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2007-10-02 04:41:15 +07:00
seq_printf(f, "%s ",
locks: minor lease cleanup Use a helper function, to simplify upcoming changes. Reviewed-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-27 07:10:51 +07:00
(lease_breaking(fl))
vfs: don't treat fl_type as a bitmap The rules for fl_type are rather convoluted. Typically it's treated as holding specific values, except in the case of LOCK_MAND, in which case it can be or'ed with LOCK_READ|LOCK_WRITE. On some arches F_WRLCK == 2 and F_UNLCK == 3, so and'ing with F_WRLCK will also catch the F_UNLCK case. It's unlikely in either case here that we'd ever see F_UNLCK since those shouldn't end up on any lists, but it's still best to be consistent. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2012-08-03 02:46:30 +07:00
? (fl->fl_type == F_UNLCK) ? "UNLCK" : "READ "
: (fl->fl_type == F_WRLCK) ? "WRITE" : "READ ");
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
if (inode) {
locks: get rid of WE_CAN_BREAK_LSLK_NOW dead code As Bruce points out, there's no compelling reason to change /proc/locks output at this point. If we did want to do this, then we'd almost certainly want to introduce a new file to display this info (maybe via debugfs?). Let's remove the dead WE_CAN_BREAK_LSLK_NOW ifdef here and just plan to stay with the legacy format. Reported-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-04-03 20:04:04 +07:00
/* userspace relies on this representation of dev_t */
pid-namespaces-vs-locks-interaction fcntl(F_GETLK,..) can return pid of process for not current pid namespace (if process is belonged to the several namespaces). It is true also for pids in /proc/locks. So correct behavior is saving pointer to the struct pid of the process lock owner. Signed-off-by: Vitaliy Gusev <vgusev@openvz.org> Acked-by: Serge Hallyn <serue@us.ibm.com> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-01-17 07:07:08 +07:00
seq_printf(f, "%d %02x:%02x:%ld ", fl_pid,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
MAJOR(inode->i_sb->s_dev),
MINOR(inode->i_sb->s_dev), inode->i_ino);
} else {
pid-namespaces-vs-locks-interaction fcntl(F_GETLK,..) can return pid of process for not current pid namespace (if process is belonged to the several namespaces). It is true also for pids in /proc/locks. So correct behavior is saving pointer to the struct pid of the process lock owner. Signed-off-by: Vitaliy Gusev <vgusev@openvz.org> Acked-by: Serge Hallyn <serue@us.ibm.com> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
2008-01-17 07:07:08 +07:00
seq_printf(f, "%d <none>:0 ", fl_pid);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
if (IS_POSIX(fl)) {
if (fl->fl_end == OFFSET_MAX)
Rework /proc/locks via seq_files and seq_list helpers Currently /proc/locks is shown with a proc_read function, but its behavior is rather complex as it has to manually handle current offset and buffer length. On the other hand, files that show objects from lists can be easily reimplemented using the sequential files and the seq_list_XXX() helpers. This saves (as usually) 16 lines of code and more than 200 from the .text section. [akpm@linux-foundation.org: no externs in C] [akpm@linux-foundation.org: warning fixes] Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2007-10-02 04:41:15 +07:00
seq_printf(f, "%Ld EOF\n", fl->fl_start);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
else
Rework /proc/locks via seq_files and seq_list helpers Currently /proc/locks is shown with a proc_read function, but its behavior is rather complex as it has to manually handle current offset and buffer length. On the other hand, files that show objects from lists can be easily reimplemented using the sequential files and the seq_list_XXX() helpers. This saves (as usually) 16 lines of code and more than 200 from the .text section. [akpm@linux-foundation.org: no externs in C] [akpm@linux-foundation.org: warning fixes] Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2007-10-02 04:41:15 +07:00
seq_printf(f, "%Ld %Ld\n", fl->fl_start, fl->fl_end);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
} else {
fs/locks.c: replace seq_printf by seq_puts Replace seq_printf where possible Cc: Jeff Layton <jlayton@redhat.com> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Fabian Frederick <fabf@skynet.be> Signed-off-by: Jeff Layton <jlayton@poochiereds.net>
2014-05-10 01:13:05 +07:00
seq_puts(f, "0 EOF\n");
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
}
Rework /proc/locks via seq_files and seq_list helpers Currently /proc/locks is shown with a proc_read function, but its behavior is rather complex as it has to manually handle current offset and buffer length. On the other hand, files that show objects from lists can be easily reimplemented using the sequential files and the seq_list_XXX() helpers. This saves (as usually) 16 lines of code and more than 200 from the .text section. [akpm@linux-foundation.org: no externs in C] [akpm@linux-foundation.org: warning fixes] Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2007-10-02 04:41:15 +07:00
static int locks_show(struct seq_file *f, void *v)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
{
locks: move file_lock_list to a set of percpu hlist_heads and convert file_lock_lock to an lglock The file_lock_list is only used for /proc/locks. The vastly common case is for locks to be put onto the list and come off again, without ever being traversed. Help optimize for this use-case by moving to percpu hlist_head-s. At the same time, we can make the locking less contentious by moving to an lglock. When iterating over the lists for /proc/locks, we must take the global lock and then iterate over each CPU's list in turn. This change necessitates a new fl_link_cpu field to keep track of which CPU the entry is on. On x86_64 at least, this field is placed within an existing hole in the struct to avoid growing the size. Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:22 +07:00
struct locks_iterator *iter = f->private;
Rework /proc/locks via seq_files and seq_list helpers Currently /proc/locks is shown with a proc_read function, but its behavior is rather complex as it has to manually handle current offset and buffer length. On the other hand, files that show objects from lists can be easily reimplemented using the sequential files and the seq_list_XXX() helpers. This saves (as usually) 16 lines of code and more than 200 from the .text section. [akpm@linux-foundation.org: no externs in C] [akpm@linux-foundation.org: warning fixes] Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2007-10-02 04:41:15 +07:00
struct file_lock *fl, *bfl;
locks: Filter /proc/locks output on proc pid ns On busy container servers reading /proc/locks shows all the locks created by all clients. This can cause large latency spikes. In my case I observed lsof taking up to 5-10 seconds while processing around 50k locks. Fix this by limiting the locks shown only to those created in the same pidns as the one the proc fs was mounted in. When reading /proc/locks from the init_pid_ns proc instance then perform no filtering [ jlayton: reformat comments for 80 columns ] Signed-off-by: Nikolay Borisov <kernel@kyup.com> Suggested-by: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2016-08-18 03:18:46 +07:00
struct pid_namespace *proc_pidns = file_inode(f->file)->i_sb->s_fs_info;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
locks: convert fl_link to a hlist_node Testing has shown that iterating over the blocked_list for deadlock detection turns out to be a bottleneck. In order to alleviate that, begin the process of turning it into a hashtable. We start by turning the fl_link into a hlist_node and the global lists into hlists. A later patch will do the conversion of the blocked_list to a hashtable. Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:17 +07:00
fl = hlist_entry(v, struct file_lock, fl_link);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
fs/locks: Remove fl_nspid and use fs-specific l_pid for remote locks Since commit c69899a17ca4 "NFSv4: Update of VFS byte range lock must be atomic with the stateid update", NFSv4 has been inserting locks in rpciod worker context. The result is that the file_lock's fl_nspid is the kworker's pid instead of the original userspace pid. The fl_nspid is only used to represent the namespaced virtual pid number when displaying locks or returning from F_GETLK. There's no reason to set it for every inserted lock, since we can usually just look it up from fl_pid. So, instead of looking up and holding struct pid for every lock, let's just look up the virtual pid number from fl_pid when it is needed. That means we can remove fl_nspid entirely. The translaton and presentation of fl_pid should handle the following four cases: 1 - F_GETLK on a remote file with a remote lock: In this case, the filesystem should determine the l_pid to return here. Filesystems should indicate that the fl_pid represents a non-local pid value that should not be translated by returning an fl_pid <= 0. 2 - F_GETLK on a local file with a remote lock: This should be the l_pid of the lock manager process, and translated. 3 - F_GETLK on a remote file with a local lock, and 4 - F_GETLK on a local file with a local lock: These should be the translated l_pid of the local locking process. Fuse was already doing the correct thing by translating the pid into the caller's namespace. With this change we must update fuse to translate to init's pid namespace, so that the locks API can then translate from init's pid namespace into the pid namespace of the caller. With this change, the locks API will expect that if a filesystem returns a remote pid as opposed to a local pid for F_GETLK, that remote pid will be <= 0. This signifies that the pid is remote, and the locks API will forego translating that pid into the pid namespace of the local calling process. Finally, we convert remote filesystems to present remote pids using negative numbers. Have lustre, 9p, ceph, cifs, and dlm negate the remote pid returned for F_GETLK lock requests. Since local pids will never be larger than PID_MAX_LIMIT (which is currently defined as <= 4 million), but pid_t is an unsigned int, we should have plenty of room to represent remote pids with negative numbers if we assume that remote pid numbers are similarly limited. If this is not the case, then we run the risk of having a remote pid returned for which there is also a corresponding local pid. This is a problem we have now, but this patch should reduce the chances of that occurring, while also returning those remote pid numbers, for whatever that may be worth. Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2017-07-16 21:28:22 +07:00
if (locks_translate_pid(fl, proc_pidns) == 0)
locks: Filter /proc/locks output on proc pid ns On busy container servers reading /proc/locks shows all the locks created by all clients. This can cause large latency spikes. In my case I observed lsof taking up to 5-10 seconds while processing around 50k locks. Fix this by limiting the locks shown only to those created in the same pidns as the one the proc fs was mounted in. When reading /proc/locks from the init_pid_ns proc instance then perform no filtering [ jlayton: reformat comments for 80 columns ] Signed-off-by: Nikolay Borisov <kernel@kyup.com> Suggested-by: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2016-08-18 03:18:46 +07:00
return 0;
locks: move file_lock_list to a set of percpu hlist_heads and convert file_lock_lock to an lglock The file_lock_list is only used for /proc/locks. The vastly common case is for locks to be put onto the list and come off again, without ever being traversed. Help optimize for this use-case by moving to percpu hlist_head-s. At the same time, we can make the locking less contentious by moving to an lglock. When iterating over the lists for /proc/locks, we must take the global lock and then iterate over each CPU's list in turn. This change necessitates a new fl_link_cpu field to keep track of which CPU the entry is on. On x86_64 at least, this field is placed within an existing hole in the struct to avoid growing the size. Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:22 +07:00
lock_get_status(f, fl, iter->li_pos, "");
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
fs/locks: rename some lists and pointers. struct file lock contains an 'fl_next' pointer which is used to point to the lock that this request is blocked waiting for. So rename it to fl_blocker. The fl_blocked list_head in an active lock is the head of a list of blocked requests. In a request it is a node in that list. These are two distinct uses, so replace with two list_heads with different names. fl_blocked_requests is the head of a list of blocked requests fl_blocked_member is a node in a member of that list. The two different list_heads are never used at the same time, but that will change in a future patch. Note that a tracepoint is changed to report fl_blocker instead of fl_next. Signed-off-by: NeilBrown <neilb@suse.com> Reviewed-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Jeff Layton <jlayton@kernel.org>
2018-11-30 06:04:08 +07:00
list_for_each_entry(bfl, &fl->fl_blocked_requests, fl_blocked_member)
locks: move file_lock_list to a set of percpu hlist_heads and convert file_lock_lock to an lglock The file_lock_list is only used for /proc/locks. The vastly common case is for locks to be put onto the list and come off again, without ever being traversed. Help optimize for this use-case by moving to percpu hlist_head-s. At the same time, we can make the locking less contentious by moving to an lglock. When iterating over the lists for /proc/locks, we must take the global lock and then iterate over each CPU's list in turn. This change necessitates a new fl_link_cpu field to keep track of which CPU the entry is on. On x86_64 at least, this field is placed within an existing hole in the struct to avoid growing the size. Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:22 +07:00
lock_get_status(f, bfl, iter->li_pos, " ->");
fs/locks.c: use list_for_each_entry() instead of list_for_each() fs/locks.c: use list_for_each_entry() instead of list_for_each() in posix_locks_deadlock() and get_locks_status() Signed-off-by: Matthias Kaehlcke <matthias.kaehlcke@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2007-10-03 01:21:34 +07:00
Rework /proc/locks via seq_files and seq_list helpers Currently /proc/locks is shown with a proc_read function, but its behavior is rather complex as it has to manually handle current offset and buffer length. On the other hand, files that show objects from lists can be easily reimplemented using the sequential files and the seq_list_XXX() helpers. This saves (as usually) 16 lines of code and more than 200 from the .text section. [akpm@linux-foundation.org: no externs in C] [akpm@linux-foundation.org: warning fixes] Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2007-10-02 04:41:15 +07:00
return 0;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
proc: show locks in /proc/pid/fdinfo/X Let's show locks which are associated with a file descriptor in its fdinfo file. Currently we don't have a reliable way to determine who holds a lock. We can find some information in /proc/locks, but PID which is reported there can be wrong. For example, a process takes a lock, then forks a child and dies. In this case /proc/locks contains the parent pid, which can be reused by another process. $ cat /proc/locks ... 6: FLOCK ADVISORY WRITE 324 00:13:13431 0 EOF ... $ ps -C rpcbind PID TTY TIME CMD 332 ? 00:00:00 rpcbind $ cat /proc/332/fdinfo/4 pos: 0 flags: 0100000 mnt_id: 22 lock: 1: FLOCK ADVISORY WRITE 324 00:13:13431 0 EOF $ ls -l /proc/332/fd/4 lr-x------ 1 root root 64 Mar 5 14:43 /proc/332/fd/4 -> /run/rpcbind.lock $ ls -l /proc/324/fd/ total 0 lrwx------ 1 root root 64 Feb 27 14:50 0 -> /dev/pts/0 lrwx------ 1 root root 64 Feb 27 14:50 1 -> /dev/pts/0 lrwx------ 1 root root 64 Feb 27 14:49 2 -> /dev/pts/0 You can see that the process with the 324 pid doesn't hold the lock. This information is required for proper dumping and restoring file locks. Signed-off-by: Andrey Vagin <avagin@openvz.org> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Acked-by: Jeff Layton <jlayton@poochiereds.net> Acked-by: "J. Bruce Fields" <bfields@fieldses.org> Acked-by: Cyrill Gorcunov <gorcunov@openvz.org> Cc: Pavel Emelyanov <xemul@parallels.com> Cc: Joe Perches <joe@perches.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-04-17 02:49:38 +07:00
static void __show_fd_locks(struct seq_file *f,
struct list_head *head, int *id,
struct file *filp, struct files_struct *files)
{
struct file_lock *fl;
list_for_each_entry(fl, head, fl_list) {
if (filp != fl->fl_file)
continue;
if (fl->fl_owner != files &&
fl->fl_owner != filp)
continue;
(*id)++;
seq_puts(f, "lock:\t");
lock_get_status(f, fl, *id, "");
}
}
void show_fd_locks(struct seq_file *f,
struct file *filp, struct files_struct *files)
{
locks: fix file locking on overlayfs This patch allows flock, posix locks, ofd locks and leases to work correctly on overlayfs. Instead of using the underlying inode for storing lock context use the overlay inode. This allows locks to be persistent across copy-up. This is done by introducing locks_inode() helper and using it instead of file_inode() to get the inode in locking code. For non-overlayfs the two are equivalent, except for an extra pointer dereference in locks_inode(). Since lock operations are in "struct file_operations" we must also make sure not to call underlying filesystem's lock operations. Introcude a super block flag MS_NOREMOTELOCK to this effect. Signed-off-by: Miklos Szeredi <mszeredi@redhat.com> Acked-by: Jeff Layton <jlayton@poochiereds.net> Cc: "J. Bruce Fields" <bfields@fieldses.org>
2016-09-16 17:44:20 +07:00
struct inode *inode = locks_inode(filp);
proc: show locks in /proc/pid/fdinfo/X Let's show locks which are associated with a file descriptor in its fdinfo file. Currently we don't have a reliable way to determine who holds a lock. We can find some information in /proc/locks, but PID which is reported there can be wrong. For example, a process takes a lock, then forks a child and dies. In this case /proc/locks contains the parent pid, which can be reused by another process. $ cat /proc/locks ... 6: FLOCK ADVISORY WRITE 324 00:13:13431 0 EOF ... $ ps -C rpcbind PID TTY TIME CMD 332 ? 00:00:00 rpcbind $ cat /proc/332/fdinfo/4 pos: 0 flags: 0100000 mnt_id: 22 lock: 1: FLOCK ADVISORY WRITE 324 00:13:13431 0 EOF $ ls -l /proc/332/fd/4 lr-x------ 1 root root 64 Mar 5 14:43 /proc/332/fd/4 -> /run/rpcbind.lock $ ls -l /proc/324/fd/ total 0 lrwx------ 1 root root 64 Feb 27 14:50 0 -> /dev/pts/0 lrwx------ 1 root root 64 Feb 27 14:50 1 -> /dev/pts/0 lrwx------ 1 root root 64 Feb 27 14:49 2 -> /dev/pts/0 You can see that the process with the 324 pid doesn't hold the lock. This information is required for proper dumping and restoring file locks. Signed-off-by: Andrey Vagin <avagin@openvz.org> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Acked-by: Jeff Layton <jlayton@poochiereds.net> Acked-by: "J. Bruce Fields" <bfields@fieldses.org> Acked-by: Cyrill Gorcunov <gorcunov@openvz.org> Cc: Pavel Emelyanov <xemul@parallels.com> Cc: Joe Perches <joe@perches.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-04-17 02:49:38 +07:00
struct file_lock_context *ctx;
int id = 0;
fs: fix data races on inode->i_flctx locks_get_lock_context() uses cmpxchg() to install i_flctx. cmpxchg() is a release operation which is correct. But it uses a plain load to load i_flctx. This is incorrect. Subsequent loads from i_flctx can hoist above the load of i_flctx pointer itself and observe uninitialized garbage there. This in turn can lead to corruption of ctx->flc_lock and other members. Documentation/memory-barriers.txt explicitly requires to use a barrier in such context: "A load-load control dependency requires a full read memory barrier". Use smp_load_acquire() in locks_get_lock_context() and in bunch of other functions that can proceed concurrently with locks_get_lock_context(). The data race was found with KernelThreadSanitizer (KTSAN). Signed-off-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-09-21 14:43:06 +07:00
ctx = smp_load_acquire(&inode->i_flctx);
proc: show locks in /proc/pid/fdinfo/X Let's show locks which are associated with a file descriptor in its fdinfo file. Currently we don't have a reliable way to determine who holds a lock. We can find some information in /proc/locks, but PID which is reported there can be wrong. For example, a process takes a lock, then forks a child and dies. In this case /proc/locks contains the parent pid, which can be reused by another process. $ cat /proc/locks ... 6: FLOCK ADVISORY WRITE 324 00:13:13431 0 EOF ... $ ps -C rpcbind PID TTY TIME CMD 332 ? 00:00:00 rpcbind $ cat /proc/332/fdinfo/4 pos: 0 flags: 0100000 mnt_id: 22 lock: 1: FLOCK ADVISORY WRITE 324 00:13:13431 0 EOF $ ls -l /proc/332/fd/4 lr-x------ 1 root root 64 Mar 5 14:43 /proc/332/fd/4 -> /run/rpcbind.lock $ ls -l /proc/324/fd/ total 0 lrwx------ 1 root root 64 Feb 27 14:50 0 -> /dev/pts/0 lrwx------ 1 root root 64 Feb 27 14:50 1 -> /dev/pts/0 lrwx------ 1 root root 64 Feb 27 14:49 2 -> /dev/pts/0 You can see that the process with the 324 pid doesn't hold the lock. This information is required for proper dumping and restoring file locks. Signed-off-by: Andrey Vagin <avagin@openvz.org> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Acked-by: Jeff Layton <jlayton@poochiereds.net> Acked-by: "J. Bruce Fields" <bfields@fieldses.org> Acked-by: Cyrill Gorcunov <gorcunov@openvz.org> Cc: Pavel Emelyanov <xemul@parallels.com> Cc: Joe Perches <joe@perches.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-04-17 02:49:38 +07:00
if (!ctx)
return;
spin_lock(&ctx->flc_lock);
__show_fd_locks(f, &ctx->flc_flock, &id, filp, files);
__show_fd_locks(f, &ctx->flc_posix, &id, filp, files);
__show_fd_locks(f, &ctx->flc_lease, &id, filp, files);
spin_unlock(&ctx->flc_lock);
}
Rework /proc/locks via seq_files and seq_list helpers Currently /proc/locks is shown with a proc_read function, but its behavior is rather complex as it has to manually handle current offset and buffer length. On the other hand, files that show objects from lists can be easily reimplemented using the sequential files and the seq_list_XXX() helpers. This saves (as usually) 16 lines of code and more than 200 from the .text section. [akpm@linux-foundation.org: no externs in C] [akpm@linux-foundation.org: warning fixes] Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2007-10-02 04:41:15 +07:00
static void *locks_start(struct seq_file *f, loff_t *pos)
locks: add __acquires and __releases annotations to locks_start and locks_stop ...to make sparse happy. Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:07 +07:00
__acquires(&blocked_lock_lock)
Rework /proc/locks via seq_files and seq_list helpers Currently /proc/locks is shown with a proc_read function, but its behavior is rather complex as it has to manually handle current offset and buffer length. On the other hand, files that show objects from lists can be easily reimplemented using the sequential files and the seq_list_XXX() helpers. This saves (as usually) 16 lines of code and more than 200 from the .text section. [akpm@linux-foundation.org: no externs in C] [akpm@linux-foundation.org: warning fixes] Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2007-10-02 04:41:15 +07:00
{
locks: move file_lock_list to a set of percpu hlist_heads and convert file_lock_lock to an lglock The file_lock_list is only used for /proc/locks. The vastly common case is for locks to be put onto the list and come off again, without ever being traversed. Help optimize for this use-case by moving to percpu hlist_head-s. At the same time, we can make the locking less contentious by moving to an lglock. When iterating over the lists for /proc/locks, we must take the global lock and then iterate over each CPU's list in turn. This change necessitates a new fl_link_cpu field to keep track of which CPU the entry is on. On x86_64 at least, this field is placed within an existing hole in the struct to avoid growing the size. Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:22 +07:00
struct locks_iterator *iter = f->private;
procfs: fix numbering in /proc/locks The lock number in /proc/locks (first field) is implemented by a counter (private field of struct seq_file) which is incremented at each call of locks_show() and reset to 1 in locks_start() whatever the offset is. It should be reset according to the actual position in the list. Because of this, the numbering erratically restarts at 1 several times when reading a long /proc/locks file. Moreover, locks_show() can be called twice to print a single line thus skipping a number. The counter should be incremented in locks_next(). And last, pos is a loff_t, which can be bigger than a pointer, so we don't use the pointer as an integer anymore, and allocate a loff_t instead. Signed-off-by: Jerome Marchand <jmarchan@redhat.com> Cc: Pavel Emelyanov <xemul@parallels.com> Cc: Matthew Wilcox <matthew@wil.cx> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-10-27 04:22:33 +07:00
locks: move file_lock_list to a set of percpu hlist_heads and convert file_lock_lock to an lglock The file_lock_list is only used for /proc/locks. The vastly common case is for locks to be put onto the list and come off again, without ever being traversed. Help optimize for this use-case by moving to percpu hlist_head-s. At the same time, we can make the locking less contentious by moving to an lglock. When iterating over the lists for /proc/locks, we must take the global lock and then iterate over each CPU's list in turn. This change necessitates a new fl_link_cpu field to keep track of which CPU the entry is on. On x86_64 at least, this field is placed within an existing hole in the struct to avoid growing the size. Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:22 +07:00
iter->li_pos = *pos + 1;
fs/locks: Replace lg_global with a percpu-rwsem Replace the global part of the lglock with a percpu-rwsem. Since fcl_lock is a spinlock and itself nests under i_lock, which too is a spinlock we cannot acquire sleeping locks at locks_{insert,remove}_global_locks(). We can however wrap all fcl_lock acquisitions with percpu_down_read such that all invocations of locks_{insert,remove}_global_locks() have that read lock held. This allows us to replace the lg_global part of the lglock with the write side of the rwsem. In the absense of writers, percpu_{down,up}_read() are free of atomic instructions. This further avoids the very long preempt-disable regions caused by lglock on larger machines. Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Al Viro <viro@ZenIV.linux.org.uk> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: dave@stgolabs.net Cc: der.herr@hofr.at Cc: paulmck@linux.vnet.ibm.com Cc: riel@redhat.com Cc: tj@kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-06-22 19:16:33 +07:00
percpu_down_write(&file_rwsem);
locks: give the blocked_hash its own spinlock There's no reason we have to protect the blocked_hash and file_lock_list with the same spinlock. With the tests I have, breaking it in two gives a barely measurable performance benefit, but it seems reasonable to make this locking as granular as possible. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:20 +07:00
spin_lock(&blocked_lock_lock);
fs/locks: Replace lg_local with a per-cpu spinlock As Oleg suggested, replace file_lock_list with a structure containing the hlist head and a spinlock. This completely removes the lglock from fs/locks. Suggested-by: Oleg Nesterov <oleg@redhat.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Al Viro <viro@ZenIV.linux.org.uk> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: dave@stgolabs.net Cc: der.herr@hofr.at Cc: paulmck@linux.vnet.ibm.com Cc: riel@redhat.com Cc: tj@kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-06-22 19:16:34 +07:00
return seq_hlist_start_percpu(&file_lock_list.hlist, &iter->li_cpu, *pos);
Rework /proc/locks via seq_files and seq_list helpers Currently /proc/locks is shown with a proc_read function, but its behavior is rather complex as it has to manually handle current offset and buffer length. On the other hand, files that show objects from lists can be easily reimplemented using the sequential files and the seq_list_XXX() helpers. This saves (as usually) 16 lines of code and more than 200 from the .text section. [akpm@linux-foundation.org: no externs in C] [akpm@linux-foundation.org: warning fixes] Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2007-10-02 04:41:15 +07:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
Rework /proc/locks via seq_files and seq_list helpers Currently /proc/locks is shown with a proc_read function, but its behavior is rather complex as it has to manually handle current offset and buffer length. On the other hand, files that show objects from lists can be easily reimplemented using the sequential files and the seq_list_XXX() helpers. This saves (as usually) 16 lines of code and more than 200 from the .text section. [akpm@linux-foundation.org: no externs in C] [akpm@linux-foundation.org: warning fixes] Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2007-10-02 04:41:15 +07:00
static void *locks_next(struct seq_file *f, void *v, loff_t *pos)
{
locks: move file_lock_list to a set of percpu hlist_heads and convert file_lock_lock to an lglock The file_lock_list is only used for /proc/locks. The vastly common case is for locks to be put onto the list and come off again, without ever being traversed. Help optimize for this use-case by moving to percpu hlist_head-s. At the same time, we can make the locking less contentious by moving to an lglock. When iterating over the lists for /proc/locks, we must take the global lock and then iterate over each CPU's list in turn. This change necessitates a new fl_link_cpu field to keep track of which CPU the entry is on. On x86_64 at least, this field is placed within an existing hole in the struct to avoid growing the size. Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:22 +07:00
struct locks_iterator *iter = f->private;
++iter->li_pos;
fs/locks: Replace lg_local with a per-cpu spinlock As Oleg suggested, replace file_lock_list with a structure containing the hlist head and a spinlock. This completely removes the lglock from fs/locks. Suggested-by: Oleg Nesterov <oleg@redhat.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Al Viro <viro@ZenIV.linux.org.uk> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: dave@stgolabs.net Cc: der.herr@hofr.at Cc: paulmck@linux.vnet.ibm.com Cc: riel@redhat.com Cc: tj@kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-06-22 19:16:34 +07:00
return seq_hlist_next_percpu(v, &file_lock_list.hlist, &iter->li_cpu, pos);
Rework /proc/locks via seq_files and seq_list helpers Currently /proc/locks is shown with a proc_read function, but its behavior is rather complex as it has to manually handle current offset and buffer length. On the other hand, files that show objects from lists can be easily reimplemented using the sequential files and the seq_list_XXX() helpers. This saves (as usually) 16 lines of code and more than 200 from the .text section. [akpm@linux-foundation.org: no externs in C] [akpm@linux-foundation.org: warning fixes] Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2007-10-02 04:41:15 +07:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
Rework /proc/locks via seq_files and seq_list helpers Currently /proc/locks is shown with a proc_read function, but its behavior is rather complex as it has to manually handle current offset and buffer length. On the other hand, files that show objects from lists can be easily reimplemented using the sequential files and the seq_list_XXX() helpers. This saves (as usually) 16 lines of code and more than 200 from the .text section. [akpm@linux-foundation.org: no externs in C] [akpm@linux-foundation.org: warning fixes] Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2007-10-02 04:41:15 +07:00
static void locks_stop(struct seq_file *f, void *v)
locks: add __acquires and __releases annotations to locks_start and locks_stop ...to make sparse happy. Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Jeff Layton <jlayton@redhat.com>
2014-02-04 00:13:07 +07:00
__releases(&blocked_lock_lock)
Rework /proc/locks via seq_files and seq_list helpers Currently /proc/locks is shown with a proc_read function, but its behavior is rather complex as it has to manually handle current offset and buffer length. On the other hand, files that show objects from lists can be easily reimplemented using the sequential files and the seq_list_XXX() helpers. This saves (as usually) 16 lines of code and more than 200 from the .text section. [akpm@linux-foundation.org: no externs in C] [akpm@linux-foundation.org: warning fixes] Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2007-10-02 04:41:15 +07:00
{
locks: give the blocked_hash its own spinlock There's no reason we have to protect the blocked_hash and file_lock_list with the same spinlock. With the tests I have, breaking it in two gives a barely measurable performance benefit, but it seems reasonable to make this locking as granular as possible. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:20 +07:00
spin_unlock(&blocked_lock_lock);
fs/locks: Replace lg_global with a percpu-rwsem Replace the global part of the lglock with a percpu-rwsem. Since fcl_lock is a spinlock and itself nests under i_lock, which too is a spinlock we cannot acquire sleeping locks at locks_{insert,remove}_global_locks(). We can however wrap all fcl_lock acquisitions with percpu_down_read such that all invocations of locks_{insert,remove}_global_locks() have that read lock held. This allows us to replace the lg_global part of the lglock with the write side of the rwsem. In the absense of writers, percpu_{down,up}_read() are free of atomic instructions. This further avoids the very long preempt-disable regions caused by lglock on larger machines. Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Al Viro <viro@ZenIV.linux.org.uk> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: dave@stgolabs.net Cc: der.herr@hofr.at Cc: paulmck@linux.vnet.ibm.com Cc: riel@redhat.com Cc: tj@kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-06-22 19:16:33 +07:00
percpu_up_write(&file_rwsem);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
}
proc: move rest of /proc/locks to fs/locks.c Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
2008-10-05 01:34:18 +07:00
static const struct seq_operations locks_seq_operations = {
Rework /proc/locks via seq_files and seq_list helpers Currently /proc/locks is shown with a proc_read function, but its behavior is rather complex as it has to manually handle current offset and buffer length. On the other hand, files that show objects from lists can be easily reimplemented using the sequential files and the seq_list_XXX() helpers. This saves (as usually) 16 lines of code and more than 200 from the .text section. [akpm@linux-foundation.org: no externs in C] [akpm@linux-foundation.org: warning fixes] Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2007-10-02 04:41:15 +07:00
.start = locks_start,
.next = locks_next,
.stop = locks_stop,
.show = locks_show,
};
proc: move rest of /proc/locks to fs/locks.c Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
2008-10-05 01:34:18 +07:00
static int __init proc_locks_init(void)
{
proc: introduce proc_create_seq_private Variant of proc_create_data that directly take a struct seq_operations argument + a private state size and drastically reduces the boilerplate code in the callers. All trivial callers converted over. Signed-off-by: Christoph Hellwig <hch@lst.de>
2018-04-24 22:05:17 +07:00
proc_create_seq_private("locks", 0, NULL, &locks_seq_operations,
sizeof(struct locks_iterator), NULL);
proc: move rest of /proc/locks to fs/locks.c Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
2008-10-05 01:34:18 +07:00
return 0;
}
fs: make locks.c explicitly non-modular The Kconfig currently controlling compilation of this code is: config FILE_LOCKING bool "Enable POSIX file locking API" if EXPERT ...meaning that it currently is not being built as a module by anyone. Lets remove the couple traces of modularity so that when reading the driver there is no doubt it is builtin-only. Since module_init translates to device_initcall in the non-modular case, the init ordering gets bumped to one level earlier when we use the more appropriate fs_initcall here. However we've made similar changes before without any fallout and none is expected here either. Cc: Jeff Layton <jlayton@poochiereds.net> Acked-by: Jeff Layton <jlayton@poochiereds.net> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: linux-fsdevel@vger.kernel.org Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com> Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
2015-12-18 02:11:03 +07:00
fs_initcall(proc_locks_init);
Rework /proc/locks via seq_files and seq_list helpers Currently /proc/locks is shown with a proc_read function, but its behavior is rather complex as it has to manually handle current offset and buffer length. On the other hand, files that show objects from lists can be easily reimplemented using the sequential files and the seq_list_XXX() helpers. This saves (as usually) 16 lines of code and more than 200 from the .text section. [akpm@linux-foundation.org: no externs in C] [akpm@linux-foundation.org: warning fixes] Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2007-10-02 04:41:15 +07:00
#endif
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
static int __init filelock_init(void)
{
locks: move file_lock_list to a set of percpu hlist_heads and convert file_lock_lock to an lglock The file_lock_list is only used for /proc/locks. The vastly common case is for locks to be put onto the list and come off again, without ever being traversed. Help optimize for this use-case by moving to percpu hlist_head-s. At the same time, we can make the locking less contentious by moving to an lglock. When iterating over the lists for /proc/locks, we must take the global lock and then iterate over each CPU's list in turn. This change necessitates a new fl_link_cpu field to keep track of which CPU the entry is on. On x86_64 at least, this field is placed within an existing hole in the struct to avoid growing the size. Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:22 +07:00
int i;
locks: add a new struct file_locking_context pointer to struct inode The current scheme of using the i_flock list is really difficult to manage. There is also a legitimate desire for a per-inode spinlock to manage these lists that isn't the i_lock. Start conversion to a new scheme to eventually replace the old i_flock list with a new "file_lock_context" object. We start by adding a new i_flctx to struct inode. For now, it lives in parallel with i_flock list, but will eventually replace it. The idea is to allocate a structure to sit in that pointer and act as a locus for all things file locking. We allocate a file_lock_context for an inode when the first lock is added to it, and it's only freed when the inode is freed. We use the i_lock to protect the assignment, but afterward it should mostly be accessed locklessly. Signed-off-by: Jeff Layton <jlayton@primarydata.com> Acked-by: Christoph Hellwig <hch@lst.de>
2015-01-17 03:05:54 +07:00
flctx_cache = kmem_cache_create("file_lock_ctx",
sizeof(struct file_lock_context), 0, SLAB_PANIC, NULL);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
filelock_cache = kmem_cache_create("file_lock_cache",
fs: locks: remove init_once From: Miklos Szeredi <mszeredi@suse.cz> Remove SLAB initialization entirely, as suggested by Bruce and Linus. Allocate with __GFP_ZERO instead and only initialize list heads. Signed-off-by: Miklos Szeredi <mszeredi@suse.cz> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
2011-07-07 18:06:09 +07:00
sizeof(struct file_lock), 0, SLAB_PANIC, NULL);
locks: move file_lock_list to a set of percpu hlist_heads and convert file_lock_lock to an lglock The file_lock_list is only used for /proc/locks. The vastly common case is for locks to be put onto the list and come off again, without ever being traversed. Help optimize for this use-case by moving to percpu hlist_head-s. At the same time, we can make the locking less contentious by moving to an lglock. When iterating over the lists for /proc/locks, we must take the global lock and then iterate over each CPU's list in turn. This change necessitates a new fl_link_cpu field to keep track of which CPU the entry is on. On x86_64 at least, this field is placed within an existing hole in the struct to avoid growing the size. Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:22 +07:00
fs/locks: Replace lg_local with a per-cpu spinlock As Oleg suggested, replace file_lock_list with a structure containing the hlist head and a spinlock. This completely removes the lglock from fs/locks. Suggested-by: Oleg Nesterov <oleg@redhat.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Al Viro <viro@ZenIV.linux.org.uk> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: dave@stgolabs.net Cc: der.herr@hofr.at Cc: paulmck@linux.vnet.ibm.com Cc: riel@redhat.com Cc: tj@kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-06-22 19:16:34 +07:00
for_each_possible_cpu(i) {
struct file_lock_list_struct *fll = per_cpu_ptr(&file_lock_list, i);
spin_lock_init(&fll->lock);
INIT_HLIST_HEAD(&fll->hlist);
}
locks: move file_lock_list to a set of percpu hlist_heads and convert file_lock_lock to an lglock The file_lock_list is only used for /proc/locks. The vastly common case is for locks to be put onto the list and come off again, without ever being traversed. Help optimize for this use-case by moving to percpu hlist_head-s. At the same time, we can make the locking less contentious by moving to an lglock. When iterating over the lists for /proc/locks, we must take the global lock and then iterate over each CPU's list in turn. This change necessitates a new fl_link_cpu field to keep track of which CPU the entry is on. On x86_64 at least, this field is placed within an existing hole in the struct to avoid growing the size. Signed-off-by: Jeff Layton <jlayton@redhat.com> Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2013-06-21 19:58:22 +07:00
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
return 0;
}
core_initcall(filelock_init);
Reference in New Issue Copy Permalink