2 Overview of the Linux Virtual File System
4 Original author: Richard Gooch <rgooch@atnf.csiro.au>
6 Last updated on June 24, 2007.
8 Copyright (C) 1999 Richard Gooch
9 Copyright (C) 2005 Pekka Enberg
11 This file is released under the GPLv2.
17 The Virtual File System (also known as the Virtual Filesystem Switch)
18 is the software layer in the kernel that provides the filesystem
19 interface to userspace programs. It also provides an abstraction
20 within the kernel which allows different filesystem implementations to
23 VFS system calls open(2), stat(2), read(2), write(2), chmod(2) and so
24 on are called from a process context. Filesystem locking is described
25 in the document Documentation/filesystems/Locking.
28 Directory Entry Cache (dcache)
29 ------------------------------
31 The VFS implements the open(2), stat(2), chmod(2), and similar system
32 calls. The pathname argument that is passed to them is used by the VFS
33 to search through the directory entry cache (also known as the dentry
34 cache or dcache). This provides a very fast look-up mechanism to
35 translate a pathname (filename) into a specific dentry. Dentries live
36 in RAM and are never saved to disc: they exist only for performance.
38 The dentry cache is meant to be a view into your entire filespace. As
39 most computers cannot fit all dentries in the RAM at the same time,
40 some bits of the cache are missing. In order to resolve your pathname
41 into a dentry, the VFS may have to resort to creating dentries along
42 the way, and then loading the inode. This is done by looking up the
49 An individual dentry usually has a pointer to an inode. Inodes are
50 filesystem objects such as regular files, directories, FIFOs and other
51 beasts. They live either on the disc (for block device filesystems)
52 or in the memory (for pseudo filesystems). Inodes that live on the
53 disc are copied into the memory when required and changes to the inode
54 are written back to disc. A single inode can be pointed to by multiple
55 dentries (hard links, for example, do this).
57 To look up an inode requires that the VFS calls the lookup() method of
58 the parent directory inode. This method is installed by the specific
59 filesystem implementation that the inode lives in. Once the VFS has
60 the required dentry (and hence the inode), we can do all those boring
61 things like open(2) the file, or stat(2) it to peek at the inode
62 data. The stat(2) operation is fairly simple: once the VFS has the
63 dentry, it peeks at the inode data and passes some of it back to
70 Opening a file requires another operation: allocation of a file
71 structure (this is the kernel-side implementation of file
72 descriptors). The freshly allocated file structure is initialized with
73 a pointer to the dentry and a set of file operation member functions.
74 These are taken from the inode data. The open() file method is then
75 called so the specific filesystem implementation can do its work. You
76 can see that this is another switch performed by the VFS. The file
77 structure is placed into the file descriptor table for the process.
79 Reading, writing and closing files (and other assorted VFS operations)
80 is done by using the userspace file descriptor to grab the appropriate
81 file structure, and then calling the required file structure method to
82 do whatever is required. For as long as the file is open, it keeps the
83 dentry in use, which in turn means that the VFS inode is still in use.
86 Registering and Mounting a Filesystem
87 =====================================
89 To register and unregister a filesystem, use the following API
94 extern int register_filesystem(struct file_system_type *);
95 extern int unregister_filesystem(struct file_system_type *);
97 The passed struct file_system_type describes your filesystem. When a
98 request is made to mount a filesystem onto a directory in your namespace,
99 the VFS will call the appropriate mount() method for the specific
100 filesystem. New vfsmount referring to the tree returned by ->mount()
101 will be attached to the mountpoint, so that when pathname resolution
102 reaches the mountpoint it will jump into the root of that vfsmount.
104 You can see all filesystems that are registered to the kernel in the
105 file /proc/filesystems.
108 struct file_system_type
109 -----------------------
111 This describes the filesystem. As of kernel 2.6.39, the following
114 struct file_system_type {
117 struct dentry *(*mount) (struct file_system_type *, int,
118 const char *, void *);
119 void (*kill_sb) (struct super_block *);
120 struct module *owner;
121 struct file_system_type * next;
122 struct list_head fs_supers;
123 struct lock_class_key s_lock_key;
124 struct lock_class_key s_umount_key;
127 name: the name of the filesystem type, such as "ext2", "iso9660",
130 fs_flags: various flags (i.e. FS_REQUIRES_DEV, FS_NO_DCACHE, etc.)
132 mount: the method to call when a new instance of this
133 filesystem should be mounted
135 kill_sb: the method to call when an instance of this filesystem
138 owner: for internal VFS use: you should initialize this to THIS_MODULE in
141 next: for internal VFS use: you should initialize this to NULL
143 s_lock_key, s_umount_key: lockdep-specific
145 The mount() method has the following arguments:
147 struct file_system_type *fs_type: describes the filesystem, partly initialized
148 by the specific filesystem code
150 int flags: mount flags
152 const char *dev_name: the device name we are mounting.
154 void *data: arbitrary mount options, usually comes as an ASCII
155 string (see "Mount Options" section)
157 The mount() method must return the root dentry of the tree requested by
158 caller. An active reference to its superblock must be grabbed and the
159 superblock must be locked. On failure it should return ERR_PTR(error).
161 The arguments match those of mount(2) and their interpretation
162 depends on filesystem type. E.g. for block filesystems, dev_name is
163 interpreted as block device name, that device is opened and if it
164 contains a suitable filesystem image the method creates and initializes
165 struct super_block accordingly, returning its root dentry to caller.
167 ->mount() may choose to return a subtree of existing filesystem - it
168 doesn't have to create a new one. The main result from the caller's
169 point of view is a reference to dentry at the root of (sub)tree to
170 be attached; creation of new superblock is a common side effect.
172 The most interesting member of the superblock structure that the
173 mount() method fills in is the "s_op" field. This is a pointer to
174 a "struct super_operations" which describes the next level of the
175 filesystem implementation.
177 Usually, a filesystem uses one of the generic mount() implementations
178 and provides a fill_super() callback instead. The generic variants are:
180 mount_bdev: mount a filesystem residing on a block device
182 mount_nodev: mount a filesystem that is not backed by a device
184 mount_single: mount a filesystem which shares the instance between
187 A fill_super() callback implementation has the following arguments:
189 struct super_block *sb: the superblock structure. The callback
190 must initialize this properly.
192 void *data: arbitrary mount options, usually comes as an ASCII
193 string (see "Mount Options" section)
195 int silent: whether or not to be silent on error
198 The Superblock Object
199 =====================
201 A superblock object represents a mounted filesystem.
204 struct super_operations
205 -----------------------
207 This describes how the VFS can manipulate the superblock of your
208 filesystem. As of kernel 2.6.22, the following members are defined:
210 struct super_operations {
211 struct inode *(*alloc_inode)(struct super_block *sb);
212 void (*destroy_inode)(struct inode *);
214 void (*dirty_inode) (struct inode *, int flags);
215 int (*write_inode) (struct inode *, int);
216 void (*drop_inode) (struct inode *);
217 void (*delete_inode) (struct inode *);
218 void (*put_super) (struct super_block *);
219 void (*write_super) (struct super_block *);
220 int (*sync_fs)(struct super_block *sb, int wait);
221 int (*freeze_fs) (struct super_block *);
222 int (*unfreeze_fs) (struct super_block *);
223 int (*statfs) (struct dentry *, struct kstatfs *);
224 int (*remount_fs) (struct super_block *, int *, char *);
225 void (*clear_inode) (struct inode *);
226 void (*umount_begin) (struct super_block *);
228 int (*show_options)(struct seq_file *, struct dentry *);
230 ssize_t (*quota_read)(struct super_block *, int, char *, size_t, loff_t);
231 ssize_t (*quota_write)(struct super_block *, int, const char *, size_t, loff_t);
232 int (*nr_cached_objects)(struct super_block *);
233 void (*free_cached_objects)(struct super_block *, int);
236 All methods are called without any locks being held, unless otherwise
237 noted. This means that most methods can block safely. All methods are
238 only called from a process context (i.e. not from an interrupt handler
241 alloc_inode: this method is called by inode_alloc() to allocate memory
242 for struct inode and initialize it. If this function is not
243 defined, a simple 'struct inode' is allocated. Normally
244 alloc_inode will be used to allocate a larger structure which
245 contains a 'struct inode' embedded within it.
247 destroy_inode: this method is called by destroy_inode() to release
248 resources allocated for struct inode. It is only required if
249 ->alloc_inode was defined and simply undoes anything done by
252 dirty_inode: this method is called by the VFS to mark an inode dirty.
254 write_inode: this method is called when the VFS needs to write an
255 inode to disc. The second parameter indicates whether the write
256 should be synchronous or not, not all filesystems check this flag.
258 drop_inode: called when the last access to the inode is dropped,
259 with the inode->i_lock spinlock held.
261 This method should be either NULL (normal UNIX filesystem
262 semantics) or "generic_delete_inode" (for filesystems that do not
263 want to cache inodes - causing "delete_inode" to always be
264 called regardless of the value of i_nlink)
266 The "generic_delete_inode()" behavior is equivalent to the
267 old practice of using "force_delete" in the put_inode() case,
268 but does not have the races that the "force_delete()" approach
271 delete_inode: called when the VFS wants to delete an inode
273 put_super: called when the VFS wishes to free the superblock
274 (i.e. unmount). This is called with the superblock lock held
276 write_super: called when the VFS superblock needs to be written to
277 disc. This method is optional
279 sync_fs: called when VFS is writing out all dirty data associated with
280 a superblock. The second parameter indicates whether the method
281 should wait until the write out has been completed. Optional.
283 freeze_fs: called when VFS is locking a filesystem and
284 forcing it into a consistent state. This method is currently
285 used by the Logical Volume Manager (LVM).
287 unfreeze_fs: called when VFS is unlocking a filesystem and making it writable
290 statfs: called when the VFS needs to get filesystem statistics.
292 remount_fs: called when the filesystem is remounted. This is called
293 with the kernel lock held
295 clear_inode: called then the VFS clears the inode. Optional
297 umount_begin: called when the VFS is unmounting a filesystem.
299 show_options: called by the VFS to show mount options for
300 /proc/<pid>/mounts. (see "Mount Options" section)
302 quota_read: called by the VFS to read from filesystem quota file.
304 quota_write: called by the VFS to write to filesystem quota file.
306 nr_cached_objects: called by the sb cache shrinking function for the
307 filesystem to return the number of freeable cached objects it contains.
310 free_cache_objects: called by the sb cache shrinking function for the
311 filesystem to scan the number of objects indicated to try to free them.
312 Optional, but any filesystem implementing this method needs to also
313 implement ->nr_cached_objects for it to be called correctly.
315 We can't do anything with any errors that the filesystem might
316 encountered, hence the void return type. This will never be called if
317 the VM is trying to reclaim under GFP_NOFS conditions, hence this
318 method does not need to handle that situation itself.
320 Implementations must include conditional reschedule calls inside any
321 scanning loop that is done. This allows the VFS to determine
322 appropriate scan batch sizes without having to worry about whether
323 implementations will cause holdoff problems due to large scan batch
326 Whoever sets up the inode is responsible for filling in the "i_op" field. This
327 is a pointer to a "struct inode_operations" which describes the methods that
328 can be performed on individual inodes.
334 An inode object represents an object within the filesystem.
337 struct inode_operations
338 -----------------------
340 This describes how the VFS can manipulate an inode in your
341 filesystem. As of kernel 2.6.22, the following members are defined:
343 struct inode_operations {
344 int (*create) (struct inode *,struct dentry *, umode_t, struct nameidata *);
345 struct dentry * (*lookup) (struct inode *,struct dentry *, struct nameidata *);
346 int (*link) (struct dentry *,struct inode *,struct dentry *);
347 int (*unlink) (struct inode *,struct dentry *);
348 int (*symlink) (struct inode *,struct dentry *,const char *);
349 int (*mkdir) (struct inode *,struct dentry *,umode_t);
350 int (*rmdir) (struct inode *,struct dentry *);
351 int (*mknod) (struct inode *,struct dentry *,umode_t,dev_t);
352 int (*rename) (struct inode *, struct dentry *,
353 struct inode *, struct dentry *);
354 int (*readlink) (struct dentry *, char __user *,int);
355 void * (*follow_link) (struct dentry *, struct nameidata *);
356 void (*put_link) (struct dentry *, struct nameidata *, void *);
357 void (*truncate) (struct inode *);
358 int (*permission) (struct inode *, int);
359 int (*get_acl)(struct inode *, int);
360 int (*setattr) (struct dentry *, struct iattr *);
361 int (*getattr) (struct vfsmount *mnt, struct dentry *, struct kstat *);
362 int (*setxattr) (struct dentry *, const char *,const void *,size_t,int);
363 ssize_t (*getxattr) (struct dentry *, const char *, void *, size_t);
364 ssize_t (*listxattr) (struct dentry *, char *, size_t);
365 int (*removexattr) (struct dentry *, const char *);
366 void (*truncate_range)(struct inode *, loff_t, loff_t);
367 void (*update_time)(struct inode *, struct timespec *, int);
370 Again, all methods are called without any locks being held, unless
373 create: called by the open(2) and creat(2) system calls. Only
374 required if you want to support regular files. The dentry you
375 get should not have an inode (i.e. it should be a negative
376 dentry). Here you will probably call d_instantiate() with the
377 dentry and the newly created inode
379 lookup: called when the VFS needs to look up an inode in a parent
380 directory. The name to look for is found in the dentry. This
381 method must call d_add() to insert the found inode into the
382 dentry. The "i_count" field in the inode structure should be
383 incremented. If the named inode does not exist a NULL inode
384 should be inserted into the dentry (this is called a negative
385 dentry). Returning an error code from this routine must only
386 be done on a real error, otherwise creating inodes with system
387 calls like create(2), mknod(2), mkdir(2) and so on will fail.
388 If you wish to overload the dentry methods then you should
389 initialise the "d_dop" field in the dentry; this is a pointer
390 to a struct "dentry_operations".
391 This method is called with the directory inode semaphore held
393 link: called by the link(2) system call. Only required if you want
394 to support hard links. You will probably need to call
395 d_instantiate() just as you would in the create() method
397 unlink: called by the unlink(2) system call. Only required if you
398 want to support deleting inodes
400 symlink: called by the symlink(2) system call. Only required if you
401 want to support symlinks. You will probably need to call
402 d_instantiate() just as you would in the create() method
404 mkdir: called by the mkdir(2) system call. Only required if you want
405 to support creating subdirectories. You will probably need to
406 call d_instantiate() just as you would in the create() method
408 rmdir: called by the rmdir(2) system call. Only required if you want
409 to support deleting subdirectories
411 mknod: called by the mknod(2) system call to create a device (char,
412 block) inode or a named pipe (FIFO) or socket. Only required
413 if you want to support creating these types of inodes. You
414 will probably need to call d_instantiate() just as you would
415 in the create() method
417 rename: called by the rename(2) system call to rename the object to
418 have the parent and name given by the second inode and dentry.
420 readlink: called by the readlink(2) system call. Only required if
421 you want to support reading symbolic links
423 follow_link: called by the VFS to follow a symbolic link to the
424 inode it points to. Only required if you want to support
425 symbolic links. This method returns a void pointer cookie
426 that is passed to put_link().
428 put_link: called by the VFS to release resources allocated by
429 follow_link(). The cookie returned by follow_link() is passed
430 to this method as the last parameter. It is used by
431 filesystems such as NFS where page cache is not stable
432 (i.e. page that was installed when the symbolic link walk
433 started might not be in the page cache at the end of the
436 truncate: Deprecated. This will not be called if ->setsize is defined.
437 Called by the VFS to change the size of a file. The
438 i_size field of the inode is set to the desired size by the
439 VFS before this method is called. This method is called by
440 the truncate(2) system call and related functionality.
442 Note: ->truncate and vmtruncate are deprecated. Do not add new
443 instances/calls of these. Filesystems should be converted to do their
444 truncate sequence via ->setattr().
446 permission: called by the VFS to check for access rights on a POSIX-like
449 May be called in rcu-walk mode (mask & MAY_NOT_BLOCK). If in rcu-walk
450 mode, the filesystem must check the permission without blocking or
451 storing to the inode.
453 If a situation is encountered that rcu-walk cannot handle, return
454 -ECHILD and it will be called again in ref-walk mode.
456 setattr: called by the VFS to set attributes for a file. This method
457 is called by chmod(2) and related system calls.
459 getattr: called by the VFS to get attributes of a file. This method
460 is called by stat(2) and related system calls.
462 setxattr: called by the VFS to set an extended attribute for a file.
463 Extended attribute is a name:value pair associated with an
464 inode. This method is called by setxattr(2) system call.
466 getxattr: called by the VFS to retrieve the value of an extended
467 attribute name. This method is called by getxattr(2) function
470 listxattr: called by the VFS to list all extended attributes for a
471 given file. This method is called by listxattr(2) system call.
473 removexattr: called by the VFS to remove an extended attribute from
474 a file. This method is called by removexattr(2) system call.
476 truncate_range: a method provided by the underlying filesystem to truncate a
477 range of blocks , i.e. punch a hole somewhere in a file.
479 update_time: called by the VFS to update a specific time or the i_version of
480 an inode. If this is not defined the VFS will update the inode itself
481 and call mark_inode_dirty_sync.
483 The Address Space Object
484 ========================
486 The address space object is used to group and manage pages in the page
487 cache. It can be used to keep track of the pages in a file (or
488 anything else) and also track the mapping of sections of the file into
489 process address spaces.
491 There are a number of distinct yet related services that an
492 address-space can provide. These include communicating memory
493 pressure, page lookup by address, and keeping track of pages tagged as
496 The first can be used independently to the others. The VM can try to
497 either write dirty pages in order to clean them, or release clean
498 pages in order to reuse them. To do this it can call the ->writepage
499 method on dirty pages, and ->releasepage on clean pages with
500 PagePrivate set. Clean pages without PagePrivate and with no external
501 references will be released without notice being given to the
504 To achieve this functionality, pages need to be placed on an LRU with
505 lru_cache_add and mark_page_active needs to be called whenever the
508 Pages are normally kept in a radix tree index by ->index. This tree
509 maintains information about the PG_Dirty and PG_Writeback status of
510 each page, so that pages with either of these flags can be found
513 The Dirty tag is primarily used by mpage_writepages - the default
514 ->writepages method. It uses the tag to find dirty pages to call
515 ->writepage on. If mpage_writepages is not used (i.e. the address
516 provides its own ->writepages) , the PAGECACHE_TAG_DIRTY tag is
517 almost unused. write_inode_now and sync_inode do use it (through
518 __sync_single_inode) to check if ->writepages has been successful in
519 writing out the whole address_space.
521 The Writeback tag is used by filemap*wait* and sync_page* functions,
522 via filemap_fdatawait_range, to wait for all writeback to
523 complete. While waiting ->sync_page (if defined) will be called on
524 each page that is found to require writeback.
526 An address_space handler may attach extra information to a page,
527 typically using the 'private' field in the 'struct page'. If such
528 information is attached, the PG_Private flag should be set. This will
529 cause various VM routines to make extra calls into the address_space
530 handler to deal with that data.
532 An address space acts as an intermediate between storage and
533 application. Data is read into the address space a whole page at a
534 time, and provided to the application either by copying of the page,
535 or by memory-mapping the page.
536 Data is written into the address space by the application, and then
537 written-back to storage typically in whole pages, however the
538 address_space has finer control of write sizes.
540 The read process essentially only requires 'readpage'. The write
541 process is more complicated and uses write_begin/write_end or
542 set_page_dirty to write data into the address_space, and writepage,
543 sync_page, and writepages to writeback data to storage.
545 Adding and removing pages to/from an address_space is protected by the
548 When data is written to a page, the PG_Dirty flag should be set. It
549 typically remains set until writepage asks for it to be written. This
550 should clear PG_Dirty and set PG_Writeback. It can be actually
551 written at any point after PG_Dirty is clear. Once it is known to be
552 safe, PG_Writeback is cleared.
554 Writeback makes use of a writeback_control structure...
556 struct address_space_operations
557 -------------------------------
559 This describes how the VFS can manipulate mapping of a file to page cache in
560 your filesystem. As of kernel 2.6.22, the following members are defined:
562 struct address_space_operations {
563 int (*writepage)(struct page *page, struct writeback_control *wbc);
564 int (*readpage)(struct file *, struct page *);
565 int (*sync_page)(struct page *);
566 int (*writepages)(struct address_space *, struct writeback_control *);
567 int (*set_page_dirty)(struct page *page);
568 int (*readpages)(struct file *filp, struct address_space *mapping,
569 struct list_head *pages, unsigned nr_pages);
570 int (*write_begin)(struct file *, struct address_space *mapping,
571 loff_t pos, unsigned len, unsigned flags,
572 struct page **pagep, void **fsdata);
573 int (*write_end)(struct file *, struct address_space *mapping,
574 loff_t pos, unsigned len, unsigned copied,
575 struct page *page, void *fsdata);
576 sector_t (*bmap)(struct address_space *, sector_t);
577 int (*invalidatepage) (struct page *, unsigned long);
578 int (*releasepage) (struct page *, int);
579 void (*freepage)(struct page *);
580 ssize_t (*direct_IO)(int, struct kiocb *, const struct iovec *iov,
581 loff_t offset, unsigned long nr_segs);
582 struct page* (*get_xip_page)(struct address_space *, sector_t,
584 /* migrate the contents of a page to the specified target */
585 int (*migratepage) (struct page *, struct page *);
586 int (*launder_page) (struct page *);
587 int (*error_remove_page) (struct mapping *mapping, struct page *page);
590 writepage: called by the VM to write a dirty page to backing store.
591 This may happen for data integrity reasons (i.e. 'sync'), or
592 to free up memory (flush). The difference can be seen in
594 The PG_Dirty flag has been cleared and PageLocked is true.
595 writepage should start writeout, should set PG_Writeback,
596 and should make sure the page is unlocked, either synchronously
597 or asynchronously when the write operation completes.
599 If wbc->sync_mode is WB_SYNC_NONE, ->writepage doesn't have to
600 try too hard if there are problems, and may choose to write out
601 other pages from the mapping if that is easier (e.g. due to
602 internal dependencies). If it chooses not to start writeout, it
603 should return AOP_WRITEPAGE_ACTIVATE so that the VM will not keep
604 calling ->writepage on that page.
606 See the file "Locking" for more details.
608 readpage: called by the VM to read a page from backing store.
609 The page will be Locked when readpage is called, and should be
610 unlocked and marked uptodate once the read completes.
611 If ->readpage discovers that it needs to unlock the page for
612 some reason, it can do so, and then return AOP_TRUNCATED_PAGE.
613 In this case, the page will be relocated, relocked and if
614 that all succeeds, ->readpage will be called again.
616 sync_page: called by the VM to notify the backing store to perform all
617 queued I/O operations for a page. I/O operations for other pages
618 associated with this address_space object may also be performed.
620 This function is optional and is called only for pages with
621 PG_Writeback set while waiting for the writeback to complete.
623 writepages: called by the VM to write out pages associated with the
624 address_space object. If wbc->sync_mode is WBC_SYNC_ALL, then
625 the writeback_control will specify a range of pages that must be
626 written out. If it is WBC_SYNC_NONE, then a nr_to_write is given
627 and that many pages should be written if possible.
628 If no ->writepages is given, then mpage_writepages is used
629 instead. This will choose pages from the address space that are
630 tagged as DIRTY and will pass them to ->writepage.
632 set_page_dirty: called by the VM to set a page dirty.
633 This is particularly needed if an address space attaches
634 private data to a page, and that data needs to be updated when
635 a page is dirtied. This is called, for example, when a memory
636 mapped page gets modified.
637 If defined, it should set the PageDirty flag, and the
638 PAGECACHE_TAG_DIRTY tag in the radix tree.
640 readpages: called by the VM to read pages associated with the address_space
641 object. This is essentially just a vector version of
642 readpage. Instead of just one page, several pages are
644 readpages is only used for read-ahead, so read errors are
645 ignored. If anything goes wrong, feel free to give up.
648 Called by the generic buffered write code to ask the filesystem to
649 prepare to write len bytes at the given offset in the file. The
650 address_space should check that the write will be able to complete,
651 by allocating space if necessary and doing any other internal
652 housekeeping. If the write will update parts of any basic-blocks on
653 storage, then those blocks should be pre-read (if they haven't been
654 read already) so that the updated blocks can be written out properly.
656 The filesystem must return the locked pagecache page for the specified
657 offset, in *pagep, for the caller to write into.
659 It must be able to cope with short writes (where the length passed to
660 write_begin is greater than the number of bytes copied into the page).
662 flags is a field for AOP_FLAG_xxx flags, described in
665 A void * may be returned in fsdata, which then gets passed into
668 Returns 0 on success; < 0 on failure (which is the error code), in
669 which case write_end is not called.
671 write_end: After a successful write_begin, and data copy, write_end must
672 be called. len is the original len passed to write_begin, and copied
673 is the amount that was able to be copied (copied == len is always true
674 if write_begin was called with the AOP_FLAG_UNINTERRUPTIBLE flag).
676 The filesystem must take care of unlocking the page and releasing it
677 refcount, and updating i_size.
679 Returns < 0 on failure, otherwise the number of bytes (<= 'copied')
680 that were able to be copied into pagecache.
682 bmap: called by the VFS to map a logical block offset within object to
683 physical block number. This method is used by the FIBMAP
684 ioctl and for working with swap-files. To be able to swap to
685 a file, the file must have a stable mapping to a block
686 device. The swap system does not go through the filesystem
687 but instead uses bmap to find out where the blocks in the file
688 are and uses those addresses directly.
691 invalidatepage: If a page has PagePrivate set, then invalidatepage
692 will be called when part or all of the page is to be removed
693 from the address space. This generally corresponds to either a
694 truncation or a complete invalidation of the address space
695 (in the latter case 'offset' will always be 0).
696 Any private data associated with the page should be updated
697 to reflect this truncation. If offset is 0, then
698 the private data should be released, because the page
699 must be able to be completely discarded. This may be done by
700 calling the ->releasepage function, but in this case the
701 release MUST succeed.
703 releasepage: releasepage is called on PagePrivate pages to indicate
704 that the page should be freed if possible. ->releasepage
705 should remove any private data from the page and clear the
706 PagePrivate flag. If releasepage() fails for some reason, it must
707 indicate failure with a 0 return value.
708 releasepage() is used in two distinct though related cases. The
709 first is when the VM finds a clean page with no active users and
710 wants to make it a free page. If ->releasepage succeeds, the
711 page will be removed from the address_space and become free.
713 The second case is when a request has been made to invalidate
714 some or all pages in an address_space. This can happen
715 through the fadvice(POSIX_FADV_DONTNEED) system call or by the
716 filesystem explicitly requesting it as nfs and 9fs do (when
717 they believe the cache may be out of date with storage) by
718 calling invalidate_inode_pages2().
719 If the filesystem makes such a call, and needs to be certain
720 that all pages are invalidated, then its releasepage will
721 need to ensure this. Possibly it can clear the PageUptodate
722 bit if it cannot free private data yet.
724 freepage: freepage is called once the page is no longer visible in
725 the page cache in order to allow the cleanup of any private
726 data. Since it may be called by the memory reclaimer, it
727 should not assume that the original address_space mapping still
728 exists, and it should not block.
730 direct_IO: called by the generic read/write routines to perform
731 direct_IO - that is IO requests which bypass the page cache
732 and transfer data directly between the storage and the
733 application's address space.
735 get_xip_page: called by the VM to translate a block number to a page.
736 The page is valid until the corresponding filesystem is unmounted.
737 Filesystems that want to use execute-in-place (XIP) need to implement
738 it. An example implementation can be found in fs/ext2/xip.c.
740 migrate_page: This is used to compact the physical memory usage.
741 If the VM wants to relocate a page (maybe off a memory card
742 that is signalling imminent failure) it will pass a new page
743 and an old page to this function. migrate_page should
744 transfer any private data across and update any references
745 that it has to the page.
747 launder_page: Called before freeing a page - it writes back the dirty page. To
748 prevent redirtying the page, it is kept locked during the whole
751 error_remove_page: normally set to generic_error_remove_page if truncation
752 is ok for this address space. Used for memory failure handling.
753 Setting this implies you deal with pages going away under you,
754 unless you have them locked or reference counts increased.
760 A file object represents a file opened by a process.
763 struct file_operations
764 ----------------------
766 This describes how the VFS can manipulate an open file. As of kernel
767 2.6.22, the following members are defined:
769 struct file_operations {
770 struct module *owner;
771 loff_t (*llseek) (struct file *, loff_t, int);
772 ssize_t (*read) (struct file *, char __user *, size_t, loff_t *);
773 ssize_t (*write) (struct file *, const char __user *, size_t, loff_t *);
774 ssize_t (*aio_read) (struct kiocb *, const struct iovec *, unsigned long, loff_t);
775 ssize_t (*aio_write) (struct kiocb *, const struct iovec *, unsigned long, loff_t);
776 int (*readdir) (struct file *, void *, filldir_t);
777 unsigned int (*poll) (struct file *, struct poll_table_struct *);
778 long (*unlocked_ioctl) (struct file *, unsigned int, unsigned long);
779 long (*compat_ioctl) (struct file *, unsigned int, unsigned long);
780 int (*mmap) (struct file *, struct vm_area_struct *);
781 int (*open) (struct inode *, struct file *);
782 int (*flush) (struct file *);
783 int (*release) (struct inode *, struct file *);
784 int (*fsync) (struct file *, loff_t, loff_t, int datasync);
785 int (*aio_fsync) (struct kiocb *, int datasync);
786 int (*fasync) (int, struct file *, int);
787 int (*lock) (struct file *, int, struct file_lock *);
788 ssize_t (*readv) (struct file *, const struct iovec *, unsigned long, loff_t *);
789 ssize_t (*writev) (struct file *, const struct iovec *, unsigned long, loff_t *);
790 ssize_t (*sendfile) (struct file *, loff_t *, size_t, read_actor_t, void *);
791 ssize_t (*sendpage) (struct file *, struct page *, int, size_t, loff_t *, int);
792 unsigned long (*get_unmapped_area)(struct file *, unsigned long, unsigned long, unsigned long, unsigned long);
793 int (*check_flags)(int);
794 int (*flock) (struct file *, int, struct file_lock *);
795 ssize_t (*splice_write)(struct pipe_inode_info *, struct file *, size_t, unsigned int);
796 ssize_t (*splice_read)(struct file *, struct pipe_inode_info *, size_t, unsigned int);
799 Again, all methods are called without any locks being held, unless
802 llseek: called when the VFS needs to move the file position index
804 read: called by read(2) and related system calls
806 aio_read: called by io_submit(2) and other asynchronous I/O operations
808 write: called by write(2) and related system calls
810 aio_write: called by io_submit(2) and other asynchronous I/O operations
812 readdir: called when the VFS needs to read the directory contents
814 poll: called by the VFS when a process wants to check if there is
815 activity on this file and (optionally) go to sleep until there
816 is activity. Called by the select(2) and poll(2) system calls
818 unlocked_ioctl: called by the ioctl(2) system call.
820 compat_ioctl: called by the ioctl(2) system call when 32 bit system calls
821 are used on 64 bit kernels.
823 mmap: called by the mmap(2) system call
825 open: called by the VFS when an inode should be opened. When the VFS
826 opens a file, it creates a new "struct file". It then calls the
827 open method for the newly allocated file structure. You might
828 think that the open method really belongs in
829 "struct inode_operations", and you may be right. I think it's
830 done the way it is because it makes filesystems simpler to
831 implement. The open() method is a good place to initialize the
832 "private_data" member in the file structure if you want to point
833 to a device structure
835 flush: called by the close(2) system call to flush a file
837 release: called when the last reference to an open file is closed
839 fsync: called by the fsync(2) system call
841 fasync: called by the fcntl(2) system call when asynchronous
842 (non-blocking) mode is enabled for a file
844 lock: called by the fcntl(2) system call for F_GETLK, F_SETLK, and F_SETLKW
847 readv: called by the readv(2) system call
849 writev: called by the writev(2) system call
851 sendfile: called by the sendfile(2) system call
853 get_unmapped_area: called by the mmap(2) system call
855 check_flags: called by the fcntl(2) system call for F_SETFL command
857 flock: called by the flock(2) system call
859 splice_write: called by the VFS to splice data from a pipe to a file. This
860 method is used by the splice(2) system call
862 splice_read: called by the VFS to splice data from file to a pipe. This
863 method is used by the splice(2) system call
865 Note that the file operations are implemented by the specific
866 filesystem in which the inode resides. When opening a device node
867 (character or block special) most filesystems will call special
868 support routines in the VFS which will locate the required device
869 driver information. These support routines replace the filesystem file
870 operations with those for the device driver, and then proceed to call
871 the new open() method for the file. This is how opening a device file
872 in the filesystem eventually ends up calling the device driver open()
876 Directory Entry Cache (dcache)
877 ==============================
880 struct dentry_operations
881 ------------------------
883 This describes how a filesystem can overload the standard dentry
884 operations. Dentries and the dcache are the domain of the VFS and the
885 individual filesystem implementations. Device drivers have no business
886 here. These methods may be set to NULL, as they are either optional or
887 the VFS uses a default. As of kernel 2.6.22, the following members are
890 struct dentry_operations {
891 int (*d_revalidate)(struct dentry *, struct nameidata *);
892 int (*d_hash)(const struct dentry *, const struct inode *,
894 int (*d_compare)(const struct dentry *, const struct inode *,
895 const struct dentry *, const struct inode *,
896 unsigned int, const char *, const struct qstr *);
897 int (*d_delete)(const struct dentry *);
898 void (*d_release)(struct dentry *);
899 void (*d_iput)(struct dentry *, struct inode *);
900 char *(*d_dname)(struct dentry *, char *, int);
901 struct vfsmount *(*d_automount)(struct path *);
902 int (*d_manage)(struct dentry *, bool);
905 d_revalidate: called when the VFS needs to revalidate a dentry. This
906 is called whenever a name look-up finds a dentry in the
907 dcache. Most filesystems leave this as NULL, because all their
908 dentries in the dcache are valid
910 d_revalidate may be called in rcu-walk mode (nd->flags & LOOKUP_RCU).
911 If in rcu-walk mode, the filesystem must revalidate the dentry without
912 blocking or storing to the dentry, d_parent and d_inode should not be
913 used without care (because they can go NULL), instead nd->inode should
916 If a situation is encountered that rcu-walk cannot handle, return
917 -ECHILD and it will be called again in ref-walk mode.
919 d_hash: called when the VFS adds a dentry to the hash table. The first
920 dentry passed to d_hash is the parent directory that the name is
921 to be hashed into. The inode is the dentry's inode.
923 Same locking and synchronisation rules as d_compare regarding
924 what is safe to dereference etc.
926 d_compare: called to compare a dentry name with a given name. The first
927 dentry is the parent of the dentry to be compared, the second is
928 the parent's inode, then the dentry and inode (may be NULL) of the
929 child dentry. len and name string are properties of the dentry to be
930 compared. qstr is the name to compare it with.
932 Must be constant and idempotent, and should not take locks if
933 possible, and should not or store into the dentry or inodes.
934 Should not dereference pointers outside the dentry or inodes without
935 lots of care (eg. d_parent, d_inode, d_name should not be used).
937 However, our vfsmount is pinned, and RCU held, so the dentries and
938 inodes won't disappear, neither will our sb or filesystem module.
939 ->i_sb and ->d_sb may be used.
941 It is a tricky calling convention because it needs to be called under
942 "rcu-walk", ie. without any locks or references on things.
944 d_delete: called when the last reference to a dentry is dropped and the
945 dcache is deciding whether or not to cache it. Return 1 to delete
946 immediately, or 0 to cache the dentry. Default is NULL which means to
947 always cache a reachable dentry. d_delete must be constant and
950 d_release: called when a dentry is really deallocated
952 d_iput: called when a dentry loses its inode (just prior to its
953 being deallocated). The default when this is NULL is that the
954 VFS calls iput(). If you define this method, you must call
957 d_dname: called when the pathname of a dentry should be generated.
958 Useful for some pseudo filesystems (sockfs, pipefs, ...) to delay
959 pathname generation. (Instead of doing it when dentry is created,
960 it's done only when the path is needed.). Real filesystems probably
961 dont want to use it, because their dentries are present in global
962 dcache hash, so their hash should be an invariant. As no lock is
963 held, d_dname() should not try to modify the dentry itself, unless
964 appropriate SMP safety is used. CAUTION : d_path() logic is quite
965 tricky. The correct way to return for example "Hello" is to put it
966 at the end of the buffer, and returns a pointer to the first char.
967 dynamic_dname() helper function is provided to take care of this.
969 d_automount: called when an automount dentry is to be traversed (optional).
970 This should create a new VFS mount record and return the record to the
971 caller. The caller is supplied with a path parameter giving the
972 automount directory to describe the automount target and the parent
973 VFS mount record to provide inheritable mount parameters. NULL should
974 be returned if someone else managed to make the automount first. If
975 the vfsmount creation failed, then an error code should be returned.
976 If -EISDIR is returned, then the directory will be treated as an
977 ordinary directory and returned to pathwalk to continue walking.
979 If a vfsmount is returned, the caller will attempt to mount it on the
980 mountpoint and will remove the vfsmount from its expiration list in
981 the case of failure. The vfsmount should be returned with 2 refs on
982 it to prevent automatic expiration - the caller will clean up the
985 This function is only used if DCACHE_NEED_AUTOMOUNT is set on the
986 dentry. This is set by __d_instantiate() if S_AUTOMOUNT is set on the
989 d_manage: called to allow the filesystem to manage the transition from a
990 dentry (optional). This allows autofs, for example, to hold up clients
991 waiting to explore behind a 'mountpoint' whilst letting the daemon go
992 past and construct the subtree there. 0 should be returned to let the
993 calling process continue. -EISDIR can be returned to tell pathwalk to
994 use this directory as an ordinary directory and to ignore anything
995 mounted on it and not to check the automount flag. Any other error
996 code will abort pathwalk completely.
998 If the 'rcu_walk' parameter is true, then the caller is doing a
999 pathwalk in RCU-walk mode. Sleeping is not permitted in this mode,
1000 and the caller can be asked to leave it and call again by returning
1003 This function is only used if DCACHE_MANAGE_TRANSIT is set on the
1004 dentry being transited from.
1008 static char *pipefs_dname(struct dentry *dent, char *buffer, int buflen)
1010 return dynamic_dname(dentry, buffer, buflen, "pipe:[%lu]",
1011 dentry->d_inode->i_ino);
1014 Each dentry has a pointer to its parent dentry, as well as a hash list
1015 of child dentries. Child dentries are basically like files in a
1019 Directory Entry Cache API
1020 --------------------------
1022 There are a number of functions defined which permit a filesystem to
1023 manipulate dentries:
1025 dget: open a new handle for an existing dentry (this just increments
1028 dput: close a handle for a dentry (decrements the usage count). If
1029 the usage count drops to 0, and the dentry is still in its
1030 parent's hash, the "d_delete" method is called to check whether
1031 it should be cached. If it should not be cached, or if the dentry
1032 is not hashed, it is deleted. Otherwise cached dentries are put
1033 into an LRU list to be reclaimed on memory shortage.
1035 d_drop: this unhashes a dentry from its parents hash list. A
1036 subsequent call to dput() will deallocate the dentry if its
1037 usage count drops to 0
1039 d_delete: delete a dentry. If there are no other open references to
1040 the dentry then the dentry is turned into a negative dentry
1041 (the d_iput() method is called). If there are other
1042 references, then d_drop() is called instead
1044 d_add: add a dentry to its parents hash list and then calls
1047 d_instantiate: add a dentry to the alias hash list for the inode and
1048 updates the "d_inode" member. The "i_count" member in the
1049 inode structure should be set/incremented. If the inode
1050 pointer is NULL, the dentry is called a "negative
1051 dentry". This function is commonly called when an inode is
1052 created for an existing negative dentry
1054 d_lookup: look up a dentry given its parent and path name component
1055 It looks up the child of that given name from the dcache
1056 hash table. If it is found, the reference count is incremented
1057 and the dentry is returned. The caller must use dput()
1058 to free the dentry when it finishes using it.
1066 On mount and remount the filesystem is passed a string containing a
1067 comma separated list of mount options. The options can have either of
1073 The <linux/parser.h> header defines an API that helps parse these
1074 options. There are plenty of examples on how to use it in existing
1080 If a filesystem accepts mount options, it must define show_options()
1081 to show all the currently active options. The rules are:
1083 - options MUST be shown which are not default or their values differ
1086 - options MAY be shown which are enabled by default or have their
1089 Options used only internally between a mount helper and the kernel
1090 (such as file descriptors), or which only have an effect during the
1091 mounting (such as ones controlling the creation of a journal) are exempt
1092 from the above rules.
1094 The underlying reason for the above rules is to make sure, that a
1095 mount can be accurately replicated (e.g. umounting and mounting again)
1096 based on the information found in /proc/mounts.
1098 A simple method of saving options at mount/remount time and showing
1099 them is provided with the save_mount_options() and
1100 generic_show_options() helper functions. Please note, that using
1101 these may have drawbacks. For more info see header comments for these
1102 functions in fs/namespace.c.
1107 (Note some of these resources are not up-to-date with the latest kernel
1110 Creating Linux virtual filesystems. 2002
1111 <http://lwn.net/Articles/13325/>
1113 The Linux Virtual File-system Layer by Neil Brown. 1999
1114 <http://www.cse.unsw.edu.au/~neilb/oss/linux-commentary/vfs.html>
1116 A tour of the Linux VFS by Michael K. Johnson. 1996
1117 <http://www.tldp.org/LDP/khg/HyperNews/get/fs/vfstour.html>
1119 A small trail through the Linux kernel by Andries Brouwer. 2001
1120 <http://www.win.tue.nl/~aeb/linux/vfs/trail.html>