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 int (*sync_fs)(struct super_block *sb, int wait);
220 int (*freeze_fs) (struct super_block *);
221 int (*unfreeze_fs) (struct super_block *);
222 int (*statfs) (struct dentry *, struct kstatfs *);
223 int (*remount_fs) (struct super_block *, int *, char *);
224 void (*clear_inode) (struct inode *);
225 void (*umount_begin) (struct super_block *);
227 int (*show_options)(struct seq_file *, struct dentry *);
229 ssize_t (*quota_read)(struct super_block *, int, char *, size_t, loff_t);
230 ssize_t (*quota_write)(struct super_block *, int, const char *, size_t, loff_t);
231 int (*nr_cached_objects)(struct super_block *);
232 void (*free_cached_objects)(struct super_block *, int);
235 All methods are called without any locks being held, unless otherwise
236 noted. This means that most methods can block safely. All methods are
237 only called from a process context (i.e. not from an interrupt handler
240 alloc_inode: this method is called by alloc_inode() to allocate memory
241 for struct inode and initialize it. If this function is not
242 defined, a simple 'struct inode' is allocated. Normally
243 alloc_inode will be used to allocate a larger structure which
244 contains a 'struct inode' embedded within it.
246 destroy_inode: this method is called by destroy_inode() to release
247 resources allocated for struct inode. It is only required if
248 ->alloc_inode was defined and simply undoes anything done by
251 dirty_inode: this method is called by the VFS to mark an inode dirty.
253 write_inode: this method is called when the VFS needs to write an
254 inode to disc. The second parameter indicates whether the write
255 should be synchronous or not, not all filesystems check this flag.
257 drop_inode: called when the last access to the inode is dropped,
258 with the inode->i_lock spinlock held.
260 This method should be either NULL (normal UNIX filesystem
261 semantics) or "generic_delete_inode" (for filesystems that do not
262 want to cache inodes - causing "delete_inode" to always be
263 called regardless of the value of i_nlink)
265 The "generic_delete_inode()" behavior is equivalent to the
266 old practice of using "force_delete" in the put_inode() case,
267 but does not have the races that the "force_delete()" approach
270 delete_inode: called when the VFS wants to delete an inode
272 put_super: called when the VFS wishes to free the superblock
273 (i.e. unmount). This is called with the superblock lock held
275 sync_fs: called when VFS is writing out all dirty data associated with
276 a superblock. The second parameter indicates whether the method
277 should wait until the write out has been completed. Optional.
279 freeze_fs: called when VFS is locking a filesystem and
280 forcing it into a consistent state. This method is currently
281 used by the Logical Volume Manager (LVM).
283 unfreeze_fs: called when VFS is unlocking a filesystem and making it writable
286 statfs: called when the VFS needs to get filesystem statistics.
288 remount_fs: called when the filesystem is remounted. This is called
289 with the kernel lock held
291 clear_inode: called then the VFS clears the inode. Optional
293 umount_begin: called when the VFS is unmounting a filesystem.
295 show_options: called by the VFS to show mount options for
296 /proc/<pid>/mounts. (see "Mount Options" section)
298 quota_read: called by the VFS to read from filesystem quota file.
300 quota_write: called by the VFS to write to filesystem quota file.
302 nr_cached_objects: called by the sb cache shrinking function for the
303 filesystem to return the number of freeable cached objects it contains.
306 free_cache_objects: called by the sb cache shrinking function for the
307 filesystem to scan the number of objects indicated to try to free them.
308 Optional, but any filesystem implementing this method needs to also
309 implement ->nr_cached_objects for it to be called correctly.
311 We can't do anything with any errors that the filesystem might
312 encountered, hence the void return type. This will never be called if
313 the VM is trying to reclaim under GFP_NOFS conditions, hence this
314 method does not need to handle that situation itself.
316 Implementations must include conditional reschedule calls inside any
317 scanning loop that is done. This allows the VFS to determine
318 appropriate scan batch sizes without having to worry about whether
319 implementations will cause holdoff problems due to large scan batch
322 Whoever sets up the inode is responsible for filling in the "i_op" field. This
323 is a pointer to a "struct inode_operations" which describes the methods that
324 can be performed on individual inodes.
330 An inode object represents an object within the filesystem.
333 struct inode_operations
334 -----------------------
336 This describes how the VFS can manipulate an inode in your
337 filesystem. As of kernel 2.6.22, the following members are defined:
339 struct inode_operations {
340 int (*create) (struct inode *,struct dentry *, umode_t, bool);
341 struct dentry * (*lookup) (struct inode *,struct dentry *, unsigned int);
342 int (*link) (struct dentry *,struct inode *,struct dentry *);
343 int (*unlink) (struct inode *,struct dentry *);
344 int (*symlink) (struct inode *,struct dentry *,const char *);
345 int (*mkdir) (struct inode *,struct dentry *,umode_t);
346 int (*rmdir) (struct inode *,struct dentry *);
347 int (*mknod) (struct inode *,struct dentry *,umode_t,dev_t);
348 int (*rename) (struct inode *, struct dentry *,
349 struct inode *, struct dentry *);
350 int (*rename2) (struct inode *, struct dentry *,
351 struct inode *, struct dentry *, unsigned int);
352 int (*readlink) (struct dentry *, char __user *,int);
353 const char *(*follow_link) (struct dentry *, void **);
354 void (*put_link) (struct inode *, void *);
355 int (*permission) (struct inode *, int);
356 int (*get_acl)(struct inode *, int);
357 int (*setattr) (struct dentry *, struct iattr *);
358 int (*getattr) (struct vfsmount *mnt, struct dentry *, struct kstat *);
359 int (*setxattr) (struct dentry *, const char *,const void *,size_t,int);
360 ssize_t (*getxattr) (struct dentry *, const char *, void *, size_t);
361 ssize_t (*listxattr) (struct dentry *, char *, size_t);
362 int (*removexattr) (struct dentry *, const char *);
363 void (*update_time)(struct inode *, struct timespec *, int);
364 int (*atomic_open)(struct inode *, struct dentry *, struct file *,
365 unsigned open_flag, umode_t create_mode, int *opened);
366 int (*tmpfile) (struct inode *, struct dentry *, umode_t);
367 int (*dentry_open)(struct dentry *, struct file *, const struct cred *);
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 rename2: this has an additional flags argument compared to rename.
421 If no flags are supported by the filesystem then this method
422 need not be implemented. If some flags are supported then the
423 filesystem must return -EINVAL for any unsupported or unknown
424 flags. Currently the following flags are implemented:
425 (1) RENAME_NOREPLACE: this flag indicates that if the target
426 of the rename exists the rename should fail with -EEXIST
427 instead of replacing the target. The VFS already checks for
428 existence, so for local filesystems the RENAME_NOREPLACE
429 implementation is equivalent to plain rename.
430 (2) RENAME_EXCHANGE: exchange source and target. Both must
431 exist; this is checked by the VFS. Unlike plain rename,
432 source and target may be of different type.
434 readlink: called by the readlink(2) system call. Only required if
435 you want to support reading symbolic links
437 follow_link: called by the VFS to follow a symbolic link to the
438 inode it points to. Only required if you want to support
439 symbolic links. This method returns the symlink body
440 to traverse (and possibly resets the current position with
441 nd_jump_link()). If the body won't go away until the inode
442 is gone, nothing else is needed; if it needs to be otherwise
443 pinned, the data needed to release whatever we'd grabbed
444 is to be stored in void * variable passed by address to
445 follow_link() instance.
447 put_link: called by the VFS to release resources allocated by
448 follow_link(). The cookie stored by follow_link() is passed
449 to this method as the last parameter; only called when
452 permission: called by the VFS to check for access rights on a POSIX-like
455 May be called in rcu-walk mode (mask & MAY_NOT_BLOCK). If in rcu-walk
456 mode, the filesystem must check the permission without blocking or
457 storing to the inode.
459 If a situation is encountered that rcu-walk cannot handle, return
460 -ECHILD and it will be called again in ref-walk mode.
462 setattr: called by the VFS to set attributes for a file. This method
463 is called by chmod(2) and related system calls.
465 getattr: called by the VFS to get attributes of a file. This method
466 is called by stat(2) and related system calls.
468 setxattr: called by the VFS to set an extended attribute for a file.
469 Extended attribute is a name:value pair associated with an
470 inode. This method is called by setxattr(2) system call.
472 getxattr: called by the VFS to retrieve the value of an extended
473 attribute name. This method is called by getxattr(2) function
476 listxattr: called by the VFS to list all extended attributes for a
477 given file. This method is called by listxattr(2) system call.
479 removexattr: called by the VFS to remove an extended attribute from
480 a file. This method is called by removexattr(2) system call.
482 update_time: called by the VFS to update a specific time or the i_version of
483 an inode. If this is not defined the VFS will update the inode itself
484 and call mark_inode_dirty_sync.
486 atomic_open: called on the last component of an open. Using this optional
487 method the filesystem can look up, possibly create and open the file in
488 one atomic operation. If it cannot perform this (e.g. the file type
489 turned out to be wrong) it may signal this by returning 1 instead of
490 usual 0 or -ve . This method is only called if the last component is
491 negative or needs lookup. Cached positive dentries are still handled by
492 f_op->open(). If the file was created, the FILE_CREATED flag should be
493 set in "opened". In case of O_EXCL the method must only succeed if the
494 file didn't exist and hence FILE_CREATED shall always be set on success.
496 tmpfile: called in the end of O_TMPFILE open(). Optional, equivalent to
497 atomically creating, opening and unlinking a file in given directory.
499 The Address Space Object
500 ========================
502 The address space object is used to group and manage pages in the page
503 cache. It can be used to keep track of the pages in a file (or
504 anything else) and also track the mapping of sections of the file into
505 process address spaces.
507 There are a number of distinct yet related services that an
508 address-space can provide. These include communicating memory
509 pressure, page lookup by address, and keeping track of pages tagged as
512 The first can be used independently to the others. The VM can try to
513 either write dirty pages in order to clean them, or release clean
514 pages in order to reuse them. To do this it can call the ->writepage
515 method on dirty pages, and ->releasepage on clean pages with
516 PagePrivate set. Clean pages without PagePrivate and with no external
517 references will be released without notice being given to the
520 To achieve this functionality, pages need to be placed on an LRU with
521 lru_cache_add and mark_page_active needs to be called whenever the
524 Pages are normally kept in a radix tree index by ->index. This tree
525 maintains information about the PG_Dirty and PG_Writeback status of
526 each page, so that pages with either of these flags can be found
529 The Dirty tag is primarily used by mpage_writepages - the default
530 ->writepages method. It uses the tag to find dirty pages to call
531 ->writepage on. If mpage_writepages is not used (i.e. the address
532 provides its own ->writepages) , the PAGECACHE_TAG_DIRTY tag is
533 almost unused. write_inode_now and sync_inode do use it (through
534 __sync_single_inode) to check if ->writepages has been successful in
535 writing out the whole address_space.
537 The Writeback tag is used by filemap*wait* and sync_page* functions,
538 via filemap_fdatawait_range, to wait for all writeback to
539 complete. While waiting ->sync_page (if defined) will be called on
540 each page that is found to require writeback.
542 An address_space handler may attach extra information to a page,
543 typically using the 'private' field in the 'struct page'. If such
544 information is attached, the PG_Private flag should be set. This will
545 cause various VM routines to make extra calls into the address_space
546 handler to deal with that data.
548 An address space acts as an intermediate between storage and
549 application. Data is read into the address space a whole page at a
550 time, and provided to the application either by copying of the page,
551 or by memory-mapping the page.
552 Data is written into the address space by the application, and then
553 written-back to storage typically in whole pages, however the
554 address_space has finer control of write sizes.
556 The read process essentially only requires 'readpage'. The write
557 process is more complicated and uses write_begin/write_end or
558 set_page_dirty to write data into the address_space, and writepage,
559 sync_page, and writepages to writeback data to storage.
561 Adding and removing pages to/from an address_space is protected by the
564 When data is written to a page, the PG_Dirty flag should be set. It
565 typically remains set until writepage asks for it to be written. This
566 should clear PG_Dirty and set PG_Writeback. It can be actually
567 written at any point after PG_Dirty is clear. Once it is known to be
568 safe, PG_Writeback is cleared.
570 Writeback makes use of a writeback_control structure...
572 struct address_space_operations
573 -------------------------------
575 This describes how the VFS can manipulate mapping of a file to page cache in
576 your filesystem. The following members are defined:
578 struct address_space_operations {
579 int (*writepage)(struct page *page, struct writeback_control *wbc);
580 int (*readpage)(struct file *, struct page *);
581 int (*writepages)(struct address_space *, struct writeback_control *);
582 int (*set_page_dirty)(struct page *page);
583 int (*readpages)(struct file *filp, struct address_space *mapping,
584 struct list_head *pages, unsigned nr_pages);
585 int (*write_begin)(struct file *, struct address_space *mapping,
586 loff_t pos, unsigned len, unsigned flags,
587 struct page **pagep, void **fsdata);
588 int (*write_end)(struct file *, struct address_space *mapping,
589 loff_t pos, unsigned len, unsigned copied,
590 struct page *page, void *fsdata);
591 sector_t (*bmap)(struct address_space *, sector_t);
592 void (*invalidatepage) (struct page *, unsigned int, unsigned int);
593 int (*releasepage) (struct page *, int);
594 void (*freepage)(struct page *);
595 ssize_t (*direct_IO)(struct kiocb *, struct iov_iter *iter, loff_t offset);
596 /* isolate a page for migration */
597 bool (*isolate_page) (struct page *, isolate_mode_t);
598 /* migrate the contents of a page to the specified target */
599 int (*migratepage) (struct page *, struct page *);
600 /* put migration-failed page back to right list */
601 void (*putback_page) (struct page *);
602 int (*launder_page) (struct page *);
604 int (*is_partially_uptodate) (struct page *, unsigned long,
606 void (*is_dirty_writeback) (struct page *, bool *, bool *);
607 int (*error_remove_page) (struct mapping *mapping, struct page *page);
608 int (*swap_activate)(struct file *);
609 int (*swap_deactivate)(struct file *);
612 writepage: called by the VM to write a dirty page to backing store.
613 This may happen for data integrity reasons (i.e. 'sync'), or
614 to free up memory (flush). The difference can be seen in
616 The PG_Dirty flag has been cleared and PageLocked is true.
617 writepage should start writeout, should set PG_Writeback,
618 and should make sure the page is unlocked, either synchronously
619 or asynchronously when the write operation completes.
621 If wbc->sync_mode is WB_SYNC_NONE, ->writepage doesn't have to
622 try too hard if there are problems, and may choose to write out
623 other pages from the mapping if that is easier (e.g. due to
624 internal dependencies). If it chooses not to start writeout, it
625 should return AOP_WRITEPAGE_ACTIVATE so that the VM will not keep
626 calling ->writepage on that page.
628 See the file "Locking" for more details.
630 readpage: called by the VM to read a page from backing store.
631 The page will be Locked when readpage is called, and should be
632 unlocked and marked uptodate once the read completes.
633 If ->readpage discovers that it needs to unlock the page for
634 some reason, it can do so, and then return AOP_TRUNCATED_PAGE.
635 In this case, the page will be relocated, relocked and if
636 that all succeeds, ->readpage will be called again.
638 writepages: called by the VM to write out pages associated with the
639 address_space object. If wbc->sync_mode is WBC_SYNC_ALL, then
640 the writeback_control will specify a range of pages that must be
641 written out. If it is WBC_SYNC_NONE, then a nr_to_write is given
642 and that many pages should be written if possible.
643 If no ->writepages is given, then mpage_writepages is used
644 instead. This will choose pages from the address space that are
645 tagged as DIRTY and will pass them to ->writepage.
647 set_page_dirty: called by the VM to set a page dirty.
648 This is particularly needed if an address space attaches
649 private data to a page, and that data needs to be updated when
650 a page is dirtied. This is called, for example, when a memory
651 mapped page gets modified.
652 If defined, it should set the PageDirty flag, and the
653 PAGECACHE_TAG_DIRTY tag in the radix tree.
655 readpages: called by the VM to read pages associated with the address_space
656 object. This is essentially just a vector version of
657 readpage. Instead of just one page, several pages are
659 readpages is only used for read-ahead, so read errors are
660 ignored. If anything goes wrong, feel free to give up.
663 Called by the generic buffered write code to ask the filesystem to
664 prepare to write len bytes at the given offset in the file. The
665 address_space should check that the write will be able to complete,
666 by allocating space if necessary and doing any other internal
667 housekeeping. If the write will update parts of any basic-blocks on
668 storage, then those blocks should be pre-read (if they haven't been
669 read already) so that the updated blocks can be written out properly.
671 The filesystem must return the locked pagecache page for the specified
672 offset, in *pagep, for the caller to write into.
674 It must be able to cope with short writes (where the length passed to
675 write_begin is greater than the number of bytes copied into the page).
677 flags is a field for AOP_FLAG_xxx flags, described in
680 A void * may be returned in fsdata, which then gets passed into
683 Returns 0 on success; < 0 on failure (which is the error code), in
684 which case write_end is not called.
686 write_end: After a successful write_begin, and data copy, write_end must
687 be called. len is the original len passed to write_begin, and copied
688 is the amount that was able to be copied (copied == len is always true
689 if write_begin was called with the AOP_FLAG_UNINTERRUPTIBLE flag).
691 The filesystem must take care of unlocking the page and releasing it
692 refcount, and updating i_size.
694 Returns < 0 on failure, otherwise the number of bytes (<= 'copied')
695 that were able to be copied into pagecache.
697 bmap: called by the VFS to map a logical block offset within object to
698 physical block number. This method is used by the FIBMAP
699 ioctl and for working with swap-files. To be able to swap to
700 a file, the file must have a stable mapping to a block
701 device. The swap system does not go through the filesystem
702 but instead uses bmap to find out where the blocks in the file
703 are and uses those addresses directly.
705 dentry_open: *WARNING: probably going away soon, do not use!* This is an
706 alternative to f_op->open(), the difference is that this method may open
707 a file not necessarily originating from the same filesystem as the one
708 i_op->open() was called on. It may be useful for stacking filesystems
709 which want to allow native I/O directly on underlying files.
712 invalidatepage: If a page has PagePrivate set, then invalidatepage
713 will be called when part or all of the page is to be removed
714 from the address space. This generally corresponds to either a
715 truncation, punch hole or a complete invalidation of the address
716 space (in the latter case 'offset' will always be 0 and 'length'
717 will be PAGE_CACHE_SIZE). Any private data associated with the page
718 should be updated to reflect this truncation. If offset is 0 and
719 length is PAGE_CACHE_SIZE, then the private data should be released,
720 because the page must be able to be completely discarded. This may
721 be done by calling the ->releasepage function, but in this case the
722 release MUST succeed.
724 releasepage: releasepage is called on PagePrivate pages to indicate
725 that the page should be freed if possible. ->releasepage
726 should remove any private data from the page and clear the
727 PagePrivate flag. If releasepage() fails for some reason, it must
728 indicate failure with a 0 return value.
729 releasepage() is used in two distinct though related cases. The
730 first is when the VM finds a clean page with no active users and
731 wants to make it a free page. If ->releasepage succeeds, the
732 page will be removed from the address_space and become free.
734 The second case is when a request has been made to invalidate
735 some or all pages in an address_space. This can happen
736 through the fadvice(POSIX_FADV_DONTNEED) system call or by the
737 filesystem explicitly requesting it as nfs and 9fs do (when
738 they believe the cache may be out of date with storage) by
739 calling invalidate_inode_pages2().
740 If the filesystem makes such a call, and needs to be certain
741 that all pages are invalidated, then its releasepage will
742 need to ensure this. Possibly it can clear the PageUptodate
743 bit if it cannot free private data yet.
745 freepage: freepage is called once the page is no longer visible in
746 the page cache in order to allow the cleanup of any private
747 data. Since it may be called by the memory reclaimer, it
748 should not assume that the original address_space mapping still
749 exists, and it should not block.
751 direct_IO: called by the generic read/write routines to perform
752 direct_IO - that is IO requests which bypass the page cache
753 and transfer data directly between the storage and the
754 application's address space.
756 isolate_page: Called by the VM when isolating a movable non-lru page.
757 If page is successfully isolated, VM marks the page as PG_isolated
758 via __SetPageIsolated.
760 migrate_page: This is used to compact the physical memory usage.
761 If the VM wants to relocate a page (maybe off a memory card
762 that is signalling imminent failure) it will pass a new page
763 and an old page to this function. migrate_page should
764 transfer any private data across and update any references
765 that it has to the page.
767 putback_page: Called by the VM when isolated page's migration fails.
769 launder_page: Called before freeing a page - it writes back the dirty page. To
770 prevent redirtying the page, it is kept locked during the whole
773 is_partially_uptodate: Called by the VM when reading a file through the
774 pagecache when the underlying blocksize != pagesize. If the required
775 block is up to date then the read can complete without needing the IO
776 to bring the whole page up to date.
778 is_dirty_writeback: Called by the VM when attempting to reclaim a page.
779 The VM uses dirty and writeback information to determine if it needs
780 to stall to allow flushers a chance to complete some IO. Ordinarily
781 it can use PageDirty and PageWriteback but some filesystems have
782 more complex state (unstable pages in NFS prevent reclaim) or
783 do not set those flags due to locking problems. This callback
784 allows a filesystem to indicate to the VM if a page should be
785 treated as dirty or writeback for the purposes of stalling.
787 error_remove_page: normally set to generic_error_remove_page if truncation
788 is ok for this address space. Used for memory failure handling.
789 Setting this implies you deal with pages going away under you,
790 unless you have them locked or reference counts increased.
792 swap_activate: Called when swapon is used on a file to allocate
793 space if necessary and pin the block lookup information in
794 memory. A return value of zero indicates success,
795 in which case this file can be used to back swapspace. The
796 swapspace operations will be proxied to this address space's
797 ->swap_{out,in} methods.
799 swap_deactivate: Called during swapoff on files where swap_activate
806 A file object represents a file opened by a process.
809 struct file_operations
810 ----------------------
812 This describes how the VFS can manipulate an open file. As of kernel
813 4.1, the following members are defined:
815 struct file_operations {
816 struct module *owner;
817 loff_t (*llseek) (struct file *, loff_t, int);
818 ssize_t (*read) (struct file *, char __user *, size_t, loff_t *);
819 ssize_t (*write) (struct file *, const char __user *, size_t, loff_t *);
820 ssize_t (*read_iter) (struct kiocb *, struct iov_iter *);
821 ssize_t (*write_iter) (struct kiocb *, struct iov_iter *);
822 int (*iterate) (struct file *, struct dir_context *);
823 unsigned int (*poll) (struct file *, struct poll_table_struct *);
824 long (*unlocked_ioctl) (struct file *, unsigned int, unsigned long);
825 long (*compat_ioctl) (struct file *, unsigned int, unsigned long);
826 int (*mmap) (struct file *, struct vm_area_struct *);
827 int (*mremap)(struct file *, struct vm_area_struct *);
828 int (*open) (struct inode *, struct file *);
829 int (*flush) (struct file *, fl_owner_t id);
830 int (*release) (struct inode *, struct file *);
831 int (*fsync) (struct file *, loff_t, loff_t, int datasync);
832 int (*aio_fsync) (struct kiocb *, int datasync);
833 int (*fasync) (int, struct file *, int);
834 int (*lock) (struct file *, int, struct file_lock *);
835 ssize_t (*sendpage) (struct file *, struct page *, int, size_t, loff_t *, int);
836 unsigned long (*get_unmapped_area)(struct file *, unsigned long, unsigned long, unsigned long, unsigned long);
837 int (*check_flags)(int);
838 int (*flock) (struct file *, int, struct file_lock *);
839 ssize_t (*splice_write)(struct pipe_inode_info *, struct file *, loff_t *, size_t, unsigned int);
840 ssize_t (*splice_read)(struct file *, loff_t *, struct pipe_inode_info *, size_t, unsigned int);
841 int (*setlease)(struct file *, long, struct file_lock **, void **);
842 long (*fallocate)(struct file *file, int mode, loff_t offset,
844 void (*show_fdinfo)(struct seq_file *m, struct file *f);
846 unsigned (*mmap_capabilities)(struct file *);
850 Again, all methods are called without any locks being held, unless
853 llseek: called when the VFS needs to move the file position index
855 read: called by read(2) and related system calls
857 read_iter: possibly asynchronous read with iov_iter as destination
859 write: called by write(2) and related system calls
861 write_iter: possibly asynchronous write with iov_iter as source
863 iterate: called when the VFS needs to read the directory contents
865 poll: called by the VFS when a process wants to check if there is
866 activity on this file and (optionally) go to sleep until there
867 is activity. Called by the select(2) and poll(2) system calls
869 unlocked_ioctl: called by the ioctl(2) system call.
871 compat_ioctl: called by the ioctl(2) system call when 32 bit system calls
872 are used on 64 bit kernels.
874 mmap: called by the mmap(2) system call
876 open: called by the VFS when an inode should be opened. When the VFS
877 opens a file, it creates a new "struct file". It then calls the
878 open method for the newly allocated file structure. You might
879 think that the open method really belongs in
880 "struct inode_operations", and you may be right. I think it's
881 done the way it is because it makes filesystems simpler to
882 implement. The open() method is a good place to initialize the
883 "private_data" member in the file structure if you want to point
884 to a device structure
886 flush: called by the close(2) system call to flush a file
888 release: called when the last reference to an open file is closed
890 fsync: called by the fsync(2) system call
892 fasync: called by the fcntl(2) system call when asynchronous
893 (non-blocking) mode is enabled for a file
895 lock: called by the fcntl(2) system call for F_GETLK, F_SETLK, and F_SETLKW
898 get_unmapped_area: called by the mmap(2) system call
900 check_flags: called by the fcntl(2) system call for F_SETFL command
902 flock: called by the flock(2) system call
904 splice_write: called by the VFS to splice data from a pipe to a file. This
905 method is used by the splice(2) system call
907 splice_read: called by the VFS to splice data from file to a pipe. This
908 method is used by the splice(2) system call
910 setlease: called by the VFS to set or release a file lock lease. setlease
911 implementations should call generic_setlease to record or remove
912 the lease in the inode after setting it.
914 fallocate: called by the VFS to preallocate blocks or punch a hole.
916 Note that the file operations are implemented by the specific
917 filesystem in which the inode resides. When opening a device node
918 (character or block special) most filesystems will call special
919 support routines in the VFS which will locate the required device
920 driver information. These support routines replace the filesystem file
921 operations with those for the device driver, and then proceed to call
922 the new open() method for the file. This is how opening a device file
923 in the filesystem eventually ends up calling the device driver open()
927 Directory Entry Cache (dcache)
928 ==============================
931 struct dentry_operations
932 ------------------------
934 This describes how a filesystem can overload the standard dentry
935 operations. Dentries and the dcache are the domain of the VFS and the
936 individual filesystem implementations. Device drivers have no business
937 here. These methods may be set to NULL, as they are either optional or
938 the VFS uses a default. As of kernel 2.6.22, the following members are
941 struct dentry_operations {
942 int (*d_revalidate)(struct dentry *, unsigned int);
943 int (*d_weak_revalidate)(struct dentry *, unsigned int);
944 int (*d_hash)(const struct dentry *, struct qstr *);
945 int (*d_compare)(const struct dentry *, const struct dentry *,
946 unsigned int, const char *, const struct qstr *);
947 int (*d_delete)(const struct dentry *);
948 void (*d_release)(struct dentry *);
949 void (*d_iput)(struct dentry *, struct inode *);
950 char *(*d_dname)(struct dentry *, char *, int);
951 struct vfsmount *(*d_automount)(struct path *);
952 int (*d_manage)(struct dentry *, bool);
955 d_revalidate: called when the VFS needs to revalidate a dentry. This
956 is called whenever a name look-up finds a dentry in the
957 dcache. Most local filesystems leave this as NULL, because all their
958 dentries in the dcache are valid. Network filesystems are different
959 since things can change on the server without the client necessarily
962 This function should return a positive value if the dentry is still
963 valid, and zero or a negative error code if it isn't.
965 d_revalidate may be called in rcu-walk mode (flags & LOOKUP_RCU).
966 If in rcu-walk mode, the filesystem must revalidate the dentry without
967 blocking or storing to the dentry, d_parent and d_inode should not be
968 used without care (because they can change and, in d_inode case, even
969 become NULL under us).
971 If a situation is encountered that rcu-walk cannot handle, return
972 -ECHILD and it will be called again in ref-walk mode.
974 d_weak_revalidate: called when the VFS needs to revalidate a "jumped" dentry.
975 This is called when a path-walk ends at dentry that was not acquired by
976 doing a lookup in the parent directory. This includes "/", "." and "..",
977 as well as procfs-style symlinks and mountpoint traversal.
979 In this case, we are less concerned with whether the dentry is still
980 fully correct, but rather that the inode is still valid. As with
981 d_revalidate, most local filesystems will set this to NULL since their
982 dcache entries are always valid.
984 This function has the same return code semantics as d_revalidate.
986 d_weak_revalidate is only called after leaving rcu-walk mode.
988 d_hash: called when the VFS adds a dentry to the hash table. The first
989 dentry passed to d_hash is the parent directory that the name is
992 Same locking and synchronisation rules as d_compare regarding
993 what is safe to dereference etc.
995 d_compare: called to compare a dentry name with a given name. The first
996 dentry is the parent of the dentry to be compared, the second is
997 the child dentry. len and name string are properties of the dentry
998 to be compared. qstr is the name to compare it with.
1000 Must be constant and idempotent, and should not take locks if
1001 possible, and should not or store into the dentry.
1002 Should not dereference pointers outside the dentry without
1003 lots of care (eg. d_parent, d_inode, d_name should not be used).
1005 However, our vfsmount is pinned, and RCU held, so the dentries and
1006 inodes won't disappear, neither will our sb or filesystem module.
1009 It is a tricky calling convention because it needs to be called under
1010 "rcu-walk", ie. without any locks or references on things.
1012 d_delete: called when the last reference to a dentry is dropped and the
1013 dcache is deciding whether or not to cache it. Return 1 to delete
1014 immediately, or 0 to cache the dentry. Default is NULL which means to
1015 always cache a reachable dentry. d_delete must be constant and
1018 d_release: called when a dentry is really deallocated
1020 d_iput: called when a dentry loses its inode (just prior to its
1021 being deallocated). The default when this is NULL is that the
1022 VFS calls iput(). If you define this method, you must call
1025 d_dname: called when the pathname of a dentry should be generated.
1026 Useful for some pseudo filesystems (sockfs, pipefs, ...) to delay
1027 pathname generation. (Instead of doing it when dentry is created,
1028 it's done only when the path is needed.). Real filesystems probably
1029 dont want to use it, because their dentries are present in global
1030 dcache hash, so their hash should be an invariant. As no lock is
1031 held, d_dname() should not try to modify the dentry itself, unless
1032 appropriate SMP safety is used. CAUTION : d_path() logic is quite
1033 tricky. The correct way to return for example "Hello" is to put it
1034 at the end of the buffer, and returns a pointer to the first char.
1035 dynamic_dname() helper function is provided to take care of this.
1037 d_automount: called when an automount dentry is to be traversed (optional).
1038 This should create a new VFS mount record and return the record to the
1039 caller. The caller is supplied with a path parameter giving the
1040 automount directory to describe the automount target and the parent
1041 VFS mount record to provide inheritable mount parameters. NULL should
1042 be returned if someone else managed to make the automount first. If
1043 the vfsmount creation failed, then an error code should be returned.
1044 If -EISDIR is returned, then the directory will be treated as an
1045 ordinary directory and returned to pathwalk to continue walking.
1047 If a vfsmount is returned, the caller will attempt to mount it on the
1048 mountpoint and will remove the vfsmount from its expiration list in
1049 the case of failure. The vfsmount should be returned with 2 refs on
1050 it to prevent automatic expiration - the caller will clean up the
1053 This function is only used if DCACHE_NEED_AUTOMOUNT is set on the
1054 dentry. This is set by __d_instantiate() if S_AUTOMOUNT is set on the
1057 d_manage: called to allow the filesystem to manage the transition from a
1058 dentry (optional). This allows autofs, for example, to hold up clients
1059 waiting to explore behind a 'mountpoint' whilst letting the daemon go
1060 past and construct the subtree there. 0 should be returned to let the
1061 calling process continue. -EISDIR can be returned to tell pathwalk to
1062 use this directory as an ordinary directory and to ignore anything
1063 mounted on it and not to check the automount flag. Any other error
1064 code will abort pathwalk completely.
1066 If the 'rcu_walk' parameter is true, then the caller is doing a
1067 pathwalk in RCU-walk mode. Sleeping is not permitted in this mode,
1068 and the caller can be asked to leave it and call again by returning
1069 -ECHILD. -EISDIR may also be returned to tell pathwalk to
1070 ignore d_automount or any mounts.
1072 This function is only used if DCACHE_MANAGE_TRANSIT is set on the
1073 dentry being transited from.
1077 static char *pipefs_dname(struct dentry *dent, char *buffer, int buflen)
1079 return dynamic_dname(dentry, buffer, buflen, "pipe:[%lu]",
1080 dentry->d_inode->i_ino);
1083 Each dentry has a pointer to its parent dentry, as well as a hash list
1084 of child dentries. Child dentries are basically like files in a
1088 Directory Entry Cache API
1089 --------------------------
1091 There are a number of functions defined which permit a filesystem to
1092 manipulate dentries:
1094 dget: open a new handle for an existing dentry (this just increments
1097 dput: close a handle for a dentry (decrements the usage count). If
1098 the usage count drops to 0, and the dentry is still in its
1099 parent's hash, the "d_delete" method is called to check whether
1100 it should be cached. If it should not be cached, or if the dentry
1101 is not hashed, it is deleted. Otherwise cached dentries are put
1102 into an LRU list to be reclaimed on memory shortage.
1104 d_drop: this unhashes a dentry from its parents hash list. A
1105 subsequent call to dput() will deallocate the dentry if its
1106 usage count drops to 0
1108 d_delete: delete a dentry. If there are no other open references to
1109 the dentry then the dentry is turned into a negative dentry
1110 (the d_iput() method is called). If there are other
1111 references, then d_drop() is called instead
1113 d_add: add a dentry to its parents hash list and then calls
1116 d_instantiate: add a dentry to the alias hash list for the inode and
1117 updates the "d_inode" member. The "i_count" member in the
1118 inode structure should be set/incremented. If the inode
1119 pointer is NULL, the dentry is called a "negative
1120 dentry". This function is commonly called when an inode is
1121 created for an existing negative dentry
1123 d_lookup: look up a dentry given its parent and path name component
1124 It looks up the child of that given name from the dcache
1125 hash table. If it is found, the reference count is incremented
1126 and the dentry is returned. The caller must use dput()
1127 to free the dentry when it finishes using it.
1135 On mount and remount the filesystem is passed a string containing a
1136 comma separated list of mount options. The options can have either of
1142 The <linux/parser.h> header defines an API that helps parse these
1143 options. There are plenty of examples on how to use it in existing
1149 If a filesystem accepts mount options, it must define show_options()
1150 to show all the currently active options. The rules are:
1152 - options MUST be shown which are not default or their values differ
1155 - options MAY be shown which are enabled by default or have their
1158 Options used only internally between a mount helper and the kernel
1159 (such as file descriptors), or which only have an effect during the
1160 mounting (such as ones controlling the creation of a journal) are exempt
1161 from the above rules.
1163 The underlying reason for the above rules is to make sure, that a
1164 mount can be accurately replicated (e.g. umounting and mounting again)
1165 based on the information found in /proc/mounts.
1167 A simple method of saving options at mount/remount time and showing
1168 them is provided with the save_mount_options() and
1169 generic_show_options() helper functions. Please note, that using
1170 these may have drawbacks. For more info see header comments for these
1171 functions in fs/namespace.c.
1176 (Note some of these resources are not up-to-date with the latest kernel
1179 Creating Linux virtual filesystems. 2002
1180 <http://lwn.net/Articles/13325/>
1182 The Linux Virtual File-system Layer by Neil Brown. 1999
1183 <http://www.cse.unsw.edu.au/~neilb/oss/linux-commentary/vfs.html>
1185 A tour of the Linux VFS by Michael K. Johnson. 1996
1186 <http://www.tldp.org/LDP/khg/HyperNews/get/fs/vfstour.html>
1188 A small trail through the Linux kernel by Andries Brouwer. 2001
1189 <http://www.win.tue.nl/~aeb/linux/vfs/trail.html>
OSDN Git Service
