1	The text below describes the locking rules for VFS-related methods.
  2It is (believed to be) up-to-date. *Please*, if you change anything in
  3prototypes or locking protocols - update this file. And update the relevant
  4instances in the tree, don't leave that to maintainers of filesystems/devices/
  5etc. At the very least, put the list of dubious cases in the end of this file.
  6Don't turn it into log - maintainers of out-of-the-tree code are supposed to
  7be able to use diff(1).
  8	Thing currently missing here: socket operations. Alexey?
  9
 10--------------------------- dentry_operations --------------------------
 11prototypes:
 12	int (*d_revalidate)(struct dentry *, struct nameidata *);
 13	int (*d_hash)(const struct dentry *, const struct inode *,
 14			struct qstr *);
 15	int (*d_compare)(const struct dentry *, const struct inode *,
 16			const struct dentry *, const struct inode *,
 17			unsigned int, const char *, const struct qstr *);
 18	int (*d_delete)(struct dentry *);
 19	void (*d_release)(struct dentry *);
 20	void (*d_iput)(struct dentry *, struct inode *);
 21	char *(*d_dname)((struct dentry *dentry, char *buffer, int buflen);
 22	struct vfsmount *(*d_automount)(struct path *path);
 23	int (*d_manage)(struct dentry *, bool);
 24
 25locking rules:
 26		rename_lock	->d_lock	may block	rcu-walk
 27d_revalidate:	no		no		yes (ref-walk)	maybe
 28d_hash		no		no		no		maybe
 29d_compare:	yes		no		no		maybe
 30d_delete:	no		yes		no		no
 31d_release:	no		no		yes		no
 32d_iput:		no		no		yes		no
 33d_dname:	no		no		no		no
 34d_automount:	no		no		yes		no
 35d_manage:	no		no		yes (ref-walk)	maybe
 36
 37--------------------------- inode_operations --------------------------- 
 38prototypes:
 39	int (*create) (struct inode *,struct dentry *,int, struct nameidata *);
 40	struct dentry * (*lookup) (struct inode *,struct dentry *, struct nameid
 41ata *);
 42	int (*link) (struct dentry *,struct inode *,struct dentry *);
 43	int (*unlink) (struct inode *,struct dentry *);
 44	int (*symlink) (struct inode *,struct dentry *,const char *);
 45	int (*mkdir) (struct inode *,struct dentry *,int);
 46	int (*rmdir) (struct inode *,struct dentry *);
 47	int (*mknod) (struct inode *,struct dentry *,int,dev_t);
 48	int (*rename) (struct inode *, struct dentry *,
 49			struct inode *, struct dentry *);
 50	int (*readlink) (struct dentry *, char __user *,int);
 51	void * (*follow_link) (struct dentry *, struct nameidata *);
 52	void (*put_link) (struct dentry *, struct nameidata *, void *);
 53	void (*truncate) (struct inode *);
 54	int (*permission) (struct inode *, int, unsigned int);
 55	int (*get_acl)(struct inode *, int);
 56	int (*setattr) (struct dentry *, struct iattr *);
 57	int (*getattr) (struct vfsmount *, struct dentry *, struct kstat *);
 58	int (*setxattr) (struct dentry *, const char *,const void *,size_t,int);
 59	ssize_t (*getxattr) (struct dentry *, const char *, void *, size_t);
 60	ssize_t (*listxattr) (struct dentry *, char *, size_t);
 61	int (*removexattr) (struct dentry *, const char *);
 62	void (*truncate_range)(struct inode *, loff_t, loff_t);
 63	int (*fiemap)(struct inode *, struct fiemap_extent_info *, u64 start, u64 len);
 64
 65locking rules:
 66	all may block
 67		i_mutex(inode)
 68lookup:		yes
 69create:		yes
 70link:		yes (both)
 71mknod:		yes
 72symlink:	yes
 73mkdir:		yes
 74unlink:		yes (both)
 75rmdir:		yes (both)	(see below)
 76rename:		yes (all)	(see below)
 77readlink:	no
 78follow_link:	no
 79put_link:	no
 80truncate:	yes		(see below)
 81setattr:	yes
 82permission:	no (may not block if called in rcu-walk mode)
 83get_acl:	no
 84getattr:	no
 85setxattr:	yes
 86getxattr:	no
 87listxattr:	no
 88removexattr:	yes
 89truncate_range:	yes
 90fiemap:		no
 91	Additionally, ->rmdir(), ->unlink() and ->rename() have ->i_mutex on
 92victim.
 93	cross-directory ->rename() has (per-superblock) ->s_vfs_rename_sem.
 94	->truncate() is never called directly - it's a callback, not a
 95method. It's called by vmtruncate() - deprecated library function used by
 96->setattr(). Locking information above applies to that call (i.e. is
 97inherited from ->setattr() - vmtruncate() is used when ATTR_SIZE had been
 98passed).
 99
100See Documentation/filesystems/directory-locking for more detailed discussion
101of the locking scheme for directory operations.
102
103--------------------------- super_operations ---------------------------
104prototypes:
105	struct inode *(*alloc_inode)(struct super_block *sb);
106	void (*destroy_inode)(struct inode *);
107	void (*dirty_inode) (struct inode *, int flags);
108	int (*write_inode) (struct inode *, struct writeback_control *wbc);
109	int (*drop_inode) (struct inode *);
110	void (*evict_inode) (struct inode *);
111	void (*put_super) (struct super_block *);
112	void (*write_super) (struct super_block *);
113	int (*sync_fs)(struct super_block *sb, int wait);
114	int (*freeze_fs) (struct super_block *);
115	int (*unfreeze_fs) (struct super_block *);
116	int (*statfs) (struct dentry *, struct kstatfs *);
117	int (*remount_fs) (struct super_block *, int *, char *);
118	void (*umount_begin) (struct super_block *);
119	int (*show_options)(struct seq_file *, struct vfsmount *);
120	ssize_t (*quota_read)(struct super_block *, int, char *, size_t, loff_t);
121	ssize_t (*quota_write)(struct super_block *, int, const char *, size_t, loff_t);
122	int (*bdev_try_to_free_page)(struct super_block*, struct page*, gfp_t);
123
124locking rules:
125	All may block [not true, see below]
126			s_umount
127alloc_inode:
128destroy_inode:
129dirty_inode:
130write_inode:
131drop_inode:				!!!inode->i_lock!!!
132evict_inode:
133put_super:		write
134write_super:		read
135sync_fs:		read
136freeze_fs:		read
137unfreeze_fs:		read
138statfs:			maybe(read)	(see below)
139remount_fs:		write
140umount_begin:		no
141show_options:		no		(namespace_sem)
142quota_read:		no		(see below)
143quota_write:		no		(see below)
144bdev_try_to_free_page:	no		(see below)
145
146->statfs() has s_umount (shared) when called by ustat(2) (native or
147compat), but that's an accident of bad API; s_umount is used to pin
148the superblock down when we only have dev_t given us by userland to
149identify the superblock.  Everything else (statfs(), fstatfs(), etc.)
150doesn't hold it when calling ->statfs() - superblock is pinned down
151by resolving the pathname passed to syscall.
152->quota_read() and ->quota_write() functions are both guaranteed to
153be the only ones operating on the quota file by the quota code (via
154dqio_sem) (unless an admin really wants to screw up something and
155writes to quota files with quotas on). For other details about locking
156see also dquot_operations section.
157->bdev_try_to_free_page is called from the ->releasepage handler of
158the block device inode.  See there for more details.
159
160--------------------------- file_system_type ---------------------------
161prototypes:
162	int (*get_sb) (struct file_system_type *, int,
163		       const char *, void *, struct vfsmount *);
164	struct dentry *(*mount) (struct file_system_type *, int,
165		       const char *, void *);
166	void (*kill_sb) (struct super_block *);
167locking rules:
168		may block
169mount		yes
170kill_sb		yes
171
172->mount() returns ERR_PTR or the root dentry; its superblock should be locked
173on return.
174->kill_sb() takes a write-locked superblock, does all shutdown work on it,
175unlocks and drops the reference.
176
177--------------------------- address_space_operations --------------------------
178prototypes:
179	int (*writepage)(struct page *page, struct writeback_control *wbc);
180	int (*readpage)(struct file *, struct page *);
181	int (*sync_page)(struct page *);
182	int (*writepages)(struct address_space *, struct writeback_control *);
183	int (*set_page_dirty)(struct page *page);
184	int (*readpages)(struct file *filp, struct address_space *mapping,
185			struct list_head *pages, unsigned nr_pages);
186	int (*write_begin)(struct file *, struct address_space *mapping,
187				loff_t pos, unsigned len, unsigned flags,
188				struct page **pagep, void **fsdata);
189	int (*write_end)(struct file *, struct address_space *mapping,
190				loff_t pos, unsigned len, unsigned copied,
191				struct page *page, void *fsdata);
192	sector_t (*bmap)(struct address_space *, sector_t);
193	int (*invalidatepage) (struct page *, unsigned long);
194	int (*releasepage) (struct page *, int);
195	void (*freepage)(struct page *);
196	int (*direct_IO)(int, struct kiocb *, const struct iovec *iov,
197			loff_t offset, unsigned long nr_segs);
198	int (*get_xip_mem)(struct address_space *, pgoff_t, int, void **,
199				unsigned long *);
200	int (*migratepage)(struct address_space *, struct page *, struct page *);
201	int (*launder_page)(struct page *);
202	int (*is_partially_uptodate)(struct page *, read_descriptor_t *, unsigned long);
203	int (*error_remove_page)(struct address_space *, struct page *);
204
205locking rules:
206	All except set_page_dirty and freepage may block
207
208			PageLocked(page)	i_mutex
209writepage:		yes, unlocks (see below)
210readpage:		yes, unlocks
211sync_page:		maybe
212writepages:
213set_page_dirty		no
214readpages:
215write_begin:		locks the page		yes
216write_end:		yes, unlocks		yes
217bmap:
218invalidatepage:		yes
219releasepage:		yes
220freepage:		yes
221direct_IO:
222get_xip_mem:					maybe
223migratepage:		yes (both)
224launder_page:		yes
225is_partially_uptodate:	yes
226error_remove_page:	yes
227
228	->write_begin(), ->write_end(), ->sync_page() and ->readpage()
229may be called from the request handler (/dev/loop).
230
231	->readpage() unlocks the page, either synchronously or via I/O
232completion.
233
234	->readpages() populates the pagecache with the passed pages and starts
235I/O against them.  They come unlocked upon I/O completion.
236
237	->writepage() is used for two purposes: for "memory cleansing" and for
238"sync".  These are quite different operations and the behaviour may differ
239depending upon the mode.
240
241If writepage is called for sync (wbc->sync_mode != WBC_SYNC_NONE) then
242it *must* start I/O against the page, even if that would involve
243blocking on in-progress I/O.
244
245If writepage is called for memory cleansing (sync_mode ==
246WBC_SYNC_NONE) then its role is to get as much writeout underway as
247possible.  So writepage should try to avoid blocking against
248currently-in-progress I/O.
249
250If the filesystem is not called for "sync" and it determines that it
251would need to block against in-progress I/O to be able to start new I/O
252against the page the filesystem should redirty the page with
253redirty_page_for_writepage(), then unlock the page and return zero.
254This may also be done to avoid internal deadlocks, but rarely.
255
256If the filesystem is called for sync then it must wait on any
257in-progress I/O and then start new I/O.
258
259The filesystem should unlock the page synchronously, before returning to the
260caller, unless ->writepage() returns special WRITEPAGE_ACTIVATE
261value. WRITEPAGE_ACTIVATE means that page cannot really be written out
262currently, and VM should stop calling ->writepage() on this page for some
263time. VM does this by moving page to the head of the active list, hence the
264name.
265
266Unless the filesystem is going to redirty_page_for_writepage(), unlock the page
267and return zero, writepage *must* run set_page_writeback() against the page,
268followed by unlocking it.  Once set_page_writeback() has been run against the
269page, write I/O can be submitted and the write I/O completion handler must run
270end_page_writeback() once the I/O is complete.  If no I/O is submitted, the
271filesystem must run end_page_writeback() against the page before returning from
272writepage.
273
274That is: after 2.5.12, pages which are under writeout are *not* locked.  Note,
275if the filesystem needs the page to be locked during writeout, that is ok, too,
276the page is allowed to be unlocked at any point in time between the calls to
277set_page_writeback() and end_page_writeback().
278
279Note, failure to run either redirty_page_for_writepage() or the combination of
280set_page_writeback()/end_page_writeback() on a page submitted to writepage
281will leave the page itself marked clean but it will be tagged as dirty in the
282radix tree.  This incoherency can lead to all sorts of hard-to-debug problems
283in the filesystem like having dirty inodes at umount and losing written data.
284
285	->sync_page() locking rules are not well-defined - usually it is called
286with lock on page, but that is not guaranteed. Considering the currently
287existing instances of this method ->sync_page() itself doesn't look
288well-defined...
289
290	->writepages() is used for periodic writeback and for syscall-initiated
291sync operations.  The address_space should start I/O against at least
292*nr_to_write pages.  *nr_to_write must be decremented for each page which is
293written.  The address_space implementation may write more (or less) pages
294than *nr_to_write asks for, but it should try to be reasonably close.  If
295nr_to_write is NULL, all dirty pages must be written.
296
297writepages should _only_ write pages which are present on
298mapping->io_pages.
299
300	->set_page_dirty() is called from various places in the kernel
301when the target page is marked as needing writeback.  It may be called
302under spinlock (it cannot block) and is sometimes called with the page
303not locked.
304
305	->bmap() is currently used by legacy ioctl() (FIBMAP) provided by some
306filesystems and by the swapper. The latter will eventually go away.  Please,
307keep it that way and don't breed new callers.
308
309	->invalidatepage() is called when the filesystem must attempt to drop
310some or all of the buffers from the page when it is being truncated.  It
311returns zero on success.  If ->invalidatepage is zero, the kernel uses
312block_invalidatepage() instead.
313
314	->releasepage() is called when the kernel is about to try to drop the
315buffers from the page in preparation for freeing it.  It returns zero to
316indicate that the buffers are (or may be) freeable.  If ->releasepage is zero,
317the kernel assumes that the fs has no private interest in the buffers.
318
319	->freepage() is called when the kernel is done dropping the page
320from the page cache.
321
322	->launder_page() may be called prior to releasing a page if
323it is still found to be dirty. It returns zero if the page was successfully
324cleaned, or an error value if not. Note that in order to prevent the page
325getting mapped back in and redirtied, it needs to be kept locked
326across the entire operation.
327
328----------------------- file_lock_operations ------------------------------
329prototypes:
330	void (*fl_copy_lock)(struct file_lock *, struct file_lock *);
331	void (*fl_release_private)(struct file_lock *);
332
333
334locking rules:
335			file_lock_lock	may block
336fl_copy_lock:		yes		no
337fl_release_private:	maybe		no
338
339----------------------- lock_manager_operations ---------------------------
340prototypes:
341	int (*lm_compare_owner)(struct file_lock *, struct file_lock *);
342	void (*lm_notify)(struct file_lock *);  /* unblock callback */
343	int (*lm_grant)(struct file_lock *, struct file_lock *, int);
344	void (*lm_release_private)(struct file_lock *);
345	void (*lm_break)(struct file_lock *); /* break_lease callback */
346	int (*lm_change)(struct file_lock **, int);
347
348locking rules:
349			file_lock_lock	may block
350lm_compare_owner:	yes		no
351lm_notify:		yes		no
352lm_grant:		no		no
353lm_release_private:	maybe		no
354lm_break:		yes		no
355lm_change		yes		no
356
357--------------------------- buffer_head -----------------------------------
358prototypes:
359	void (*b_end_io)(struct buffer_head *bh, int uptodate);
360
361locking rules:
362	called from interrupts. In other words, extreme care is needed here.
363bh is locked, but that's all warranties we have here. Currently only RAID1,
364highmem, fs/buffer.c, and fs/ntfs/aops.c are providing these. Block devices
365call this method upon the IO completion.
366
367--------------------------- block_device_operations -----------------------
368prototypes:
369	int (*open) (struct block_device *, fmode_t);
370	int (*release) (struct gendisk *, fmode_t);
371	int (*ioctl) (struct block_device *, fmode_t, unsigned, unsigned long);
372	int (*compat_ioctl) (struct block_device *, fmode_t, unsigned, unsigned long);
373	int (*direct_access) (struct block_device *, sector_t, void **, unsigned long *);
374	int (*media_changed) (struct gendisk *);
375	void (*unlock_native_capacity) (struct gendisk *);
376	int (*revalidate_disk) (struct gendisk *);
377	int (*getgeo)(struct block_device *, struct hd_geometry *);
378	void (*swap_slot_free_notify) (struct block_device *, unsigned long);
379
380locking rules:
381			bd_mutex
382open:			yes
383release:		yes
384ioctl:			no
385compat_ioctl:		no
386direct_access:		no
387media_changed:		no
388unlock_native_capacity:	no
389revalidate_disk:	no
390getgeo:			no
391swap_slot_free_notify:	no	(see below)
392
393media_changed, unlock_native_capacity and revalidate_disk are called only from
394check_disk_change().
395
396swap_slot_free_notify is called with swap_lock and sometimes the page lock
397held.
398
399
400--------------------------- file_operations -------------------------------
401prototypes:
402	loff_t (*llseek) (struct file *, loff_t, int);
403	ssize_t (*read) (struct file *, char __user *, size_t, loff_t *);
404	ssize_t (*write) (struct file *, const char __user *, size_t, loff_t *);
405	ssize_t (*aio_read) (struct kiocb *, const struct iovec *, unsigned long, loff_t);
406	ssize_t (*aio_write) (struct kiocb *, const struct iovec *, unsigned long, loff_t);
407	int (*readdir) (struct file *, void *, filldir_t);
408	unsigned int (*poll) (struct file *, struct poll_table_struct *);
409	long (*unlocked_ioctl) (struct file *, unsigned int, unsigned long);
410	long (*compat_ioctl) (struct file *, unsigned int, unsigned long);
411	int (*mmap) (struct file *, struct vm_area_struct *);
412	int (*open) (struct inode *, struct file *);
413	int (*flush) (struct file *);
414	int (*release) (struct inode *, struct file *);
415	int (*fsync) (struct file *, loff_t start, loff_t end, int datasync);
416	int (*aio_fsync) (struct kiocb *, int datasync);
417	int (*fasync) (int, struct file *, int);
418	int (*lock) (struct file *, int, struct file_lock *);
419	ssize_t (*readv) (struct file *, const struct iovec *, unsigned long,
420			loff_t *);
421	ssize_t (*writev) (struct file *, const struct iovec *, unsigned long,
422			loff_t *);
423	ssize_t (*sendfile) (struct file *, loff_t *, size_t, read_actor_t,
424			void __user *);
425	ssize_t (*sendpage) (struct file *, struct page *, int, size_t,
426			loff_t *, int);
427	unsigned long (*get_unmapped_area)(struct file *, unsigned long,
428			unsigned long, unsigned long, unsigned long);
429	int (*check_flags)(int);
430	int (*flock) (struct file *, int, struct file_lock *);
431	ssize_t (*splice_write)(struct pipe_inode_info *, struct file *, loff_t *,
432			size_t, unsigned int);
433	ssize_t (*splice_read)(struct file *, loff_t *, struct pipe_inode_info *,
434			size_t, unsigned int);
435	int (*setlease)(struct file *, long, struct file_lock **);
436	long (*fallocate)(struct file *, int, loff_t, loff_t);
437};
438
439locking rules:
440	All may block except for ->setlease.
441	No VFS locks held on entry except for ->setlease.
442
443->setlease has the file_list_lock held and must not sleep.
444
445->llseek() locking has moved from llseek to the individual llseek
446implementations.  If your fs is not using generic_file_llseek, you
447need to acquire and release the appropriate locks in your ->llseek().
448For many filesystems, it is probably safe to acquire the inode
449mutex or just to use i_size_read() instead.
450Note: this does not protect the file->f_pos against concurrent modifications
451since this is something the userspace has to take care about.
452
453->fasync() is responsible for maintaining the FASYNC bit in filp->f_flags.
454Most instances call fasync_helper(), which does that maintenance, so it's
455not normally something one needs to worry about.  Return values > 0 will be
456mapped to zero in the VFS layer.
457
458->readdir() and ->ioctl() on directories must be changed. Ideally we would
459move ->readdir() to inode_operations and use a separate method for directory
460->ioctl() or kill the latter completely. One of the problems is that for
461anything that resembles union-mount we won't have a struct file for all
462components. And there are other reasons why the current interface is a mess...
463
464->read on directories probably must go away - we should just enforce -EISDIR
465in sys_read() and friends.
466
467--------------------------- dquot_operations -------------------------------
468prototypes:
469	int (*write_dquot) (struct dquot *);
470	int (*acquire_dquot) (struct dquot *);
471	int (*release_dquot) (struct dquot *);
472	int (*mark_dirty) (struct dquot *);
473	int (*write_info) (struct super_block *, int);
474
475These operations are intended to be more or less wrapping functions that ensure
476a proper locking wrt the filesystem and call the generic quota operations.
477
478What filesystem should expect from the generic quota functions:
479
480		FS recursion	Held locks when called
481write_dquot:	yes		dqonoff_sem or dqptr_sem
482acquire_dquot:	yes		dqonoff_sem or dqptr_sem
483release_dquot:	yes		dqonoff_sem or dqptr_sem
484mark_dirty:	no		-
485write_info:	yes		dqonoff_sem
486
487FS recursion means calling ->quota_read() and ->quota_write() from superblock
488operations.
489
490More details about quota locking can be found in fs/dquot.c.
491
492--------------------------- vm_operations_struct -----------------------------
493prototypes:
494	void (*open)(struct vm_area_struct*);
495	void (*close)(struct vm_area_struct*);
496	int (*fault)(struct vm_area_struct*, struct vm_fault *);
497	int (*page_mkwrite)(struct vm_area_struct *, struct vm_fault *);
498	int (*access)(struct vm_area_struct *, unsigned long, void*, int, int);
499
500locking rules:
501		mmap_sem	PageLocked(page)
502open:		yes
503close:		yes
504fault:		yes		can return with page locked
505page_mkwrite:	yes		can return with page locked
506access:		yes
507
508	->fault() is called when a previously not present pte is about
509to be faulted in. The filesystem must find and return the page associated
510with the passed in "pgoff" in the vm_fault structure. If it is possible that
511the page may be truncated and/or invalidated, then the filesystem must lock
512the page, then ensure it is not already truncated (the page lock will block
513subsequent truncate), and then return with VM_FAULT_LOCKED, and the page
514locked. The VM will unlock the page.
515
516	->page_mkwrite() is called when a previously read-only pte is
517about to become writeable. The filesystem again must ensure that there are
518no truncate/invalidate races, and then return with the page locked. If
519the page has been truncated, the filesystem should not look up a new page
520like the ->fault() handler, but simply return with VM_FAULT_NOPAGE, which
521will cause the VM to retry the fault.
522
523	->access() is called when get_user_pages() fails in
524acces_process_vm(), typically used to debug a process through
525/proc/pid/mem or ptrace.  This function is needed only for
526VM_IO | VM_PFNMAP VMAs.
527
528================================================================================
529			Dubious stuff
530
531(if you break something or notice that it is broken and do not fix it yourself
532- at least put it here)