1.. SPDX-License-Identifier: GPL-2.0 2 3========================================= 4Overview of the Linux Virtual File System 5========================================= 6 7Original author: Richard Gooch <rgooch@atnf.csiro.au> 8 9- Copyright (C) 1999 Richard Gooch 10- Copyright (C) 2005 Pekka Enberg 11 12 13Introduction 14============ 15 16The Virtual File System (also known as the Virtual Filesystem Switch) is 17the software layer in the kernel that provides the filesystem interface 18to userspace programs. It also provides an abstraction within the 19kernel which allows different filesystem implementations to coexist. 20 21VFS system calls open(2), stat(2), read(2), write(2), chmod(2) and so on 22are called from a process context. Filesystem locking is described in 23the document Documentation/filesystems/locking.rst. 24 25 26Directory Entry Cache (dcache) 27------------------------------ 28 29The VFS implements the open(2), stat(2), chmod(2), and similar system 30calls. The pathname argument that is passed to them is used by the VFS 31to search through the directory entry cache (also known as the dentry 32cache or dcache). This provides a very fast look-up mechanism to 33translate a pathname (filename) into a specific dentry. Dentries live 34in RAM and are never saved to disc: they exist only for performance. 35 36The dentry cache is meant to be a view into your entire filespace. As 37most computers cannot fit all dentries in the RAM at the same time, some 38bits of the cache are missing. In order to resolve your pathname into a 39dentry, the VFS may have to resort to creating dentries along the way, 40and then loading the inode. This is done by looking up the inode. 41 42 43The Inode Object 44---------------- 45 46An individual dentry usually has a pointer to an inode. Inodes are 47filesystem objects such as regular files, directories, FIFOs and other 48beasts. They live either on the disc (for block device filesystems) or 49in the memory (for pseudo filesystems). Inodes that live on the disc 50are copied into the memory when required and changes to the inode are 51written back to disc. A single inode can be pointed to by multiple 52dentries (hard links, for example, do this). 53 54To look up an inode requires that the VFS calls the lookup() method of 55the parent directory inode. This method is installed by the specific 56filesystem implementation that the inode lives in. Once the VFS has the 57required dentry (and hence the inode), we can do all those boring things 58like open(2) the file, or stat(2) it to peek at the inode data. The 59stat(2) operation is fairly simple: once the VFS has the dentry, it 60peeks at the inode data and passes some of it back to userspace. 61 62 63The File Object 64--------------- 65 66Opening a file requires another operation: allocation of a file 67structure (this is the kernel-side implementation of file descriptors). 68The freshly allocated file structure is initialized with a pointer to 69the dentry and a set of file operation member functions. These are 70taken from the inode data. The open() file method is then called so the 71specific filesystem implementation can do its work. You can see that 72this is another switch performed by the VFS. The file structure is 73placed into the file descriptor table for the process. 74 75Reading, writing and closing files (and other assorted VFS operations) 76is done by using the userspace file descriptor to grab the appropriate 77file structure, and then calling the required file structure method to 78do whatever is required. For as long as the file is open, it keeps the 79dentry in use, which in turn means that the VFS inode is still in use. 80 81 82Registering and Mounting a Filesystem 83===================================== 84 85To register and unregister a filesystem, use the following API 86functions: 87 88.. code-block:: c 89 90 #include <linux/fs.h> 91 92 extern int register_filesystem(struct file_system_type *); 93 extern int unregister_filesystem(struct file_system_type *); 94 95The passed struct file_system_type describes your filesystem. When a 96request is made to mount a filesystem onto a directory in your 97namespace, the VFS will call the appropriate mount() method for the 98specific filesystem. New vfsmount referring to the tree returned by 99->mount() will be attached to the mountpoint, so that when pathname 100resolution reaches the mountpoint it will jump into the root of that 101vfsmount. 102 103You can see all filesystems that are registered to the kernel in the 104file /proc/filesystems. 105 106 107struct file_system_type 108----------------------- 109 110This describes the filesystem. The following 111members are defined: 112 113.. code-block:: c 114 115 struct file_system_type { 116 const char *name; 117 int fs_flags; 118 int (*init_fs_context)(struct fs_context *); 119 const struct fs_parameter_spec *parameters; 120 struct dentry *(*mount) (struct file_system_type *, int, 121 const char *, void *); 122 void (*kill_sb) (struct super_block *); 123 struct module *owner; 124 struct file_system_type * next; 125 struct hlist_head fs_supers; 126 127 struct lock_class_key s_lock_key; 128 struct lock_class_key s_umount_key; 129 struct lock_class_key s_vfs_rename_key; 130 struct lock_class_key s_writers_key[SB_FREEZE_LEVELS]; 131 132 struct lock_class_key i_lock_key; 133 struct lock_class_key i_mutex_key; 134 struct lock_class_key invalidate_lock_key; 135 struct lock_class_key i_mutex_dir_key; 136 }; 137 138``name`` 139 the name of the filesystem type, such as "ext2", "iso9660", 140 "msdos" and so on 141 142``fs_flags`` 143 various flags (i.e. FS_REQUIRES_DEV, FS_NO_DCACHE, etc.) 144 145``init_fs_context`` 146 Initializes 'struct fs_context' ->ops and ->fs_private fields with 147 filesystem-specific data. 148 149``parameters`` 150 Pointer to the array of filesystem parameters descriptors 151 'struct fs_parameter_spec'. 152 More info in Documentation/filesystems/mount_api.rst. 153 154``mount`` 155 the method to call when a new instance of this filesystem should 156 be mounted 157 158``kill_sb`` 159 the method to call when an instance of this filesystem should be 160 shut down 161 162 163``owner`` 164 for internal VFS use: you should initialize this to THIS_MODULE 165 in most cases. 166 167``next`` 168 for internal VFS use: you should initialize this to NULL 169 170``fs_supers`` 171 for internal VFS use: hlist of filesystem instances (superblocks) 172 173 s_lock_key, s_umount_key, s_vfs_rename_key, s_writers_key, 174 i_lock_key, i_mutex_key, invalidate_lock_key, i_mutex_dir_key: lockdep-specific 175 176The mount() method has the following arguments: 177 178``struct file_system_type *fs_type`` 179 describes the filesystem, partly initialized by the specific 180 filesystem code 181 182``int flags`` 183 mount flags 184 185``const char *dev_name`` 186 the device name we are mounting. 187 188``void *data`` 189 arbitrary mount options, usually comes as an ASCII string (see 190 "Mount Options" section) 191 192The mount() method must return the root dentry of the tree requested by 193caller. An active reference to its superblock must be grabbed and the 194superblock must be locked. On failure it should return ERR_PTR(error). 195 196The arguments match those of mount(2) and their interpretation depends 197on filesystem type. E.g. for block filesystems, dev_name is interpreted 198as block device name, that device is opened and if it contains a 199suitable filesystem image the method creates and initializes struct 200super_block accordingly, returning its root dentry to caller. 201 202->mount() may choose to return a subtree of existing filesystem - it 203doesn't have to create a new one. The main result from the caller's 204point of view is a reference to dentry at the root of (sub)tree to be 205attached; creation of new superblock is a common side effect. 206 207The most interesting member of the superblock structure that the mount() 208method fills in is the "s_op" field. This is a pointer to a "struct 209super_operations" which describes the next level of the filesystem 210implementation. 211 212Usually, a filesystem uses one of the generic mount() implementations 213and provides a fill_super() callback instead. The generic variants are: 214 215``mount_bdev`` 216 mount a filesystem residing on a block device 217 218``mount_nodev`` 219 mount a filesystem that is not backed by a device 220 221``mount_single`` 222 mount a filesystem which shares the instance between all mounts 223 224A fill_super() callback implementation has the following arguments: 225 226``struct super_block *sb`` 227 the superblock structure. The callback must initialize this 228 properly. 229 230``void *data`` 231 arbitrary mount options, usually comes as an ASCII string (see 232 "Mount Options" section) 233 234``int silent`` 235 whether or not to be silent on error 236 237 238The Superblock Object 239===================== 240 241A superblock object represents a mounted filesystem. 242 243 244struct super_operations 245----------------------- 246 247This describes how the VFS can manipulate the superblock of your 248filesystem. The following members are defined: 249 250.. code-block:: c 251 252 struct super_operations { 253 struct inode *(*alloc_inode)(struct super_block *sb); 254 void (*destroy_inode)(struct inode *); 255 void (*free_inode)(struct inode *); 256 257 void (*dirty_inode) (struct inode *, int flags); 258 int (*write_inode) (struct inode *, struct writeback_control *wbc); 259 int (*drop_inode) (struct inode *); 260 void (*evict_inode) (struct inode *); 261 void (*put_super) (struct super_block *); 262 int (*sync_fs)(struct super_block *sb, int wait); 263 int (*freeze_super) (struct super_block *sb, 264 enum freeze_holder who); 265 int (*freeze_fs) (struct super_block *); 266 int (*thaw_super) (struct super_block *sb, 267 enum freeze_wholder who); 268 int (*unfreeze_fs) (struct super_block *); 269 int (*statfs) (struct dentry *, struct kstatfs *); 270 int (*remount_fs) (struct super_block *, int *, char *); 271 void (*umount_begin) (struct super_block *); 272 273 int (*show_options)(struct seq_file *, struct dentry *); 274 int (*show_devname)(struct seq_file *, struct dentry *); 275 int (*show_path)(struct seq_file *, struct dentry *); 276 int (*show_stats)(struct seq_file *, struct dentry *); 277 278 ssize_t (*quota_read)(struct super_block *, int, char *, size_t, loff_t); 279 ssize_t (*quota_write)(struct super_block *, int, const char *, size_t, loff_t); 280 struct dquot **(*get_dquots)(struct inode *); 281 282 long (*nr_cached_objects)(struct super_block *, 283 struct shrink_control *); 284 long (*free_cached_objects)(struct super_block *, 285 struct shrink_control *); 286 }; 287 288All methods are called without any locks being held, unless otherwise 289noted. This means that most methods can block safely. All methods are 290only called from a process context (i.e. not from an interrupt handler 291or bottom half). 292 293``alloc_inode`` 294 this method is called by alloc_inode() to allocate memory for 295 struct inode and initialize it. If this function is not 296 defined, a simple 'struct inode' is allocated. Normally 297 alloc_inode will be used to allocate a larger structure which 298 contains a 'struct inode' embedded within it. 299 300``destroy_inode`` 301 this method is called by destroy_inode() to release resources 302 allocated for struct inode. It is only required if 303 ->alloc_inode was defined and simply undoes anything done by 304 ->alloc_inode. 305 306``free_inode`` 307 this method is called from RCU callback. If you use call_rcu() 308 in ->destroy_inode to free 'struct inode' memory, then it's 309 better to release memory in this method. 310 311``dirty_inode`` 312 this method is called by the VFS when an inode is marked dirty. 313 This is specifically for the inode itself being marked dirty, 314 not its data. If the update needs to be persisted by fdatasync(), 315 then I_DIRTY_DATASYNC will be set in the flags argument. 316 I_DIRTY_TIME will be set in the flags in case lazytime is enabled 317 and struct inode has times updated since the last ->dirty_inode 318 call. 319 320``write_inode`` 321 this method is called when the VFS needs to write an inode to 322 disc. The second parameter indicates whether the write should 323 be synchronous or not, not all filesystems check this flag. 324 325``drop_inode`` 326 called when the last access to the inode is dropped, with the 327 inode->i_lock spinlock held. 328 329 This method should be either NULL (normal UNIX filesystem 330 semantics) or "generic_delete_inode" (for filesystems that do 331 not want to cache inodes - causing "delete_inode" to always be 332 called regardless of the value of i_nlink) 333 334 The "generic_delete_inode()" behavior is equivalent to the old 335 practice of using "force_delete" in the put_inode() case, but 336 does not have the races that the "force_delete()" approach had. 337 338``evict_inode`` 339 called when the VFS wants to evict an inode. Caller does 340 *not* evict the pagecache or inode-associated metadata buffers; 341 the method has to use truncate_inode_pages_final() to get rid 342 of those. Caller makes sure async writeback cannot be running for 343 the inode while (or after) ->evict_inode() is called. Optional. 344 345``put_super`` 346 called when the VFS wishes to free the superblock 347 (i.e. unmount). This is called with the superblock lock held 348 349``sync_fs`` 350 called when VFS is writing out all dirty data associated with a 351 superblock. The second parameter indicates whether the method 352 should wait until the write out has been completed. Optional. 353 354``freeze_super`` 355 Called instead of ->freeze_fs callback if provided. 356 Main difference is that ->freeze_super is called without taking 357 down_write(&sb->s_umount). If filesystem implements it and wants 358 ->freeze_fs to be called too, then it has to call ->freeze_fs 359 explicitly from this callback. Optional. 360 361``freeze_fs`` 362 called when VFS is locking a filesystem and forcing it into a 363 consistent state. This method is currently used by the Logical 364 Volume Manager (LVM) and ioctl(FIFREEZE). Optional. 365 366``thaw_super`` 367 called when VFS is unlocking a filesystem and making it writable 368 again after ->freeze_super. Optional. 369 370``unfreeze_fs`` 371 called when VFS is unlocking a filesystem and making it writable 372 again after ->freeze_fs. Optional. 373 374``statfs`` 375 called when the VFS needs to get filesystem statistics. 376 377``remount_fs`` 378 called when the filesystem is remounted. This is called with 379 the kernel lock held 380 381``umount_begin`` 382 called when the VFS is unmounting a filesystem. 383 384``show_options`` 385 called by the VFS to show mount options for /proc/<pid>/mounts 386 and /proc/<pid>/mountinfo. 387 (see "Mount Options" section) 388 389``show_devname`` 390 Optional. Called by the VFS to show device name for 391 /proc/<pid>/{mounts,mountinfo,mountstats}. If not provided then 392 '(struct mount).mnt_devname' will be used. 393 394``show_path`` 395 Optional. Called by the VFS (for /proc/<pid>/mountinfo) to show 396 the mount root dentry path relative to the filesystem root. 397 398``show_stats`` 399 Optional. Called by the VFS (for /proc/<pid>/mountstats) to show 400 filesystem-specific mount statistics. 401 402``quota_read`` 403 called by the VFS to read from filesystem quota file. 404 405``quota_write`` 406 called by the VFS to write to filesystem quota file. 407 408``get_dquots`` 409 called by quota to get 'struct dquot' array for a particular inode. 410 Optional. 411 412``nr_cached_objects`` 413 called by the sb cache shrinking function for the filesystem to 414 return the number of freeable cached objects it contains. 415 Optional. 416 417``free_cache_objects`` 418 called by the sb cache shrinking function for the filesystem to 419 scan the number of objects indicated to try to free them. 420 Optional, but any filesystem implementing this method needs to 421 also implement ->nr_cached_objects for it to be called 422 correctly. 423 424 We can't do anything with any errors that the filesystem might 425 encountered, hence the void return type. This will never be 426 called if the VM is trying to reclaim under GFP_NOFS conditions, 427 hence this method does not need to handle that situation itself. 428 429 Implementations must include conditional reschedule calls inside 430 any scanning loop that is done. This allows the VFS to 431 determine appropriate scan batch sizes without having to worry 432 about whether implementations will cause holdoff problems due to 433 large scan batch sizes. 434 435Whoever sets up the inode is responsible for filling in the "i_op" 436field. This is a pointer to a "struct inode_operations" which describes 437the methods that can be performed on individual inodes. 438 439 440struct xattr_handlers 441--------------------- 442 443On filesystems that support extended attributes (xattrs), the s_xattr 444superblock field points to a NULL-terminated array of xattr handlers. 445Extended attributes are name:value pairs. 446 447``name`` 448 Indicates that the handler matches attributes with the specified 449 name (such as "system.posix_acl_access"); the prefix field must 450 be NULL. 451 452``prefix`` 453 Indicates that the handler matches all attributes with the 454 specified name prefix (such as "user."); the name field must be 455 NULL. 456 457``list`` 458 Determine if attributes matching this xattr handler should be 459 listed for a particular dentry. Used by some listxattr 460 implementations like generic_listxattr. 461 462``get`` 463 Called by the VFS to get the value of a particular extended 464 attribute. This method is called by the getxattr(2) system 465 call. 466 467``set`` 468 Called by the VFS to set the value of a particular extended 469 attribute. When the new value is NULL, called to remove a 470 particular extended attribute. This method is called by the 471 setxattr(2) and removexattr(2) system calls. 472 473When none of the xattr handlers of a filesystem match the specified 474attribute name or when a filesystem doesn't support extended attributes, 475the various ``*xattr(2)`` system calls return -EOPNOTSUPP. 476 477 478The Inode Object 479================ 480 481An inode object represents an object within the filesystem. 482 483 484struct inode_operations 485----------------------- 486 487This describes how the VFS can manipulate an inode in your filesystem. 488As of kernel 2.6.22, the following members are defined: 489 490.. code-block:: c 491 492 struct inode_operations { 493 int (*create) (struct mnt_idmap *, struct inode *,struct dentry *, umode_t, bool); 494 struct dentry * (*lookup) (struct inode *,struct dentry *, unsigned int); 495 int (*link) (struct dentry *,struct inode *,struct dentry *); 496 int (*unlink) (struct inode *,struct dentry *); 497 int (*symlink) (struct mnt_idmap *, struct inode *,struct dentry *,const char *); 498 int (*mkdir) (struct mnt_idmap *, struct inode *,struct dentry *,umode_t); 499 int (*rmdir) (struct inode *,struct dentry *); 500 int (*mknod) (struct mnt_idmap *, struct inode *,struct dentry *,umode_t,dev_t); 501 int (*rename) (struct mnt_idmap *, struct inode *, struct dentry *, 502 struct inode *, struct dentry *, unsigned int); 503 int (*readlink) (struct dentry *, char __user *,int); 504 const char *(*get_link) (struct dentry *, struct inode *, 505 struct delayed_call *); 506 int (*permission) (struct mnt_idmap *, struct inode *, int); 507 struct posix_acl * (*get_inode_acl)(struct inode *, int, bool); 508 int (*setattr) (struct mnt_idmap *, struct dentry *, struct iattr *); 509 int (*getattr) (struct mnt_idmap *, const struct path *, struct kstat *, u32, unsigned int); 510 ssize_t (*listxattr) (struct dentry *, char *, size_t); 511 void (*update_time)(struct inode *, struct timespec *, int); 512 int (*atomic_open)(struct inode *, struct dentry *, struct file *, 513 unsigned open_flag, umode_t create_mode); 514 int (*tmpfile) (struct mnt_idmap *, struct inode *, struct file *, umode_t); 515 struct posix_acl * (*get_acl)(struct mnt_idmap *, struct dentry *, int); 516 int (*set_acl)(struct mnt_idmap *, struct dentry *, struct posix_acl *, int); 517 int (*fileattr_set)(struct mnt_idmap *idmap, 518 struct dentry *dentry, struct fileattr *fa); 519 int (*fileattr_get)(struct dentry *dentry, struct fileattr *fa); 520 }; 521 522Again, all methods are called without any locks being held, unless 523otherwise noted. 524 525``create`` 526 called by the open(2) and creat(2) system calls. Only required 527 if you want to support regular files. The dentry you get should 528 not have an inode (i.e. it should be a negative dentry). Here 529 you will probably call d_instantiate() with the dentry and the 530 newly created inode 531 532``lookup`` 533 called when the VFS needs to look up an inode in a parent 534 directory. The name to look for is found in the dentry. This 535 method must call d_add() to insert the found inode into the 536 dentry. The "i_count" field in the inode structure should be 537 incremented. If the named inode does not exist a NULL inode 538 should be inserted into the dentry (this is called a negative 539 dentry). Returning an error code from this routine must only be 540 done on a real error, otherwise creating inodes with system 541 calls like create(2), mknod(2), mkdir(2) and so on will fail. 542 If you wish to overload the dentry methods then you should 543 initialise the "d_dop" field in the dentry; this is a pointer to 544 a struct "dentry_operations". This method is called with the 545 directory inode semaphore held 546 547``link`` 548 called by the link(2) system call. Only required if you want to 549 support hard links. You will probably need to call 550 d_instantiate() just as you would in the create() method 551 552``unlink`` 553 called by the unlink(2) system call. Only required if you want 554 to support deleting inodes 555 556``symlink`` 557 called by the symlink(2) system call. Only required if you want 558 to support symlinks. You will probably need to call 559 d_instantiate() just as you would in the create() method 560 561``mkdir`` 562 called by the mkdir(2) system call. Only required if you want 563 to support creating subdirectories. You will probably need to 564 call d_instantiate() just as you would in the create() method 565 566``rmdir`` 567 called by the rmdir(2) system call. Only required if you want 568 to support deleting subdirectories 569 570``mknod`` 571 called by the mknod(2) system call to create a device (char, 572 block) inode or a named pipe (FIFO) or socket. Only required if 573 you want to support creating these types of inodes. You will 574 probably need to call d_instantiate() just as you would in the 575 create() method 576 577``rename`` 578 called by the rename(2) system call to rename the object to have 579 the parent and name given by the second inode and dentry. 580 581 The filesystem must return -EINVAL for any unsupported or 582 unknown flags. Currently the following flags are implemented: 583 (1) RENAME_NOREPLACE: this flag indicates that if the target of 584 the rename exists the rename should fail with -EEXIST instead of 585 replacing the target. The VFS already checks for existence, so 586 for local filesystems the RENAME_NOREPLACE implementation is 587 equivalent to plain rename. 588 (2) RENAME_EXCHANGE: exchange source and target. Both must 589 exist; this is checked by the VFS. Unlike plain rename, source 590 and target may be of different type. 591 592``get_link`` 593 called by the VFS to follow a symbolic link to the inode it 594 points to. Only required if you want to support symbolic links. 595 This method returns the symlink body to traverse (and possibly 596 resets the current position with nd_jump_link()). If the body 597 won't go away until the inode is gone, nothing else is needed; 598 if it needs to be otherwise pinned, arrange for its release by 599 having get_link(..., ..., done) do set_delayed_call(done, 600 destructor, argument). In that case destructor(argument) will 601 be called once VFS is done with the body you've returned. May 602 be called in RCU mode; that is indicated by NULL dentry 603 argument. If request can't be handled without leaving RCU mode, 604 have it return ERR_PTR(-ECHILD). 605 606 If the filesystem stores the symlink target in ->i_link, the 607 VFS may use it directly without calling ->get_link(); however, 608 ->get_link() must still be provided. ->i_link must not be 609 freed until after an RCU grace period. Writing to ->i_link 610 post-iget() time requires a 'release' memory barrier. 611 612``readlink`` 613 this is now just an override for use by readlink(2) for the 614 cases when ->get_link uses nd_jump_link() or object is not in 615 fact a symlink. Normally filesystems should only implement 616 ->get_link for symlinks and readlink(2) will automatically use 617 that. 618 619``permission`` 620 called by the VFS to check for access rights on a POSIX-like 621 filesystem. 622 623 May be called in rcu-walk mode (mask & MAY_NOT_BLOCK). If in 624 rcu-walk mode, the filesystem must check the permission without 625 blocking or storing to the inode. 626 627 If a situation is encountered that rcu-walk cannot handle, 628 return 629 -ECHILD and it will be called again in ref-walk mode. 630 631``setattr`` 632 called by the VFS to set attributes for a file. This method is 633 called by chmod(2) and related system calls. 634 635``getattr`` 636 called by the VFS to get attributes of a file. This method is 637 called by stat(2) and related system calls. 638 639``listxattr`` 640 called by the VFS to list all extended attributes for a given 641 file. This method is called by the listxattr(2) system call. 642 643``update_time`` 644 called by the VFS to update a specific time or the i_version of 645 an inode. If this is not defined the VFS will update the inode 646 itself and call mark_inode_dirty_sync. 647 648``atomic_open`` 649 called on the last component of an open. Using this optional 650 method the filesystem can look up, possibly create and open the 651 file in one atomic operation. If it wants to leave actual 652 opening to the caller (e.g. if the file turned out to be a 653 symlink, device, or just something filesystem won't do atomic 654 open for), it may signal this by returning finish_no_open(file, 655 dentry). This method is only called if the last component is 656 negative or needs lookup. Cached positive dentries are still 657 handled by f_op->open(). If the file was created, FMODE_CREATED 658 flag should be set in file->f_mode. In case of O_EXCL the 659 method must only succeed if the file didn't exist and hence 660 FMODE_CREATED shall always be set on success. 661 662``tmpfile`` 663 called in the end of O_TMPFILE open(). Optional, equivalent to 664 atomically creating, opening and unlinking a file in given 665 directory. On success needs to return with the file already 666 open; this can be done by calling finish_open_simple() right at 667 the end. 668 669``fileattr_get`` 670 called on ioctl(FS_IOC_GETFLAGS) and ioctl(FS_IOC_FSGETXATTR) to 671 retrieve miscellaneous file flags and attributes. Also called 672 before the relevant SET operation to check what is being changed 673 (in this case with i_rwsem locked exclusive). If unset, then 674 fall back to f_op->ioctl(). 675 676``fileattr_set`` 677 called on ioctl(FS_IOC_SETFLAGS) and ioctl(FS_IOC_FSSETXATTR) to 678 change miscellaneous file flags and attributes. Callers hold 679 i_rwsem exclusive. If unset, then fall back to f_op->ioctl(). 680 681 682The Address Space Object 683======================== 684 685The address space object is used to group and manage pages in the page 686cache. It can be used to keep track of the pages in a file (or anything 687else) and also track the mapping of sections of the file into process 688address spaces. 689 690There are a number of distinct yet related services that an 691address-space can provide. These include communicating memory pressure, 692page lookup by address, and keeping track of pages tagged as Dirty or 693Writeback. 694 695The first can be used independently to the others. The VM can try to 696either write dirty pages in order to clean them, or release clean pages 697in order to reuse them. To do this it can call the ->writepage method 698on dirty pages, and ->release_folio on clean folios with the private 699flag set. Clean pages without PagePrivate and with no external references 700will be released without notice being given to the address_space. 701 702To achieve this functionality, pages need to be placed on an LRU with 703lru_cache_add and mark_page_active needs to be called whenever the page 704is used. 705 706Pages are normally kept in a radix tree index by ->index. This tree 707maintains information about the PG_Dirty and PG_Writeback status of each 708page, so that pages with either of these flags can be found quickly. 709 710The Dirty tag is primarily used by mpage_writepages - the default 711->writepages method. It uses the tag to find dirty pages to call 712->writepage on. If mpage_writepages is not used (i.e. the address 713provides its own ->writepages) , the PAGECACHE_TAG_DIRTY tag is almost 714unused. write_inode_now and sync_inode do use it (through 715__sync_single_inode) to check if ->writepages has been successful in 716writing out the whole address_space. 717 718The Writeback tag is used by filemap*wait* and sync_page* functions, via 719filemap_fdatawait_range, to wait for all writeback to complete. 720 721An address_space handler may attach extra information to a page, 722typically using the 'private' field in the 'struct page'. If such 723information is attached, the PG_Private flag should be set. This will 724cause various VM routines to make extra calls into the address_space 725handler to deal with that data. 726 727An address space acts as an intermediate between storage and 728application. Data is read into the address space a whole page at a 729time, and provided to the application either by copying of the page, or 730by memory-mapping the page. Data is written into the address space by 731the application, and then written-back to storage typically in whole 732pages, however the address_space has finer control of write sizes. 733 734The read process essentially only requires 'read_folio'. The write 735process is more complicated and uses write_begin/write_end or 736dirty_folio to write data into the address_space, and writepage and 737writepages to writeback data to storage. 738 739Adding and removing pages to/from an address_space is protected by the 740inode's i_mutex. 741 742When data is written to a page, the PG_Dirty flag should be set. It 743typically remains set until writepage asks for it to be written. This 744should clear PG_Dirty and set PG_Writeback. It can be actually written 745at any point after PG_Dirty is clear. Once it is known to be safe, 746PG_Writeback is cleared. 747 748Writeback makes use of a writeback_control structure to direct the 749operations. This gives the writepage and writepages operations some 750information about the nature of and reason for the writeback request, 751and the constraints under which it is being done. It is also used to 752return information back to the caller about the result of a writepage or 753writepages request. 754 755 756Handling errors during writeback 757-------------------------------- 758 759Most applications that do buffered I/O will periodically call a file 760synchronization call (fsync, fdatasync, msync or sync_file_range) to 761ensure that data written has made it to the backing store. When there 762is an error during writeback, they expect that error to be reported when 763a file sync request is made. After an error has been reported on one 764request, subsequent requests on the same file descriptor should return 7650, unless further writeback errors have occurred since the previous file 766syncronization. 767 768Ideally, the kernel would report errors only on file descriptions on 769which writes were done that subsequently failed to be written back. The 770generic pagecache infrastructure does not track the file descriptions 771that have dirtied each individual page however, so determining which 772file descriptors should get back an error is not possible. 773 774Instead, the generic writeback error tracking infrastructure in the 775kernel settles for reporting errors to fsync on all file descriptions 776that were open at the time that the error occurred. In a situation with 777multiple writers, all of them will get back an error on a subsequent 778fsync, even if all of the writes done through that particular file 779descriptor succeeded (or even if there were no writes on that file 780descriptor at all). 781 782Filesystems that wish to use this infrastructure should call 783mapping_set_error to record the error in the address_space when it 784occurs. Then, after writing back data from the pagecache in their 785file->fsync operation, they should call file_check_and_advance_wb_err to 786ensure that the struct file's error cursor has advanced to the correct 787point in the stream of errors emitted by the backing device(s). 788 789 790struct address_space_operations 791------------------------------- 792 793This describes how the VFS can manipulate mapping of a file to page 794cache in your filesystem. The following members are defined: 795 796.. code-block:: c 797 798 struct address_space_operations { 799 int (*writepage)(struct page *page, struct writeback_control *wbc); 800 int (*read_folio)(struct file *, struct folio *); 801 int (*writepages)(struct address_space *, struct writeback_control *); 802 bool (*dirty_folio)(struct address_space *, struct folio *); 803 void (*readahead)(struct readahead_control *); 804 int (*write_begin)(struct file *, struct address_space *mapping, 805 loff_t pos, unsigned len, 806 struct page **pagep, void **fsdata); 807 int (*write_end)(struct file *, struct address_space *mapping, 808 loff_t pos, unsigned len, unsigned copied, 809 struct page *page, void *fsdata); 810 sector_t (*bmap)(struct address_space *, sector_t); 811 void (*invalidate_folio) (struct folio *, size_t start, size_t len); 812 bool (*release_folio)(struct folio *, gfp_t); 813 void (*free_folio)(struct folio *); 814 ssize_t (*direct_IO)(struct kiocb *, struct iov_iter *iter); 815 int (*migrate_folio)(struct mapping *, struct folio *dst, 816 struct folio *src, enum migrate_mode); 817 int (*launder_folio) (struct folio *); 818 819 bool (*is_partially_uptodate) (struct folio *, size_t from, 820 size_t count); 821 void (*is_dirty_writeback)(struct folio *, bool *, bool *); 822 int (*error_remove_page) (struct mapping *mapping, struct page *page); 823 int (*swap_activate)(struct swap_info_struct *sis, struct file *f, sector_t *span) 824 int (*swap_deactivate)(struct file *); 825 int (*swap_rw)(struct kiocb *iocb, struct iov_iter *iter); 826 }; 827 828``writepage`` 829 called by the VM to write a dirty page to backing store. This 830 may happen for data integrity reasons (i.e. 'sync'), or to free 831 up memory (flush). The difference can be seen in 832 wbc->sync_mode. The PG_Dirty flag has been cleared and 833 PageLocked is true. writepage should start writeout, should set 834 PG_Writeback, and should make sure the page is unlocked, either 835 synchronously or asynchronously when the write operation 836 completes. 837 838 If wbc->sync_mode is WB_SYNC_NONE, ->writepage doesn't have to 839 try too hard if there are problems, and may choose to write out 840 other pages from the mapping if that is easier (e.g. due to 841 internal dependencies). If it chooses not to start writeout, it 842 should return AOP_WRITEPAGE_ACTIVATE so that the VM will not 843 keep calling ->writepage on that page. 844 845 See the file "Locking" for more details. 846 847``read_folio`` 848 Called by the page cache to read a folio from the backing store. 849 The 'file' argument supplies authentication information to network 850 filesystems, and is generally not used by block based filesystems. 851 It may be NULL if the caller does not have an open file (eg if 852 the kernel is performing a read for itself rather than on behalf 853 of a userspace process with an open file). 854 855 If the mapping does not support large folios, the folio will 856 contain a single page. The folio will be locked when read_folio 857 is called. If the read completes successfully, the folio should 858 be marked uptodate. The filesystem should unlock the folio 859 once the read has completed, whether it was successful or not. 860 The filesystem does not need to modify the refcount on the folio; 861 the page cache holds a reference count and that will not be 862 released until the folio is unlocked. 863 864 Filesystems may implement ->read_folio() synchronously. 865 In normal operation, folios are read through the ->readahead() 866 method. Only if this fails, or if the caller needs to wait for 867 the read to complete will the page cache call ->read_folio(). 868 Filesystems should not attempt to perform their own readahead 869 in the ->read_folio() operation. 870 871 If the filesystem cannot perform the read at this time, it can 872 unlock the folio, do whatever action it needs to ensure that the 873 read will succeed in the future and return AOP_TRUNCATED_PAGE. 874 In this case, the caller should look up the folio, lock it, 875 and call ->read_folio again. 876 877 Callers may invoke the ->read_folio() method directly, but using 878 read_mapping_folio() will take care of locking, waiting for the 879 read to complete and handle cases such as AOP_TRUNCATED_PAGE. 880 881``writepages`` 882 called by the VM to write out pages associated with the 883 address_space object. If wbc->sync_mode is WB_SYNC_ALL, then 884 the writeback_control will specify a range of pages that must be 885 written out. If it is WB_SYNC_NONE, then a nr_to_write is 886 given and that many pages should be written if possible. If no 887 ->writepages is given, then mpage_writepages is used instead. 888 This will choose pages from the address space that are tagged as 889 DIRTY and will pass them to ->writepage. 890 891``dirty_folio`` 892 called by the VM to mark a folio as dirty. This is particularly 893 needed if an address space attaches private data to a folio, and 894 that data needs to be updated when a folio is dirtied. This is 895 called, for example, when a memory mapped page gets modified. 896 If defined, it should set the folio dirty flag, and the 897 PAGECACHE_TAG_DIRTY search mark in i_pages. 898 899``readahead`` 900 Called by the VM to read pages associated with the address_space 901 object. The pages are consecutive in the page cache and are 902 locked. The implementation should decrement the page refcount 903 after starting I/O on each page. Usually the page will be 904 unlocked by the I/O completion handler. The set of pages are 905 divided into some sync pages followed by some async pages, 906 rac->ra->async_size gives the number of async pages. The 907 filesystem should attempt to read all sync pages but may decide 908 to stop once it reaches the async pages. If it does decide to 909 stop attempting I/O, it can simply return. The caller will 910 remove the remaining pages from the address space, unlock them 911 and decrement the page refcount. Set PageUptodate if the I/O 912 completes successfully. Setting PageError on any page will be 913 ignored; simply unlock the page if an I/O error occurs. 914 915``write_begin`` 916 Called by the generic buffered write code to ask the filesystem 917 to prepare to write len bytes at the given offset in the file. 918 The address_space should check that the write will be able to 919 complete, by allocating space if necessary and doing any other 920 internal housekeeping. If the write will update parts of any 921 basic-blocks on storage, then those blocks should be pre-read 922 (if they haven't been read already) so that the updated blocks 923 can be written out properly. 924 925 The filesystem must return the locked pagecache page for the 926 specified offset, in ``*pagep``, for the caller to write into. 927 928 It must be able to cope with short writes (where the length 929 passed to write_begin is greater than the number of bytes copied 930 into the page). 931 932 A void * may be returned in fsdata, which then gets passed into 933 write_end. 934 935 Returns 0 on success; < 0 on failure (which is the error code), 936 in which case write_end is not called. 937 938``write_end`` 939 After a successful write_begin, and data copy, write_end must be 940 called. len is the original len passed to write_begin, and 941 copied is the amount that was able to be copied. 942 943 The filesystem must take care of unlocking the page and 944 releasing it refcount, and updating i_size. 945 946 Returns < 0 on failure, otherwise the number of bytes (<= 947 'copied') that were able to be copied into pagecache. 948 949``bmap`` 950 called by the VFS to map a logical block offset within object to 951 physical block number. This method is used by the FIBMAP ioctl 952 and for working with swap-files. To be able to swap to a file, 953 the file must have a stable mapping to a block device. The swap 954 system does not go through the filesystem but instead uses bmap 955 to find out where the blocks in the file are and uses those 956 addresses directly. 957 958``invalidate_folio`` 959 If a folio has private data, then invalidate_folio will be 960 called when part or all of the folio is to be removed from the 961 address space. This generally corresponds to either a 962 truncation, punch hole or a complete invalidation of the address 963 space (in the latter case 'offset' will always be 0 and 'length' 964 will be folio_size()). Any private data associated with the folio 965 should be updated to reflect this truncation. If offset is 0 966 and length is folio_size(), then the private data should be 967 released, because the folio must be able to be completely 968 discarded. This may be done by calling the ->release_folio 969 function, but in this case the release MUST succeed. 970 971``release_folio`` 972 release_folio is called on folios with private data to tell the 973 filesystem that the folio is about to be freed. ->release_folio 974 should remove any private data from the folio and clear the 975 private flag. If release_folio() fails, it should return false. 976 release_folio() is used in two distinct though related cases. 977 The first is when the VM wants to free a clean folio with no 978 active users. If ->release_folio succeeds, the folio will be 979 removed from the address_space and be freed. 980 981 The second case is when a request has been made to invalidate 982 some or all folios in an address_space. This can happen 983 through the fadvise(POSIX_FADV_DONTNEED) system call or by the 984 filesystem explicitly requesting it as nfs and 9p do (when they 985 believe the cache may be out of date with storage) by calling 986 invalidate_inode_pages2(). If the filesystem makes such a call, 987 and needs to be certain that all folios are invalidated, then 988 its release_folio will need to ensure this. Possibly it can 989 clear the uptodate flag if it cannot free private data yet. 990 991``free_folio`` 992 free_folio is called once the folio is no longer visible in the 993 page cache in order to allow the cleanup of any private data. 994 Since it may be called by the memory reclaimer, it should not 995 assume that the original address_space mapping still exists, and 996 it should not block. 997 998``direct_IO`` 999 called by the generic read/write routines to perform direct_IO - 1000 that is IO requests which bypass the page cache and transfer 1001 data directly between the storage and the application's address 1002 space. 1003 1004``migrate_folio`` 1005 This is used to compact the physical memory usage. If the VM 1006 wants to relocate a folio (maybe from a memory device that is 1007 signalling imminent failure) it will pass a new folio and an old 1008 folio to this function. migrate_folio should transfer any private 1009 data across and update any references that it has to the folio. 1010 1011``launder_folio`` 1012 Called before freeing a folio - it writes back the dirty folio. 1013 To prevent redirtying the folio, it is kept locked during the 1014 whole operation. 1015 1016``is_partially_uptodate`` 1017 Called by the VM when reading a file through the pagecache when 1018 the underlying blocksize is smaller than the size of the folio. 1019 If the required block is up to date then the read can complete 1020 without needing I/O to bring the whole page up to date. 1021 1022``is_dirty_writeback`` 1023 Called by the VM when attempting to reclaim a folio. The VM uses 1024 dirty and writeback information to determine if it needs to 1025 stall to allow flushers a chance to complete some IO. 1026 Ordinarily it can use folio_test_dirty and folio_test_writeback but 1027 some filesystems have more complex state (unstable folios in NFS 1028 prevent reclaim) or do not set those flags due to locking 1029 problems. This callback allows a filesystem to indicate to the 1030 VM if a folio should be treated as dirty or writeback for the 1031 purposes of stalling. 1032 1033``error_remove_page`` 1034 normally set to generic_error_remove_page if truncation is ok 1035 for this address space. Used for memory failure handling. 1036 Setting this implies you deal with pages going away under you, 1037 unless you have them locked or reference counts increased. 1038 1039``swap_activate`` 1040 1041 Called to prepare the given file for swap. It should perform 1042 any validation and preparation necessary to ensure that writes 1043 can be performed with minimal memory allocation. It should call 1044 add_swap_extent(), or the helper iomap_swapfile_activate(), and 1045 return the number of extents added. If IO should be submitted 1046 through ->swap_rw(), it should set SWP_FS_OPS, otherwise IO will 1047 be submitted directly to the block device ``sis->bdev``. 1048 1049``swap_deactivate`` 1050 Called during swapoff on files where swap_activate was 1051 successful. 1052 1053``swap_rw`` 1054 Called to read or write swap pages when SWP_FS_OPS is set. 1055 1056The File Object 1057=============== 1058 1059A file object represents a file opened by a process. This is also known 1060as an "open file description" in POSIX parlance. 1061 1062 1063struct file_operations 1064---------------------- 1065 1066This describes how the VFS can manipulate an open file. As of kernel 10674.18, the following members are defined: 1068 1069.. code-block:: c 1070 1071 struct file_operations { 1072 struct module *owner; 1073 loff_t (*llseek) (struct file *, loff_t, int); 1074 ssize_t (*read) (struct file *, char __user *, size_t, loff_t *); 1075 ssize_t (*write) (struct file *, const char __user *, size_t, loff_t *); 1076 ssize_t (*read_iter) (struct kiocb *, struct iov_iter *); 1077 ssize_t (*write_iter) (struct kiocb *, struct iov_iter *); 1078 int (*iopoll)(struct kiocb *kiocb, bool spin); 1079 int (*iterate) (struct file *, struct dir_context *); 1080 int (*iterate_shared) (struct file *, struct dir_context *); 1081 __poll_t (*poll) (struct file *, struct poll_table_struct *); 1082 long (*unlocked_ioctl) (struct file *, unsigned int, unsigned long); 1083 long (*compat_ioctl) (struct file *, unsigned int, unsigned long); 1084 int (*mmap) (struct file *, struct vm_area_struct *); 1085 int (*open) (struct inode *, struct file *); 1086 int (*flush) (struct file *, fl_owner_t id); 1087 int (*release) (struct inode *, struct file *); 1088 int (*fsync) (struct file *, loff_t, loff_t, int datasync); 1089 int (*fasync) (int, struct file *, int); 1090 int (*lock) (struct file *, int, struct file_lock *); 1091 unsigned long (*get_unmapped_area)(struct file *, unsigned long, unsigned long, unsigned long, unsigned long); 1092 int (*check_flags)(int); 1093 int (*flock) (struct file *, int, struct file_lock *); 1094 ssize_t (*splice_write)(struct pipe_inode_info *, struct file *, loff_t *, size_t, unsigned int); 1095 ssize_t (*splice_read)(struct file *, loff_t *, struct pipe_inode_info *, size_t, unsigned int); 1096 int (*setlease)(struct file *, long, struct file_lock **, void **); 1097 long (*fallocate)(struct file *file, int mode, loff_t offset, 1098 loff_t len); 1099 void (*show_fdinfo)(struct seq_file *m, struct file *f); 1100 #ifndef CONFIG_MMU 1101 unsigned (*mmap_capabilities)(struct file *); 1102 #endif 1103 ssize_t (*copy_file_range)(struct file *, loff_t, struct file *, loff_t, size_t, unsigned int); 1104 loff_t (*remap_file_range)(struct file *file_in, loff_t pos_in, 1105 struct file *file_out, loff_t pos_out, 1106 loff_t len, unsigned int remap_flags); 1107 int (*fadvise)(struct file *, loff_t, loff_t, int); 1108 }; 1109 1110Again, all methods are called without any locks being held, unless 1111otherwise noted. 1112 1113``llseek`` 1114 called when the VFS needs to move the file position index 1115 1116``read`` 1117 called by read(2) and related system calls 1118 1119``read_iter`` 1120 possibly asynchronous read with iov_iter as destination 1121 1122``write`` 1123 called by write(2) and related system calls 1124 1125``write_iter`` 1126 possibly asynchronous write with iov_iter as source 1127 1128``iopoll`` 1129 called when aio wants to poll for completions on HIPRI iocbs 1130 1131``iterate`` 1132 called when the VFS needs to read the directory contents 1133 1134``iterate_shared`` 1135 called when the VFS needs to read the directory contents when 1136 filesystem supports concurrent dir iterators 1137 1138``poll`` 1139 called by the VFS when a process wants to check if there is 1140 activity on this file and (optionally) go to sleep until there 1141 is activity. Called by the select(2) and poll(2) system calls 1142 1143``unlocked_ioctl`` 1144 called by the ioctl(2) system call. 1145 1146``compat_ioctl`` 1147 called by the ioctl(2) system call when 32 bit system calls are 1148 used on 64 bit kernels. 1149 1150``mmap`` 1151 called by the mmap(2) system call 1152 1153``open`` 1154 called by the VFS when an inode should be opened. When the VFS 1155 opens a file, it creates a new "struct file". It then calls the 1156 open method for the newly allocated file structure. You might 1157 think that the open method really belongs in "struct 1158 inode_operations", and you may be right. I think it's done the 1159 way it is because it makes filesystems simpler to implement. 1160 The open() method is a good place to initialize the 1161 "private_data" member in the file structure if you want to point 1162 to a device structure 1163 1164``flush`` 1165 called by the close(2) system call to flush a file 1166 1167``release`` 1168 called when the last reference to an open file is closed 1169 1170``fsync`` 1171 called by the fsync(2) system call. Also see the section above 1172 entitled "Handling errors during writeback". 1173 1174``fasync`` 1175 called by the fcntl(2) system call when asynchronous 1176 (non-blocking) mode is enabled for a file 1177 1178``lock`` 1179 called by the fcntl(2) system call for F_GETLK, F_SETLK, and 1180 F_SETLKW commands 1181 1182``get_unmapped_area`` 1183 called by the mmap(2) system call 1184 1185``check_flags`` 1186 called by the fcntl(2) system call for F_SETFL command 1187 1188``flock`` 1189 called by the flock(2) system call 1190 1191``splice_write`` 1192 called by the VFS to splice data from a pipe to a file. This 1193 method is used by the splice(2) system call 1194 1195``splice_read`` 1196 called by the VFS to splice data from file to a pipe. This 1197 method is used by the splice(2) system call 1198 1199``setlease`` 1200 called by the VFS to set or release a file lock lease. setlease 1201 implementations should call generic_setlease to record or remove 1202 the lease in the inode after setting it. 1203 1204``fallocate`` 1205 called by the VFS to preallocate blocks or punch a hole. 1206 1207``copy_file_range`` 1208 called by the copy_file_range(2) system call. 1209 1210``remap_file_range`` 1211 called by the ioctl(2) system call for FICLONERANGE and FICLONE 1212 and FIDEDUPERANGE commands to remap file ranges. An 1213 implementation should remap len bytes at pos_in of the source 1214 file into the dest file at pos_out. Implementations must handle 1215 callers passing in len == 0; this means "remap to the end of the 1216 source file". The return value should the number of bytes 1217 remapped, or the usual negative error code if errors occurred 1218 before any bytes were remapped. The remap_flags parameter 1219 accepts REMAP_FILE_* flags. If REMAP_FILE_DEDUP is set then the 1220 implementation must only remap if the requested file ranges have 1221 identical contents. If REMAP_FILE_CAN_SHORTEN is set, the caller is 1222 ok with the implementation shortening the request length to 1223 satisfy alignment or EOF requirements (or any other reason). 1224 1225``fadvise`` 1226 possibly called by the fadvise64() system call. 1227 1228Note that the file operations are implemented by the specific 1229filesystem in which the inode resides. When opening a device node 1230(character or block special) most filesystems will call special 1231support routines in the VFS which will locate the required device 1232driver information. These support routines replace the filesystem file 1233operations with those for the device driver, and then proceed to call 1234the new open() method for the file. This is how opening a device file 1235in the filesystem eventually ends up calling the device driver open() 1236method. 1237 1238 1239Directory Entry Cache (dcache) 1240============================== 1241 1242 1243struct dentry_operations 1244------------------------ 1245 1246This describes how a filesystem can overload the standard dentry 1247operations. Dentries and the dcache are the domain of the VFS and the 1248individual filesystem implementations. Device drivers have no business 1249here. These methods may be set to NULL, as they are either optional or 1250the VFS uses a default. As of kernel 2.6.22, the following members are 1251defined: 1252 1253.. code-block:: c 1254 1255 struct dentry_operations { 1256 int (*d_revalidate)(struct dentry *, unsigned int); 1257 int (*d_weak_revalidate)(struct dentry *, unsigned int); 1258 int (*d_hash)(const struct dentry *, struct qstr *); 1259 int (*d_compare)(const struct dentry *, 1260 unsigned int, const char *, const struct qstr *); 1261 int (*d_delete)(const struct dentry *); 1262 int (*d_init)(struct dentry *); 1263 void (*d_release)(struct dentry *); 1264 void (*d_iput)(struct dentry *, struct inode *); 1265 char *(*d_dname)(struct dentry *, char *, int); 1266 struct vfsmount *(*d_automount)(struct path *); 1267 int (*d_manage)(const struct path *, bool); 1268 struct dentry *(*d_real)(struct dentry *, const struct inode *); 1269 }; 1270 1271``d_revalidate`` 1272 called when the VFS needs to revalidate a dentry. This is 1273 called whenever a name look-up finds a dentry in the dcache. 1274 Most local filesystems leave this as NULL, because all their 1275 dentries in the dcache are valid. Network filesystems are 1276 different since things can change on the server without the 1277 client necessarily being aware of it. 1278 1279 This function should return a positive value if the dentry is 1280 still valid, and zero or a negative error code if it isn't. 1281 1282 d_revalidate may be called in rcu-walk mode (flags & 1283 LOOKUP_RCU). If in rcu-walk mode, the filesystem must 1284 revalidate the dentry without blocking or storing to the dentry, 1285 d_parent and d_inode should not be used without care (because 1286 they can change and, in d_inode case, even become NULL under 1287 us). 1288 1289 If a situation is encountered that rcu-walk cannot handle, 1290 return 1291 -ECHILD and it will be called again in ref-walk mode. 1292 1293``d_weak_revalidate`` 1294 called when the VFS needs to revalidate a "jumped" dentry. This 1295 is called when a path-walk ends at dentry that was not acquired 1296 by doing a lookup in the parent directory. This includes "/", 1297 "." and "..", as well as procfs-style symlinks and mountpoint 1298 traversal. 1299 1300 In this case, we are less concerned with whether the dentry is 1301 still fully correct, but rather that the inode is still valid. 1302 As with d_revalidate, most local filesystems will set this to 1303 NULL since their dcache entries are always valid. 1304 1305 This function has the same return code semantics as 1306 d_revalidate. 1307 1308 d_weak_revalidate is only called after leaving rcu-walk mode. 1309 1310``d_hash`` 1311 called when the VFS adds a dentry to the hash table. The first 1312 dentry passed to d_hash is the parent directory that the name is 1313 to be hashed into. 1314 1315 Same locking and synchronisation rules as d_compare regarding 1316 what is safe to dereference etc. 1317 1318``d_compare`` 1319 called to compare a dentry name with a given name. The first 1320 dentry is the parent of the dentry to be compared, the second is 1321 the child dentry. len and name string are properties of the 1322 dentry to be compared. qstr is the name to compare it with. 1323 1324 Must be constant and idempotent, and should not take locks if 1325 possible, and should not or store into the dentry. Should not 1326 dereference pointers outside the dentry without lots of care 1327 (eg. d_parent, d_inode, d_name should not be used). 1328 1329 However, our vfsmount is pinned, and RCU held, so the dentries 1330 and inodes won't disappear, neither will our sb or filesystem 1331 module. ->d_sb may be used. 1332 1333 It is a tricky calling convention because it needs to be called 1334 under "rcu-walk", ie. without any locks or references on things. 1335 1336``d_delete`` 1337 called when the last reference to a dentry is dropped and the 1338 dcache is deciding whether or not to cache it. Return 1 to 1339 delete immediately, or 0 to cache the dentry. Default is NULL 1340 which means to always cache a reachable dentry. d_delete must 1341 be constant and idempotent. 1342 1343``d_init`` 1344 called when a dentry is allocated 1345 1346``d_release`` 1347 called when a dentry is really deallocated 1348 1349``d_iput`` 1350 called when a dentry loses its inode (just prior to its being 1351 deallocated). The default when this is NULL is that the VFS 1352 calls iput(). If you define this method, you must call iput() 1353 yourself 1354 1355``d_dname`` 1356 called when the pathname of a dentry should be generated. 1357 Useful for some pseudo filesystems (sockfs, pipefs, ...) to 1358 delay pathname generation. (Instead of doing it when dentry is 1359 created, it's done only when the path is needed.). Real 1360 filesystems probably dont want to use it, because their dentries 1361 are present in global dcache hash, so their hash should be an 1362 invariant. As no lock is held, d_dname() should not try to 1363 modify the dentry itself, unless appropriate SMP safety is used. 1364 CAUTION : d_path() logic is quite tricky. The correct way to 1365 return for example "Hello" is to put it at the end of the 1366 buffer, and returns a pointer to the first char. 1367 dynamic_dname() helper function is provided to take care of 1368 this. 1369 1370 Example : 1371 1372.. code-block:: c 1373 1374 static char *pipefs_dname(struct dentry *dent, char *buffer, int buflen) 1375 { 1376 return dynamic_dname(dentry, buffer, buflen, "pipe:[%lu]", 1377 dentry->d_inode->i_ino); 1378 } 1379 1380``d_automount`` 1381 called when an automount dentry is to be traversed (optional). 1382 This should create a new VFS mount record and return the record 1383 to the caller. The caller is supplied with a path parameter 1384 giving the automount directory to describe the automount target 1385 and the parent VFS mount record to provide inheritable mount 1386 parameters. NULL should be returned if someone else managed to 1387 make the automount first. If the vfsmount creation failed, then 1388 an error code should be returned. If -EISDIR is returned, then 1389 the directory will be treated as an ordinary directory and 1390 returned to pathwalk to continue walking. 1391 1392 If a vfsmount is returned, the caller will attempt to mount it 1393 on the mountpoint and will remove the vfsmount from its 1394 expiration list in the case of failure. The vfsmount should be 1395 returned with 2 refs on it to prevent automatic expiration - the 1396 caller will clean up the additional ref. 1397 1398 This function is only used if DCACHE_NEED_AUTOMOUNT is set on 1399 the dentry. This is set by __d_instantiate() if S_AUTOMOUNT is 1400 set on the inode being added. 1401 1402``d_manage`` 1403 called to allow the filesystem to manage the transition from a 1404 dentry (optional). This allows autofs, for example, to hold up 1405 clients waiting to explore behind a 'mountpoint' while letting 1406 the daemon go past and construct the subtree there. 0 should be 1407 returned to let the calling process continue. -EISDIR can be 1408 returned to tell pathwalk to use this directory as an ordinary 1409 directory and to ignore anything mounted on it and not to check 1410 the automount flag. Any other error code will abort pathwalk 1411 completely. 1412 1413 If the 'rcu_walk' parameter is true, then the caller is doing a 1414 pathwalk in RCU-walk mode. Sleeping is not permitted in this 1415 mode, and the caller can be asked to leave it and call again by 1416 returning -ECHILD. -EISDIR may also be returned to tell 1417 pathwalk to ignore d_automount or any mounts. 1418 1419 This function is only used if DCACHE_MANAGE_TRANSIT is set on 1420 the dentry being transited from. 1421 1422``d_real`` 1423 overlay/union type filesystems implement this method to return 1424 one of the underlying dentries hidden by the overlay. It is 1425 used in two different modes: 1426 1427 Called from file_dentry() it returns the real dentry matching 1428 the inode argument. The real dentry may be from a lower layer 1429 already copied up, but still referenced from the file. This 1430 mode is selected with a non-NULL inode argument. 1431 1432 With NULL inode the topmost real underlying dentry is returned. 1433 1434Each dentry has a pointer to its parent dentry, as well as a hash list 1435of child dentries. Child dentries are basically like files in a 1436directory. 1437 1438 1439Directory Entry Cache API 1440-------------------------- 1441 1442There are a number of functions defined which permit a filesystem to 1443manipulate dentries: 1444 1445``dget`` 1446 open a new handle for an existing dentry (this just increments 1447 the usage count) 1448 1449``dput`` 1450 close a handle for a dentry (decrements the usage count). If 1451 the usage count drops to 0, and the dentry is still in its 1452 parent's hash, the "d_delete" method is called to check whether 1453 it should be cached. If it should not be cached, or if the 1454 dentry is not hashed, it is deleted. Otherwise cached dentries 1455 are put into an LRU list to be reclaimed on memory shortage. 1456 1457``d_drop`` 1458 this unhashes a dentry from its parents hash list. A subsequent 1459 call to dput() will deallocate the dentry if its usage count 1460 drops to 0 1461 1462``d_delete`` 1463 delete a dentry. If there are no other open references to the 1464 dentry then the dentry is turned into a negative dentry (the 1465 d_iput() method is called). If there are other references, then 1466 d_drop() is called instead 1467 1468``d_add`` 1469 add a dentry to its parents hash list and then calls 1470 d_instantiate() 1471 1472``d_instantiate`` 1473 add a dentry to the alias hash list for the inode and updates 1474 the "d_inode" member. The "i_count" member in the inode 1475 structure should be set/incremented. If the inode pointer is 1476 NULL, the dentry is called a "negative dentry". This function 1477 is commonly called when an inode is created for an existing 1478 negative dentry 1479 1480``d_lookup`` 1481 look up a dentry given its parent and path name component It 1482 looks up the child of that given name from the dcache hash 1483 table. If it is found, the reference count is incremented and 1484 the dentry is returned. The caller must use dput() to free the 1485 dentry when it finishes using it. 1486 1487 1488Mount Options 1489============= 1490 1491 1492Parsing options 1493--------------- 1494 1495On mount and remount the filesystem is passed a string containing a 1496comma separated list of mount options. The options can have either of 1497these forms: 1498 1499 option 1500 option=value 1501 1502The <linux/parser.h> header defines an API that helps parse these 1503options. There are plenty of examples on how to use it in existing 1504filesystems. 1505 1506 1507Showing options 1508--------------- 1509 1510If a filesystem accepts mount options, it must define show_options() to 1511show all the currently active options. The rules are: 1512 1513 - options MUST be shown which are not default or their values differ 1514 from the default 1515 1516 - options MAY be shown which are enabled by default or have their 1517 default value 1518 1519Options used only internally between a mount helper and the kernel (such 1520as file descriptors), or which only have an effect during the mounting 1521(such as ones controlling the creation of a journal) are exempt from the 1522above rules. 1523 1524The underlying reason for the above rules is to make sure, that a mount 1525can be accurately replicated (e.g. umounting and mounting again) based 1526on the information found in /proc/mounts. 1527 1528 1529Resources 1530========= 1531 1532(Note some of these resources are not up-to-date with the latest kernel 1533 version.) 1534 1535Creating Linux virtual filesystems. 2002 1536 <https://lwn.net/Articles/13325/> 1537 1538The Linux Virtual File-system Layer by Neil Brown. 1999 1539 <http://www.cse.unsw.edu.au/~neilb/oss/linux-commentary/vfs.html> 1540 1541A tour of the Linux VFS by Michael K. Johnson. 1996 1542 <https://www.tldp.org/LDP/khg/HyperNews/get/fs/vfstour.html> 1543 1544A small trail through the Linux kernel by Andries Brouwer. 2001 1545 <https://www.win.tue.nl/~aeb/linux/vfs/trail.html> 1546