Documentation/filesystems/Locking

   1         The text below describes the locking rules for VFS-related methods.
   2 It is (believed to be) up-to-date. *Please*, if you change anything in
   3 prototypes or locking protocols - update this file. And update the relevant
   4 instances in the tree, don't leave that to maintainers of filesystems/devices/
   5 etc. At the very least, put the list of dubious cases in the end of this file.
   6 Don't turn it into log - maintainers of out-of-the-tree code are supposed to
   7 be able to use diff(1).
   8         Thing currently missing here: socket operations. Alexey?
   9
  10 --------------------------- dentry_operations --------------------------
  11 prototypes:
  12         int (*d_revalidate)(struct dentry *, unsigned int);
  13         int (*d_weak_revalidate)(struct dentry *, unsigned int);
  14         int (*d_hash)(const struct dentry *, struct qstr *);
  15         int (*d_compare)(const struct dentry *, const struct dentry *,
  16                         unsigned int, const char *, const struct qstr *);
  17         int (*d_delete)(struct dentry *);
  18         void (*d_release)(struct dentry *);
  19         void (*d_iput)(struct dentry *, struct inode *);
  20         char *(*d_dname)((struct dentry *dentry, char *buffer, int buflen);
  21         struct vfsmount *(*d_automount)(struct path *path);
  22         int (*d_manage)(struct dentry *, bool);
  23         struct dentry *(*d_real)(struct dentry *, const struct inode *,
  24                                  unsigned int);
  25
  26 locking rules:
  27                 rename_lock     ->d_lock        may block       rcu-walk
  28 d_revalidate:   no              no              yes (ref-walk)  maybe
  29 d_weak_revalidate:no            no              yes             no
  30 d_hash          no              no              no              maybe
  31 d_compare:      yes             no              no              maybe
  32 d_delete:       no              yes             no              no
  33 d_release:      no              no              yes             no
  34 d_prune:        no              yes             no              no
  35 d_iput:         no              no              yes             no
  36 d_dname:        no              no              no              no
  37 d_automount:    no              no              yes             no
  38 d_manage:       no              no              yes (ref-walk)  maybe
  39 d_real          no              no              yes             no
  40
  41 --------------------------- inode_operations ---------------------------
  42 prototypes:
  43         int (*create) (struct inode *,struct dentry *,umode_t, bool);
  44         struct dentry * (*lookup) (struct inode *,struct dentry *, unsigned int);
  45         int (*link) (struct dentry *,struct inode *,struct dentry *);
  46         int (*unlink) (struct inode *,struct dentry *);
  47         int (*symlink) (struct inode *,struct dentry *,const char *);
  48         int (*mkdir) (struct inode *,struct dentry *,umode_t);
  49         int (*rmdir) (struct inode *,struct dentry *);
  50         int (*mknod) (struct inode *,struct dentry *,umode_t,dev_t);
  51         int (*rename) (struct inode *, struct dentry *,
  52                         struct inode *, struct dentry *);
  53         int (*rename2) (struct inode *, struct dentry *,
  54                         struct inode *, struct dentry *, unsigned int);
  55         int (*readlink) (struct dentry *, char __user *,int);
  56         const char *(*get_link) (struct dentry *, struct inode *, void **);
  57         void (*truncate) (struct inode *);
  58         int (*permission) (struct inode *, int, unsigned int);
  59         int (*get_acl)(struct inode *, int);
  60         int (*setattr) (struct dentry *, struct iattr *);
  61         int (*getattr) (struct vfsmount *, struct dentry *, struct kstat *);
  62         int (*setxattr) (struct dentry *, const char *,const void *,size_t,int);
  63         ssize_t (*getxattr) (struct dentry *, const char *, void *, size_t);
  64         ssize_t (*listxattr) (struct dentry *, char *, size_t);
  65         int (*removexattr) (struct dentry *, const char *);
  66         int (*fiemap)(struct inode *, struct fiemap_extent_info *, u64 start, u64 len);
  67         void (*update_time)(struct inode *, struct timespec *, int);
  68         int (*atomic_open)(struct inode *, struct dentry *,
  69                                 struct file *, unsigned open_flag,
  70                                 umode_t create_mode, int *opened);
  71         int (*tmpfile) (struct inode *, struct dentry *, umode_t);
  72         int (*dentry_open)(struct dentry *, struct file *, const struct cred *);
  73
  74 locking rules:
  75         all may block
  76                 i_mutex(inode)
  77 lookup:         yes
  78 create:         yes
  79 link:           yes (both)
  80 mknod:          yes
  81 symlink:        yes
  82 mkdir:          yes
  83 unlink:         yes (both)
  84 rmdir:          yes (both)      (see below)
  85 rename:         yes (all)       (see below)
  86 rename2:        yes (all)       (see below)
  87 readlink:       no
  88 get_link:       no
  89 setattr:        yes
  90 permission:     no (may not block if called in rcu-walk mode)
  91 get_acl:        no
  92 getattr:        no
  93 setxattr:       yes
  94 getxattr:       no
  95 listxattr:      no
  96 removexattr:    yes
  97 fiemap:         no
  98 update_time:    no
  99 atomic_open:    yes
 100 tmpfile:        no
 101 dentry_open:    no
 102
 103         Additionally, ->rmdir(), ->unlink() and ->rename() have ->i_mutex on
 104 victim.
 105         cross-directory ->rename() and rename2() has (per-superblock)
 106 ->s_vfs_rename_sem.
 107
 108 See Documentation/filesystems/directory-locking for more detailed discussion
 109 of the locking scheme for directory operations.
 110
 111 --------------------------- super_operations ---------------------------
 112 prototypes:
 113         struct inode *(*alloc_inode)(struct super_block *sb);
 114         void (*destroy_inode)(struct inode *);
 115         void (*dirty_inode) (struct inode *, int flags);
 116         int (*write_inode) (struct inode *, struct writeback_control *wbc);
 117         int (*drop_inode) (struct inode *);
 118         void (*evict_inode) (struct inode *);
 119         void (*put_super) (struct super_block *);
 120         int (*sync_fs)(struct super_block *sb, int wait);
 121         int (*freeze_fs) (struct super_block *);
 122         int (*unfreeze_fs) (struct super_block *);
 123         int (*statfs) (struct dentry *, struct kstatfs *);
 124         int (*remount_fs) (struct super_block *, int *, char *);
 125         void (*umount_begin) (struct super_block *);
 126         int (*show_options)(struct seq_file *, struct dentry *);
 127         ssize_t (*quota_read)(struct super_block *, int, char *, size_t, loff_t);
 128         ssize_t (*quota_write)(struct super_block *, int, const char *, size_t, loff_t);
 129         int (*bdev_try_to_free_page)(struct super_block*, struct page*, gfp_t);
 130
 131 locking rules:
 132         All may block [not true, see below]
 133                         s_umount
 134 alloc_inode:
 135 destroy_inode:
 136 dirty_inode:
 137 write_inode:
 138 drop_inode:                             !!!inode->i_lock!!!
 139 evict_inode:
 140 put_super:              write
 141 sync_fs:                read
 142 freeze_fs:              write
 143 unfreeze_fs:            write
 144 statfs:                 maybe(read)     (see below)
 145 remount_fs:             write
 146 umount_begin:           no
 147 show_options:           no              (namespace_sem)
 148 quota_read:             no              (see below)
 149 quota_write:            no              (see below)
 150 bdev_try_to_free_page:  no              (see below)
 151
 152 ->statfs() has s_umount (shared) when called by ustat(2) (native or
 153 compat), but that's an accident of bad API; s_umount is used to pin
 154 the superblock down when we only have dev_t given us by userland to
 155 identify the superblock.  Everything else (statfs(), fstatfs(), etc.)
 156 doesn't hold it when calling ->statfs() - superblock is pinned down
 157 by resolving the pathname passed to syscall.
 158 ->quota_read() and ->quota_write() functions are both guaranteed to
 159 be the only ones operating on the quota file by the quota code (via
 160 dqio_sem) (unless an admin really wants to screw up something and
 161 writes to quota files with quotas on). For other details about locking
 162 see also dquot_operations section.
 163 ->bdev_try_to_free_page is called from the ->releasepage handler of
 164 the block device inode.  See there for more details.
 165
 166 --------------------------- file_system_type ---------------------------
 167 prototypes:
 168         struct dentry *(*mount) (struct file_system_type *, int,
 169                        const char *, void *);
 170         void (*kill_sb) (struct super_block *);
 171 locking rules:
 172                 may block
 173 mount           yes
 174 kill_sb         yes
 175
 176 ->mount() returns ERR_PTR or the root dentry; its superblock should be locked
 177 on return.
 178 ->kill_sb() takes a write-locked superblock, does all shutdown work on it,
 179 unlocks and drops the reference.
 180
 181 --------------------------- address_space_operations --------------------------
 182 prototypes:
 183         int (*writepage)(struct page *page, struct writeback_control *wbc);
 184         int (*readpage)(struct file *, struct page *);
 185         int (*sync_page)(struct page *);
 186         int (*writepages)(struct address_space *, struct writeback_control *);
 187         int (*set_page_dirty)(struct page *page);
 188         int (*readpages)(struct file *filp, struct address_space *mapping,
 189                         struct list_head *pages, unsigned nr_pages);
 190         int (*write_begin)(struct file *, struct address_space *mapping,
 191                                 loff_t pos, unsigned len, unsigned flags,
 192                                 struct page **pagep, void **fsdata);
 193         int (*write_end)(struct file *, struct address_space *mapping,
 194                                 loff_t pos, unsigned len, unsigned copied,
 195                                 struct page *page, void *fsdata);
 196         sector_t (*bmap)(struct address_space *, sector_t);
 197         void (*invalidatepage) (struct page *, unsigned int, unsigned int);
 198         int (*releasepage) (struct page *, int);
 199         void (*freepage)(struct page *);
 200         int (*direct_IO)(struct kiocb *, struct iov_iter *iter);
 201         int (*migratepage)(struct address_space *, struct page *, struct page *);
 202         int (*launder_page)(struct page *);
 203         int (*is_partially_uptodate)(struct page *, unsigned long, unsigned long);
 204         int (*error_remove_page)(struct address_space *, struct page *);
 205         int (*swap_activate)(struct file *);
 206         int (*swap_deactivate)(struct file *);
 207
 208 locking rules:
 209         All except set_page_dirty and freepage may block
 210
 211                         PageLocked(page)        i_mutex
 212 writepage:              yes, unlocks (see below)
 213 readpage:               yes, unlocks
 214 sync_page:              maybe
 215 writepages:
 216 set_page_dirty          no
 217 readpages:
 218 write_begin:            locks the page          yes
 219 write_end:              yes, unlocks            yes
 220 bmap:
 221 invalidatepage:         yes
 222 releasepage:            yes
 223 freepage:               yes
 224 direct_IO:
 225 migratepage:            yes (both)
 226 launder_page:           yes
 227 is_partially_uptodate:  yes
 228 error_remove_page:      yes
 229 swap_activate:          no
 230 swap_deactivate:        no
 231
 232         ->write_begin(), ->write_end(), ->sync_page() and ->readpage()
 233 may be called from the request handler (/dev/loop).
 234
 235         ->readpage() unlocks the page, either synchronously or via I/O
 236 completion.
 237
 238         ->readpages() populates the pagecache with the passed pages and starts
 239 I/O against them.  They come unlocked upon I/O completion.
 240
 241         ->writepage() is used for two purposes: for "memory cleansing" and for
 242 "sync".  These are quite different operations and the behaviour may differ
 243 depending upon the mode.
 244
 245 If writepage is called for sync (wbc->sync_mode != WBC_SYNC_NONE) then
 246 it *must* start I/O against the page, even if that would involve
 247 blocking on in-progress I/O.
 248
 249 If writepage is called for memory cleansing (sync_mode ==
 250 WBC_SYNC_NONE) then its role is to get as much writeout underway as
 251 possible.  So writepage should try to avoid blocking against
 252 currently-in-progress I/O.
 253
 254 If the filesystem is not called for "sync" and it determines that it
 255 would need to block against in-progress I/O to be able to start new I/O
 256 against the page the filesystem should redirty the page with
 257 redirty_page_for_writepage(), then unlock the page and return zero.
 258 This may also be done to avoid internal deadlocks, but rarely.
 259
 260 If the filesystem is called for sync then it must wait on any
 261 in-progress I/O and then start new I/O.
 262
 263 The filesystem should unlock the page synchronously, before returning to the
 264 caller, unless ->writepage() returns special WRITEPAGE_ACTIVATE
 265 value. WRITEPAGE_ACTIVATE means that page cannot really be written out
 266 currently, and VM should stop calling ->writepage() on this page for some
 267 time. VM does this by moving page to the head of the active list, hence the
 268 name.
 269
 270 Unless the filesystem is going to redirty_page_for_writepage(), unlock the page
 271 and return zero, writepage *must* run set_page_writeback() against the page,
 272 followed by unlocking it.  Once set_page_writeback() has been run against the
 273 page, write I/O can be submitted and the write I/O completion handler must run
 274 end_page_writeback() once the I/O is complete.  If no I/O is submitted, the
 275 filesystem must run end_page_writeback() against the page before returning from
 276 writepage.
 277
 278 That is: after 2.5.12, pages which are under writeout are *not* locked.  Note,
 279 if the filesystem needs the page to be locked during writeout, that is ok, too,
 280 the page is allowed to be unlocked at any point in time between the calls to
 281 set_page_writeback() and end_page_writeback().
 282
 283 Note, failure to run either redirty_page_for_writepage() or the combination of
 284 set_page_writeback()/end_page_writeback() on a page submitted to writepage
 285 will leave the page itself marked clean but it will be tagged as dirty in the
 286 radix tree.  This incoherency can lead to all sorts of hard-to-debug problems
 287 in the filesystem like having dirty inodes at umount and losing written data.
 288
 289         ->sync_page() locking rules are not well-defined - usually it is called
 290 with lock on page, but that is not guaranteed. Considering the currently
 291 existing instances of this method ->sync_page() itself doesn't look
 292 well-defined...
 293
 294         ->writepages() is used for periodic writeback and for syscall-initiated
 295 sync operations.  The address_space should start I/O against at least
 296 *nr_to_write pages.  *nr_to_write must be decremented for each page which is
 297 written.  The address_space implementation may write more (or less) pages
 298 than *nr_to_write asks for, but it should try to be reasonably close.  If
 299 nr_to_write is NULL, all dirty pages must be written.
 300
 301 writepages should _only_ write pages which are present on
 302 mapping->io_pages.
 303
 304         ->set_page_dirty() is called from various places in the kernel
 305 when the target page is marked as needing writeback.  It may be called
 306 under spinlock (it cannot block) and is sometimes called with the page
 307 not locked.
 308
 309         ->bmap() is currently used by legacy ioctl() (FIBMAP) provided by some
 310 filesystems and by the swapper. The latter will eventually go away.  Please,
 311 keep it that way and don't breed new callers.
 312
 313         ->invalidatepage() is called when the filesystem must attempt to drop
 314 some or all of the buffers from the page when it is being truncated. It
 315 returns zero on success. If ->invalidatepage is zero, the kernel uses
 316 block_invalidatepage() instead.
 317
 318         ->releasepage() is called when the kernel is about to try to drop the
 319 buffers from the page in preparation for freeing it.  It returns zero to
 320 indicate that the buffers are (or may be) freeable.  If ->releasepage is zero,
 321 the kernel assumes that the fs has no private interest in the buffers.
 322
 323         ->freepage() is called when the kernel is done dropping the page
 324 from the page cache.
 325
 326         ->launder_page() may be called prior to releasing a page if
 327 it is still found to be dirty. It returns zero if the page was successfully
 328 cleaned, or an error value if not. Note that in order to prevent the page
 329 getting mapped back in and redirtied, it needs to be kept locked
 330 across the entire operation.
 331
 332         ->swap_activate will be called with a non-zero argument on
 333 files backing (non block device backed) swapfiles. A return value
 334 of zero indicates success, in which case this file can be used for
 335 backing swapspace. The swapspace operations will be proxied to the
 336 address space operations.
 337
 338         ->swap_deactivate() will be called in the sys_swapoff()
 339 path after ->swap_activate() returned success.
 340
 341 ----------------------- file_lock_operations ------------------------------
 342 prototypes:
 343         void (*fl_copy_lock)(struct file_lock *, struct file_lock *);
 344         void (*fl_release_private)(struct file_lock *);
 345
 346
 347 locking rules:
 348                         inode->i_lock   may block
 349 fl_copy_lock:           yes             no
 350 fl_release_private:     maybe           maybe[1]
 351
 352 [1]:    ->fl_release_private for flock or POSIX locks is currently allowed
 353 to block. Leases however can still be freed while the i_lock is held and
 354 so fl_release_private called on a lease should not block.
 355
 356 ----------------------- lock_manager_operations ---------------------------
 357 prototypes:
 358         int (*lm_compare_owner)(struct file_lock *, struct file_lock *);
 359         unsigned long (*lm_owner_key)(struct file_lock *);
 360         void (*lm_notify)(struct file_lock *);  /* unblock callback */
 361         int (*lm_grant)(struct file_lock *, struct file_lock *, int);
 362         void (*lm_break)(struct file_lock *); /* break_lease callback */
 363         int (*lm_change)(struct file_lock **, int);
 364
 365 locking rules:
 366
 367                         inode->i_lock   blocked_lock_lock       may block
 368 lm_compare_owner:       yes[1]          maybe                   no
 369 lm_owner_key            yes[1]          yes                     no
 370 lm_notify:              yes             yes                     no
 371 lm_grant:               no              no                      no
 372 lm_break:               yes             no                      no
 373 lm_change               yes             no                      no
 374
 375 [1]:    ->lm_compare_owner and ->lm_owner_key are generally called with
 376 *an* inode->i_lock held. It may not be the i_lock of the inode
 377 associated with either file_lock argument! This is the case with deadlock
 378 detection, since the code has to chase down the owners of locks that may
 379 be entirely unrelated to the one on which the lock is being acquired.
 380 For deadlock detection however, the blocked_lock_lock is also held. The
 381 fact that these locks are held ensures that the file_locks do not
 382 disappear out from under you while doing the comparison or generating an
 383 owner key.
 384
 385 --------------------------- buffer_head -----------------------------------
 386 prototypes:
 387         void (*b_end_io)(struct buffer_head *bh, int uptodate);
 388
 389 locking rules:
 390         called from interrupts. In other words, extreme care is needed here.
 391 bh is locked, but that's all warranties we have here. Currently only RAID1,
 392 highmem, fs/buffer.c, and fs/ntfs/aops.c are providing these. Block devices
 393 call this method upon the IO completion.
 394
 395 --------------------------- block_device_operations -----------------------
 396 prototypes:
 397         int (*open) (struct block_device *, fmode_t);
 398         int (*release) (struct gendisk *, fmode_t);
 399         int (*ioctl) (struct block_device *, fmode_t, unsigned, unsigned long);
 400         int (*compat_ioctl) (struct block_device *, fmode_t, unsigned, unsigned long);
 401         int (*direct_access) (struct block_device *, sector_t, void __pmem **,
 402                                 unsigned long *);
 403         int (*media_changed) (struct gendisk *);
 404         void (*unlock_native_capacity) (struct gendisk *);
 405         int (*revalidate_disk) (struct gendisk *);
 406         int (*getgeo)(struct block_device *, struct hd_geometry *);
 407         void (*swap_slot_free_notify) (struct block_device *, unsigned long);
 408
 409 locking rules:
 410                         bd_mutex
 411 open:                   yes
 412 release:                yes
 413 ioctl:                  no
 414 compat_ioctl:           no
 415 direct_access:          no
 416 media_changed:          no
 417 unlock_native_capacity: no
 418 revalidate_disk:        no
 419 getgeo:                 no
 420 swap_slot_free_notify:  no      (see below)
 421
 422 media_changed, unlock_native_capacity and revalidate_disk are called only from
 423 check_disk_change().
 424
 425 swap_slot_free_notify is called with swap_lock and sometimes the page lock
 426 held.
 427
 428
 429 --------------------------- file_operations -------------------------------
 430 prototypes:
 431         loff_t (*llseek) (struct file *, loff_t, int);
 432         ssize_t (*read) (struct file *, char __user *, size_t, loff_t *);
 433         ssize_t (*write) (struct file *, const char __user *, size_t, loff_t *);
 434         ssize_t (*read_iter) (struct kiocb *, struct iov_iter *);
 435         ssize_t (*write_iter) (struct kiocb *, struct iov_iter *);
 436         int (*iterate) (struct file *, struct dir_context *);
 437         unsigned int (*poll) (struct file *, struct poll_table_struct *);
 438         long (*unlocked_ioctl) (struct file *, unsigned int, unsigned long);
 439         long (*compat_ioctl) (struct file *, unsigned int, unsigned long);
 440         int (*mmap) (struct file *, struct vm_area_struct *);
 441         int (*open) (struct inode *, struct file *);
 442         int (*flush) (struct file *);
 443         int (*release) (struct inode *, struct file *);
 444         int (*fsync) (struct file *, loff_t start, loff_t end, int datasync);
 445         int (*aio_fsync) (struct kiocb *, int datasync);
 446         int (*fasync) (int, struct file *, int);
 447         int (*lock) (struct file *, int, struct file_lock *);
 448         ssize_t (*readv) (struct file *, const struct iovec *, unsigned long,
 449                         loff_t *);
 450         ssize_t (*writev) (struct file *, const struct iovec *, unsigned long,
 451                         loff_t *);
 452         ssize_t (*sendfile) (struct file *, loff_t *, size_t, read_actor_t,
 453                         void __user *);
 454         ssize_t (*sendpage) (struct file *, struct page *, int, size_t,
 455                         loff_t *, int);
 456         unsigned long (*get_unmapped_area)(struct file *, unsigned long,
 457                         unsigned long, unsigned long, unsigned long);
 458         int (*check_flags)(int);
 459         int (*flock) (struct file *, int, struct file_lock *);
 460         ssize_t (*splice_write)(struct pipe_inode_info *, struct file *, loff_t *,
 461                         size_t, unsigned int);
 462         ssize_t (*splice_read)(struct file *, loff_t *, struct pipe_inode_info *,
 463                         size_t, unsigned int);
 464         int (*setlease)(struct file *, long, struct file_lock **, void **);
 465         long (*fallocate)(struct file *, int, loff_t, loff_t);
 466 };
 467
 468 locking rules:
 469         All may block.
 470
 471 ->llseek() locking has moved from llseek to the individual llseek
 472 implementations.  If your fs is not using generic_file_llseek, you
 473 need to acquire and release the appropriate locks in your ->llseek().
 474 For many filesystems, it is probably safe to acquire the inode
 475 mutex or just to use i_size_read() instead.
 476 Note: this does not protect the file->f_pos against concurrent modifications
 477 since this is something the userspace has to take care about.
 478
 479 ->fasync() is responsible for maintaining the FASYNC bit in filp->f_flags.
 480 Most instances call fasync_helper(), which does that maintenance, so it's
 481 not normally something one needs to worry about.  Return values > 0 will be
 482 mapped to zero in the VFS layer.
 483
 484 ->readdir() and ->ioctl() on directories must be changed. Ideally we would
 485 move ->readdir() to inode_operations and use a separate method for directory
 486 ->ioctl() or kill the latter completely. One of the problems is that for
 487 anything that resembles union-mount we won't have a struct file for all
 488 components. And there are other reasons why the current interface is a mess...
 489
 490 ->read on directories probably must go away - we should just enforce -EISDIR
 491 in sys_read() and friends.
 492
 493 ->setlease operations should call generic_setlease() before or after setting
 494 the lease within the individual filesystem to record the result of the
 495 operation
 496
 497 --------------------------- dquot_operations -------------------------------
 498 prototypes:
 499         int (*write_dquot) (struct dquot *);
 500         int (*acquire_dquot) (struct dquot *);
 501         int (*release_dquot) (struct dquot *);
 502         int (*mark_dirty) (struct dquot *);
 503         int (*write_info) (struct super_block *, int);
 504
 505 These operations are intended to be more or less wrapping functions that ensure
 506 a proper locking wrt the filesystem and call the generic quota operations.
 507
 508 What filesystem should expect from the generic quota functions:
 509
 510                 FS recursion    Held locks when called
 511 write_dquot:    yes             dqonoff_sem or dqptr_sem
 512 acquire_dquot:  yes             dqonoff_sem or dqptr_sem
 513 release_dquot:  yes             dqonoff_sem or dqptr_sem
 514 mark_dirty:     no              -
 515 write_info:     yes             dqonoff_sem
 516
 517 FS recursion means calling ->quota_read() and ->quota_write() from superblock
 518 operations.
 519
 520 More details about quota locking can be found in fs/dquot.c.
 521
 522 --------------------------- vm_operations_struct -----------------------------
 523 prototypes:
 524         void (*open)(struct vm_area_struct*);
 525         void (*close)(struct vm_area_struct*);
 526         int (*fault)(struct vm_area_struct*, struct vm_fault *);
 527         int (*page_mkwrite)(struct vm_area_struct *, struct vm_fault *);
 528         int (*pfn_mkwrite)(struct vm_area_struct *, struct vm_fault *);
 529         int (*access)(struct vm_area_struct *, unsigned long, void*, int, int);
 530
 531 locking rules:
 532                 mmap_sem        PageLocked(page)
 533 open:           yes
 534 close:          yes
 535 fault:          yes             can return with page locked
 536 map_pages:      yes
 537 page_mkwrite:   yes             can return with page locked
 538 pfn_mkwrite:    yes
 539 access:         yes
 540
 541         ->fault() is called when a previously not present pte is about
 542 to be faulted in. The filesystem must find and return the page associated
 543 with the passed in "pgoff" in the vm_fault structure. If it is possible that
 544 the page may be truncated and/or invalidated, then the filesystem must lock
 545 the page, then ensure it is not already truncated (the page lock will block
 546 subsequent truncate), and then return with VM_FAULT_LOCKED, and the page
 547 locked. The VM will unlock the page.
 548
 549         ->map_pages() is called when VM asks to map easy accessible pages.
 550 Filesystem should find and map pages associated with offsets from "pgoff"
 551 till "max_pgoff". ->map_pages() is called with page table locked and must
 552 not block.  If it's not possible to reach a page without blocking,
 553 filesystem should skip it. Filesystem should use do_set_pte() to setup
 554 page table entry. Pointer to entry associated with offset "pgoff" is
 555 passed in "pte" field in vm_fault structure. Pointers to entries for other
 556 offsets should be calculated relative to "pte".
 557
 558         ->page_mkwrite() is called when a previously read-only pte is
 559 about to become writeable. The filesystem again must ensure that there are
 560 no truncate/invalidate races, and then return with the page locked. If
 561 the page has been truncated, the filesystem should not look up a new page
 562 like the ->fault() handler, but simply return with VM_FAULT_NOPAGE, which
 563 will cause the VM to retry the fault.
 564
 565         ->pfn_mkwrite() is the same as page_mkwrite but when the pte is
 566 VM_PFNMAP or VM_MIXEDMAP with a page-less entry. Expected return is
 567 VM_FAULT_NOPAGE. Or one of the VM_FAULT_ERROR types. The default behavior
 568 after this call is to make the pte read-write, unless pfn_mkwrite returns
 569 an error.
 570
 571         ->access() is called when get_user_pages() fails in
 572 access_process_vm(), typically used to debug a process through
 573 /proc/pid/mem or ptrace.  This function is needed only for
 574 VM_IO | VM_PFNMAP VMAs.
 575
 576 ================================================================================
 577                         Dubious stuff
 578
 579 (if you break something or notice that it is broken and do not fix it yourself
 580 - at least put it here)