1 .. SPDX-License-Identifier: GPL-2.0 1 .. SPDX-License-Identifier: GPL-2.0
2 2
3 ========================================= 3 =========================================
4 Overview of the Linux Virtual File System 4 Overview of the Linux Virtual File System
5 ========================================= 5 =========================================
6 6
7 Original author: Richard Gooch <rgooch@atnf.csi 7 Original author: Richard Gooch <rgooch@atnf.csiro.au>
8 8
9 - Copyright (C) 1999 Richard Gooch 9 - Copyright (C) 1999 Richard Gooch
10 - Copyright (C) 2005 Pekka Enberg 10 - Copyright (C) 2005 Pekka Enberg
11 11
12 12
13 Introduction 13 Introduction
14 ============ 14 ============
15 15
16 The Virtual File System (also known as the Vir 16 The Virtual File System (also known as the Virtual Filesystem Switch) is
17 the software layer in the kernel that provides 17 the software layer in the kernel that provides the filesystem interface
18 to userspace programs. It also provides an ab 18 to userspace programs. It also provides an abstraction within the
19 kernel which allows different filesystem imple 19 kernel which allows different filesystem implementations to coexist.
20 20
21 VFS system calls open(2), stat(2), read(2), wr 21 VFS system calls open(2), stat(2), read(2), write(2), chmod(2) and so on
22 are called from a process context. Filesystem 22 are called from a process context. Filesystem locking is described in
23 the document Documentation/filesystems/locking 23 the document Documentation/filesystems/locking.rst.
24 24
25 25
26 Directory Entry Cache (dcache) 26 Directory Entry Cache (dcache)
27 ------------------------------ 27 ------------------------------
28 28
29 The VFS implements the open(2), stat(2), chmod 29 The VFS implements the open(2), stat(2), chmod(2), and similar system
30 calls. The pathname argument that is passed t 30 calls. The pathname argument that is passed to them is used by the VFS
31 to search through the directory entry cache (a 31 to search through the directory entry cache (also known as the dentry
32 cache or dcache). This provides a very fast l 32 cache or dcache). This provides a very fast look-up mechanism to
33 translate a pathname (filename) into a specifi 33 translate a pathname (filename) into a specific dentry. Dentries live
34 in RAM and are never saved to disc: they exist 34 in RAM and are never saved to disc: they exist only for performance.
35 35
36 The dentry cache is meant to be a view into yo 36 The dentry cache is meant to be a view into your entire filespace. As
37 most computers cannot fit all dentries in the 37 most computers cannot fit all dentries in the RAM at the same time, some
38 bits of the cache are missing. In order to re 38 bits of the cache are missing. In order to resolve your pathname into a
39 dentry, the VFS may have to resort to creating 39 dentry, the VFS may have to resort to creating dentries along the way,
40 and then loading the inode. This is done by l 40 and then loading the inode. This is done by looking up the inode.
41 41
42 42
43 The Inode Object 43 The Inode Object
44 ---------------- 44 ----------------
45 45
46 An individual dentry usually has a pointer to 46 An individual dentry usually has a pointer to an inode. Inodes are
47 filesystem objects such as regular files, dire 47 filesystem objects such as regular files, directories, FIFOs and other
48 beasts. They live either on the disc (for blo 48 beasts. They live either on the disc (for block device filesystems) or
49 in the memory (for pseudo filesystems). Inode 49 in the memory (for pseudo filesystems). Inodes that live on the disc
50 are copied into the memory when required and c 50 are copied into the memory when required and changes to the inode are
51 written back to disc. A single inode can be p 51 written back to disc. A single inode can be pointed to by multiple
52 dentries (hard links, for example, do this). 52 dentries (hard links, for example, do this).
53 53
54 To look up an inode requires that the VFS call 54 To look up an inode requires that the VFS calls the lookup() method of
55 the parent directory inode. This method is in 55 the parent directory inode. This method is installed by the specific
56 filesystem implementation that the inode lives 56 filesystem implementation that the inode lives in. Once the VFS has the
57 required dentry (and hence the inode), we can 57 required dentry (and hence the inode), we can do all those boring things
58 like open(2) the file, or stat(2) it to peek a 58 like open(2) the file, or stat(2) it to peek at the inode data. The
59 stat(2) operation is fairly simple: once the V 59 stat(2) operation is fairly simple: once the VFS has the dentry, it
60 peeks at the inode data and passes some of it 60 peeks at the inode data and passes some of it back to userspace.
61 61
62 62
63 The File Object 63 The File Object
64 --------------- 64 ---------------
65 65
66 Opening a file requires another operation: all 66 Opening a file requires another operation: allocation of a file
67 structure (this is the kernel-side implementat 67 structure (this is the kernel-side implementation of file descriptors).
68 The freshly allocated file structure is initia 68 The freshly allocated file structure is initialized with a pointer to
69 the dentry and a set of file operation member 69 the dentry and a set of file operation member functions. These are
70 taken from the inode data. The open() file me 70 taken from the inode data. The open() file method is then called so the
71 specific filesystem implementation can do its 71 specific filesystem implementation can do its work. You can see that
72 this is another switch performed by the VFS. 72 this is another switch performed by the VFS. The file structure is
73 placed into the file descriptor table for the 73 placed into the file descriptor table for the process.
74 74
75 Reading, writing and closing files (and other 75 Reading, writing and closing files (and other assorted VFS operations)
76 is done by using the userspace file descriptor 76 is done by using the userspace file descriptor to grab the appropriate
77 file structure, and then calling the required 77 file structure, and then calling the required file structure method to
78 do whatever is required. For as long as the f 78 do whatever is required. For as long as the file is open, it keeps the
79 dentry in use, which in turn means that the VF 79 dentry in use, which in turn means that the VFS inode is still in use.
80 80
81 81
82 Registering and Mounting a Filesystem 82 Registering and Mounting a Filesystem
83 ===================================== 83 =====================================
84 84
85 To register and unregister a filesystem, use t 85 To register and unregister a filesystem, use the following API
86 functions: 86 functions:
87 87
88 .. code-block:: c 88 .. code-block:: c
89 89
90 #include <linux/fs.h> 90 #include <linux/fs.h>
91 91
92 extern int register_filesystem(struct 92 extern int register_filesystem(struct file_system_type *);
93 extern int unregister_filesystem(struc 93 extern int unregister_filesystem(struct file_system_type *);
94 94
95 The passed struct file_system_type describes y 95 The passed struct file_system_type describes your filesystem. When a
96 request is made to mount a filesystem onto a d 96 request is made to mount a filesystem onto a directory in your
97 namespace, the VFS will call the appropriate m 97 namespace, the VFS will call the appropriate mount() method for the
98 specific filesystem. New vfsmount referring t 98 specific filesystem. New vfsmount referring to the tree returned by
99 ->mount() will be attached to the mountpoint, 99 ->mount() will be attached to the mountpoint, so that when pathname
100 resolution reaches the mountpoint it will jump 100 resolution reaches the mountpoint it will jump into the root of that
101 vfsmount. 101 vfsmount.
102 102
103 You can see all filesystems that are registere 103 You can see all filesystems that are registered to the kernel in the
104 file /proc/filesystems. 104 file /proc/filesystems.
105 105
106 106
107 struct file_system_type 107 struct file_system_type
108 ----------------------- 108 -----------------------
109 109
110 This describes the filesystem. The following 110 This describes the filesystem. The following
111 members are defined: 111 members are defined:
112 112
113 .. code-block:: c 113 .. code-block:: c
114 114
115 struct file_system_type { 115 struct file_system_type {
116 const char *name; 116 const char *name;
117 int fs_flags; 117 int fs_flags;
118 int (*init_fs_context)(struct 118 int (*init_fs_context)(struct fs_context *);
119 const struct fs_parameter_spec 119 const struct fs_parameter_spec *parameters;
120 struct dentry *(*mount) (struc 120 struct dentry *(*mount) (struct file_system_type *, int,
121 const char *, void *); 121 const char *, void *);
122 void (*kill_sb) (struct super_ 122 void (*kill_sb) (struct super_block *);
123 struct module *owner; 123 struct module *owner;
124 struct file_system_type * next 124 struct file_system_type * next;
125 struct hlist_head fs_supers; 125 struct hlist_head fs_supers;
126 126
127 struct lock_class_key s_lock_k 127 struct lock_class_key s_lock_key;
128 struct lock_class_key s_umount 128 struct lock_class_key s_umount_key;
129 struct lock_class_key s_vfs_re 129 struct lock_class_key s_vfs_rename_key;
130 struct lock_class_key s_writer 130 struct lock_class_key s_writers_key[SB_FREEZE_LEVELS];
131 131
132 struct lock_class_key i_lock_k 132 struct lock_class_key i_lock_key;
133 struct lock_class_key i_mutex_ 133 struct lock_class_key i_mutex_key;
134 struct lock_class_key invalida 134 struct lock_class_key invalidate_lock_key;
135 struct lock_class_key i_mutex_ 135 struct lock_class_key i_mutex_dir_key;
136 }; 136 };
137 137
138 ``name`` 138 ``name``
139 the name of the filesystem type, such 139 the name of the filesystem type, such as "ext2", "iso9660",
140 "msdos" and so on 140 "msdos" and so on
141 141
142 ``fs_flags`` 142 ``fs_flags``
143 various flags (i.e. FS_REQUIRES_DEV, F 143 various flags (i.e. FS_REQUIRES_DEV, FS_NO_DCACHE, etc.)
144 144
145 ``init_fs_context`` 145 ``init_fs_context``
146 Initializes 'struct fs_context' ->ops 146 Initializes 'struct fs_context' ->ops and ->fs_private fields with
147 filesystem-specific data. 147 filesystem-specific data.
148 148
149 ``parameters`` 149 ``parameters``
150 Pointer to the array of filesystem par 150 Pointer to the array of filesystem parameters descriptors
151 'struct fs_parameter_spec'. 151 'struct fs_parameter_spec'.
152 More info in Documentation/filesystems 152 More info in Documentation/filesystems/mount_api.rst.
153 153
154 ``mount`` 154 ``mount``
155 the method to call when a new instance 155 the method to call when a new instance of this filesystem should
156 be mounted 156 be mounted
157 157
158 ``kill_sb`` 158 ``kill_sb``
159 the method to call when an instance of 159 the method to call when an instance of this filesystem should be
160 shut down 160 shut down
161 161
162 162
163 ``owner`` 163 ``owner``
164 for internal VFS use: you should initi 164 for internal VFS use: you should initialize this to THIS_MODULE
165 in most cases. 165 in most cases.
166 166
167 ``next`` 167 ``next``
168 for internal VFS use: you should initi 168 for internal VFS use: you should initialize this to NULL
169 169
170 ``fs_supers`` 170 ``fs_supers``
171 for internal VFS use: hlist of filesys 171 for internal VFS use: hlist of filesystem instances (superblocks)
172 172
173 s_lock_key, s_umount_key, s_vfs_rename_key, 173 s_lock_key, s_umount_key, s_vfs_rename_key, s_writers_key,
174 i_lock_key, i_mutex_key, invalidate_lock_key 174 i_lock_key, i_mutex_key, invalidate_lock_key, i_mutex_dir_key: lockdep-specific
175 175
176 The mount() method has the following arguments 176 The mount() method has the following arguments:
177 177
178 ``struct file_system_type *fs_type`` 178 ``struct file_system_type *fs_type``
179 describes the filesystem, partly initi 179 describes the filesystem, partly initialized by the specific
180 filesystem code 180 filesystem code
181 181
182 ``int flags`` 182 ``int flags``
183 mount flags 183 mount flags
184 184
185 ``const char *dev_name`` 185 ``const char *dev_name``
186 the device name we are mounting. 186 the device name we are mounting.
187 187
188 ``void *data`` 188 ``void *data``
189 arbitrary mount options, usually comes 189 arbitrary mount options, usually comes as an ASCII string (see
190 "Mount Options" section) 190 "Mount Options" section)
191 191
192 The mount() method must return the root dentry 192 The mount() method must return the root dentry of the tree requested by
193 caller. An active reference to its superblock 193 caller. An active reference to its superblock must be grabbed and the
194 superblock must be locked. On failure it shou 194 superblock must be locked. On failure it should return ERR_PTR(error).
195 195
196 The arguments match those of mount(2) and thei 196 The arguments match those of mount(2) and their interpretation depends
197 on filesystem type. E.g. for block filesystem 197 on filesystem type. E.g. for block filesystems, dev_name is interpreted
198 as block device name, that device is opened an 198 as block device name, that device is opened and if it contains a
199 suitable filesystem image the method creates a 199 suitable filesystem image the method creates and initializes struct
200 super_block accordingly, returning its root de 200 super_block accordingly, returning its root dentry to caller.
201 201
202 ->mount() may choose to return a subtree of ex 202 ->mount() may choose to return a subtree of existing filesystem - it
203 doesn't have to create a new one. The main re 203 doesn't have to create a new one. The main result from the caller's
204 point of view is a reference to dentry at the 204 point of view is a reference to dentry at the root of (sub)tree to be
205 attached; creation of new superblock is a comm 205 attached; creation of new superblock is a common side effect.
206 206
207 The most interesting member of the superblock 207 The most interesting member of the superblock structure that the mount()
208 method fills in is the "s_op" field. This is 208 method fills in is the "s_op" field. This is a pointer to a "struct
209 super_operations" which describes the next lev 209 super_operations" which describes the next level of the filesystem
210 implementation. 210 implementation.
211 211
212 Usually, a filesystem uses one of the generic 212 Usually, a filesystem uses one of the generic mount() implementations
213 and provides a fill_super() callback instead. 213 and provides a fill_super() callback instead. The generic variants are:
214 214
215 ``mount_bdev`` 215 ``mount_bdev``
216 mount a filesystem residing on a block 216 mount a filesystem residing on a block device
217 217
218 ``mount_nodev`` 218 ``mount_nodev``
219 mount a filesystem that is not backed 219 mount a filesystem that is not backed by a device
220 220
221 ``mount_single`` 221 ``mount_single``
222 mount a filesystem which shares the in 222 mount a filesystem which shares the instance between all mounts
223 223
224 A fill_super() callback implementation has the 224 A fill_super() callback implementation has the following arguments:
225 225
226 ``struct super_block *sb`` 226 ``struct super_block *sb``
227 the superblock structure. The callbac 227 the superblock structure. The callback must initialize this
228 properly. 228 properly.
229 229
230 ``void *data`` 230 ``void *data``
231 arbitrary mount options, usually comes 231 arbitrary mount options, usually comes as an ASCII string (see
232 "Mount Options" section) 232 "Mount Options" section)
233 233
234 ``int silent`` 234 ``int silent``
235 whether or not to be silent on error 235 whether or not to be silent on error
236 236
237 237
238 The Superblock Object 238 The Superblock Object
239 ===================== 239 =====================
240 240
241 A superblock object represents a mounted files 241 A superblock object represents a mounted filesystem.
242 242
243 243
244 struct super_operations 244 struct super_operations
245 ----------------------- 245 -----------------------
246 246
247 This describes how the VFS can manipulate the 247 This describes how the VFS can manipulate the superblock of your
248 filesystem. The following members are defined 248 filesystem. The following members are defined:
249 249
250 .. code-block:: c 250 .. code-block:: c
251 251
252 struct super_operations { 252 struct super_operations {
253 struct inode *(*alloc_inode)(s 253 struct inode *(*alloc_inode)(struct super_block *sb);
254 void (*destroy_inode)(struct i 254 void (*destroy_inode)(struct inode *);
255 void (*free_inode)(struct inod 255 void (*free_inode)(struct inode *);
256 256
257 void (*dirty_inode) (struct in 257 void (*dirty_inode) (struct inode *, int flags);
258 int (*write_inode) (struct ino 258 int (*write_inode) (struct inode *, struct writeback_control *wbc);
259 int (*drop_inode) (struct inod 259 int (*drop_inode) (struct inode *);
260 void (*evict_inode) (struct in 260 void (*evict_inode) (struct inode *);
261 void (*put_super) (struct supe 261 void (*put_super) (struct super_block *);
262 int (*sync_fs)(struct super_bl 262 int (*sync_fs)(struct super_block *sb, int wait);
263 int (*freeze_super) (struct su !! 263 int (*freeze_super) (struct super_block *);
264 enum f <<
265 int (*freeze_fs) (struct super 264 int (*freeze_fs) (struct super_block *);
266 int (*thaw_super) (struct supe !! 265 int (*thaw_super) (struct super_block *);
267 enum f <<
268 int (*unfreeze_fs) (struct sup 266 int (*unfreeze_fs) (struct super_block *);
269 int (*statfs) (struct dentry * 267 int (*statfs) (struct dentry *, struct kstatfs *);
270 int (*remount_fs) (struct supe 268 int (*remount_fs) (struct super_block *, int *, char *);
271 void (*umount_begin) (struct s 269 void (*umount_begin) (struct super_block *);
272 270
273 int (*show_options)(struct seq 271 int (*show_options)(struct seq_file *, struct dentry *);
274 int (*show_devname)(struct seq 272 int (*show_devname)(struct seq_file *, struct dentry *);
275 int (*show_path)(struct seq_fi 273 int (*show_path)(struct seq_file *, struct dentry *);
276 int (*show_stats)(struct seq_f 274 int (*show_stats)(struct seq_file *, struct dentry *);
277 275
278 ssize_t (*quota_read)(struct s 276 ssize_t (*quota_read)(struct super_block *, int, char *, size_t, loff_t);
279 ssize_t (*quota_write)(struct 277 ssize_t (*quota_write)(struct super_block *, int, const char *, size_t, loff_t);
280 struct dquot **(*get_dquots)(s 278 struct dquot **(*get_dquots)(struct inode *);
281 279
282 long (*nr_cached_objects)(stru 280 long (*nr_cached_objects)(struct super_block *,
283 struct 281 struct shrink_control *);
284 long (*free_cached_objects)(st 282 long (*free_cached_objects)(struct super_block *,
285 struct 283 struct shrink_control *);
286 }; 284 };
287 285
288 All methods are called without any locks being 286 All methods are called without any locks being held, unless otherwise
289 noted. This means that most methods can block 287 noted. This means that most methods can block safely. All methods are
290 only called from a process context (i.e. not f 288 only called from a process context (i.e. not from an interrupt handler
291 or bottom half). 289 or bottom half).
292 290
293 ``alloc_inode`` 291 ``alloc_inode``
294 this method is called by alloc_inode() 292 this method is called by alloc_inode() to allocate memory for
295 struct inode and initialize it. If th 293 struct inode and initialize it. If this function is not
296 defined, a simple 'struct inode' is al 294 defined, a simple 'struct inode' is allocated. Normally
297 alloc_inode will be used to allocate a 295 alloc_inode will be used to allocate a larger structure which
298 contains a 'struct inode' embedded wit 296 contains a 'struct inode' embedded within it.
299 297
300 ``destroy_inode`` 298 ``destroy_inode``
301 this method is called by destroy_inode 299 this method is called by destroy_inode() to release resources
302 allocated for struct inode. It is onl 300 allocated for struct inode. It is only required if
303 ->alloc_inode was defined and simply u 301 ->alloc_inode was defined and simply undoes anything done by
304 ->alloc_inode. 302 ->alloc_inode.
305 303
306 ``free_inode`` 304 ``free_inode``
307 this method is called from RCU callbac 305 this method is called from RCU callback. If you use call_rcu()
308 in ->destroy_inode to free 'struct ino 306 in ->destroy_inode to free 'struct inode' memory, then it's
309 better to release memory in this metho 307 better to release memory in this method.
310 308
311 ``dirty_inode`` 309 ``dirty_inode``
312 this method is called by the VFS when 310 this method is called by the VFS when an inode is marked dirty.
313 This is specifically for the inode its 311 This is specifically for the inode itself being marked dirty,
314 not its data. If the update needs to 312 not its data. If the update needs to be persisted by fdatasync(),
315 then I_DIRTY_DATASYNC will be set in t 313 then I_DIRTY_DATASYNC will be set in the flags argument.
316 I_DIRTY_TIME will be set in the flags 314 I_DIRTY_TIME will be set in the flags in case lazytime is enabled
317 and struct inode has times updated sin 315 and struct inode has times updated since the last ->dirty_inode
318 call. 316 call.
319 317
320 ``write_inode`` 318 ``write_inode``
321 this method is called when the VFS nee 319 this method is called when the VFS needs to write an inode to
322 disc. The second parameter indicates 320 disc. The second parameter indicates whether the write should
323 be synchronous or not, not all filesys 321 be synchronous or not, not all filesystems check this flag.
324 322
325 ``drop_inode`` 323 ``drop_inode``
326 called when the last access to the ino 324 called when the last access to the inode is dropped, with the
327 inode->i_lock spinlock held. 325 inode->i_lock spinlock held.
328 326
329 This method should be either NULL (nor 327 This method should be either NULL (normal UNIX filesystem
330 semantics) or "generic_delete_inode" ( 328 semantics) or "generic_delete_inode" (for filesystems that do
331 not want to cache inodes - causing "de 329 not want to cache inodes - causing "delete_inode" to always be
332 called regardless of the value of i_nl 330 called regardless of the value of i_nlink)
333 331
334 The "generic_delete_inode()" behavior 332 The "generic_delete_inode()" behavior is equivalent to the old
335 practice of using "force_delete" in th 333 practice of using "force_delete" in the put_inode() case, but
336 does not have the races that the "forc 334 does not have the races that the "force_delete()" approach had.
337 335
338 ``evict_inode`` 336 ``evict_inode``
339 called when the VFS wants to evict an 337 called when the VFS wants to evict an inode. Caller does
340 *not* evict the pagecache or inode-ass 338 *not* evict the pagecache or inode-associated metadata buffers;
341 the method has to use truncate_inode_p 339 the method has to use truncate_inode_pages_final() to get rid
342 of those. Caller makes sure async writ 340 of those. Caller makes sure async writeback cannot be running for
343 the inode while (or after) ->evict_ino 341 the inode while (or after) ->evict_inode() is called. Optional.
344 342
345 ``put_super`` 343 ``put_super``
346 called when the VFS wishes to free the 344 called when the VFS wishes to free the superblock
347 (i.e. unmount). This is called with t 345 (i.e. unmount). This is called with the superblock lock held
348 346
349 ``sync_fs`` 347 ``sync_fs``
350 called when VFS is writing out all dir 348 called when VFS is writing out all dirty data associated with a
351 superblock. The second parameter indi 349 superblock. The second parameter indicates whether the method
352 should wait until the write out has be 350 should wait until the write out has been completed. Optional.
353 351
354 ``freeze_super`` 352 ``freeze_super``
355 Called instead of ->freeze_fs callback 353 Called instead of ->freeze_fs callback if provided.
356 Main difference is that ->freeze_super 354 Main difference is that ->freeze_super is called without taking
357 down_write(&sb->s_umount). If filesyst 355 down_write(&sb->s_umount). If filesystem implements it and wants
358 ->freeze_fs to be called too, then it 356 ->freeze_fs to be called too, then it has to call ->freeze_fs
359 explicitly from this callback. Optiona 357 explicitly from this callback. Optional.
360 358
361 ``freeze_fs`` 359 ``freeze_fs``
362 called when VFS is locking a filesyste 360 called when VFS is locking a filesystem and forcing it into a
363 consistent state. This method is curr 361 consistent state. This method is currently used by the Logical
364 Volume Manager (LVM) and ioctl(FIFREEZ 362 Volume Manager (LVM) and ioctl(FIFREEZE). Optional.
365 363
366 ``thaw_super`` 364 ``thaw_super``
367 called when VFS is unlocking a filesys 365 called when VFS is unlocking a filesystem and making it writable
368 again after ->freeze_super. Optional. 366 again after ->freeze_super. Optional.
369 367
370 ``unfreeze_fs`` 368 ``unfreeze_fs``
371 called when VFS is unlocking a filesys 369 called when VFS is unlocking a filesystem and making it writable
372 again after ->freeze_fs. Optional. 370 again after ->freeze_fs. Optional.
373 371
374 ``statfs`` 372 ``statfs``
375 called when the VFS needs to get files 373 called when the VFS needs to get filesystem statistics.
376 374
377 ``remount_fs`` 375 ``remount_fs``
378 called when the filesystem is remounte 376 called when the filesystem is remounted. This is called with
379 the kernel lock held 377 the kernel lock held
380 378
381 ``umount_begin`` 379 ``umount_begin``
382 called when the VFS is unmounting a fi 380 called when the VFS is unmounting a filesystem.
383 381
384 ``show_options`` 382 ``show_options``
385 called by the VFS to show mount option 383 called by the VFS to show mount options for /proc/<pid>/mounts
386 and /proc/<pid>/mountinfo. 384 and /proc/<pid>/mountinfo.
387 (see "Mount Options" section) 385 (see "Mount Options" section)
388 386
389 ``show_devname`` 387 ``show_devname``
390 Optional. Called by the VFS to show de 388 Optional. Called by the VFS to show device name for
391 /proc/<pid>/{mounts,mountinfo,mountsta 389 /proc/<pid>/{mounts,mountinfo,mountstats}. If not provided then
392 '(struct mount).mnt_devname' will be u 390 '(struct mount).mnt_devname' will be used.
393 391
394 ``show_path`` 392 ``show_path``
395 Optional. Called by the VFS (for /proc 393 Optional. Called by the VFS (for /proc/<pid>/mountinfo) to show
396 the mount root dentry path relative to 394 the mount root dentry path relative to the filesystem root.
397 395
398 ``show_stats`` 396 ``show_stats``
399 Optional. Called by the VFS (for /proc 397 Optional. Called by the VFS (for /proc/<pid>/mountstats) to show
400 filesystem-specific mount statistics. 398 filesystem-specific mount statistics.
401 399
402 ``quota_read`` 400 ``quota_read``
403 called by the VFS to read from filesys 401 called by the VFS to read from filesystem quota file.
404 402
405 ``quota_write`` 403 ``quota_write``
406 called by the VFS to write to filesyst 404 called by the VFS to write to filesystem quota file.
407 405
408 ``get_dquots`` 406 ``get_dquots``
409 called by quota to get 'struct dquot' 407 called by quota to get 'struct dquot' array for a particular inode.
410 Optional. 408 Optional.
411 409
412 ``nr_cached_objects`` 410 ``nr_cached_objects``
413 called by the sb cache shrinking funct 411 called by the sb cache shrinking function for the filesystem to
414 return the number of freeable cached o 412 return the number of freeable cached objects it contains.
415 Optional. 413 Optional.
416 414
417 ``free_cache_objects`` 415 ``free_cache_objects``
418 called by the sb cache shrinking funct 416 called by the sb cache shrinking function for the filesystem to
419 scan the number of objects indicated t 417 scan the number of objects indicated to try to free them.
420 Optional, but any filesystem implement 418 Optional, but any filesystem implementing this method needs to
421 also implement ->nr_cached_objects for 419 also implement ->nr_cached_objects for it to be called
422 correctly. 420 correctly.
423 421
424 We can't do anything with any errors t 422 We can't do anything with any errors that the filesystem might
425 encountered, hence the void return typ 423 encountered, hence the void return type. This will never be
426 called if the VM is trying to reclaim 424 called if the VM is trying to reclaim under GFP_NOFS conditions,
427 hence this method does not need to han 425 hence this method does not need to handle that situation itself.
428 426
429 Implementations must include condition 427 Implementations must include conditional reschedule calls inside
430 any scanning loop that is done. This 428 any scanning loop that is done. This allows the VFS to
431 determine appropriate scan batch sizes 429 determine appropriate scan batch sizes without having to worry
432 about whether implementations will cau 430 about whether implementations will cause holdoff problems due to
433 large scan batch sizes. 431 large scan batch sizes.
434 432
435 Whoever sets up the inode is responsible for f 433 Whoever sets up the inode is responsible for filling in the "i_op"
436 field. This is a pointer to a "struct inode_o 434 field. This is a pointer to a "struct inode_operations" which describes
437 the methods that can be performed on individua 435 the methods that can be performed on individual inodes.
438 436
439 437
440 struct xattr_handler !! 438 struct xattr_handlers
441 --------------------- 439 ---------------------
442 440
443 On filesystems that support extended attribute 441 On filesystems that support extended attributes (xattrs), the s_xattr
444 superblock field points to a NULL-terminated a 442 superblock field points to a NULL-terminated array of xattr handlers.
445 Extended attributes are name:value pairs. 443 Extended attributes are name:value pairs.
446 444
447 ``name`` 445 ``name``
448 Indicates that the handler matches att 446 Indicates that the handler matches attributes with the specified
449 name (such as "system.posix_acl_access 447 name (such as "system.posix_acl_access"); the prefix field must
450 be NULL. 448 be NULL.
451 449
452 ``prefix`` 450 ``prefix``
453 Indicates that the handler matches all 451 Indicates that the handler matches all attributes with the
454 specified name prefix (such as "user." 452 specified name prefix (such as "user."); the name field must be
455 NULL. 453 NULL.
456 454
457 ``list`` 455 ``list``
458 Determine if attributes matching this 456 Determine if attributes matching this xattr handler should be
459 listed for a particular dentry. Used 457 listed for a particular dentry. Used by some listxattr
460 implementations like generic_listxattr 458 implementations like generic_listxattr.
461 459
462 ``get`` 460 ``get``
463 Called by the VFS to get the value of 461 Called by the VFS to get the value of a particular extended
464 attribute. This method is called by t 462 attribute. This method is called by the getxattr(2) system
465 call. 463 call.
466 464
467 ``set`` 465 ``set``
468 Called by the VFS to set the value of 466 Called by the VFS to set the value of a particular extended
469 attribute. When the new value is NULL 467 attribute. When the new value is NULL, called to remove a
470 particular extended attribute. This m 468 particular extended attribute. This method is called by the
471 setxattr(2) and removexattr(2) system 469 setxattr(2) and removexattr(2) system calls.
472 470
473 When none of the xattr handlers of a filesyste 471 When none of the xattr handlers of a filesystem match the specified
474 attribute name or when a filesystem doesn't su 472 attribute name or when a filesystem doesn't support extended attributes,
475 the various ``*xattr(2)`` system calls return 473 the various ``*xattr(2)`` system calls return -EOPNOTSUPP.
476 474
477 475
478 The Inode Object 476 The Inode Object
479 ================ 477 ================
480 478
481 An inode object represents an object within th 479 An inode object represents an object within the filesystem.
482 480
483 481
484 struct inode_operations 482 struct inode_operations
485 ----------------------- 483 -----------------------
486 484
487 This describes how the VFS can manipulate an i 485 This describes how the VFS can manipulate an inode in your filesystem.
488 As of kernel 2.6.22, the following members are 486 As of kernel 2.6.22, the following members are defined:
489 487
490 .. code-block:: c 488 .. code-block:: c
491 489
492 struct inode_operations { 490 struct inode_operations {
493 int (*create) (struct mnt_idma 491 int (*create) (struct mnt_idmap *, struct inode *,struct dentry *, umode_t, bool);
494 struct dentry * (*lookup) (str 492 struct dentry * (*lookup) (struct inode *,struct dentry *, unsigned int);
495 int (*link) (struct dentry *,s 493 int (*link) (struct dentry *,struct inode *,struct dentry *);
496 int (*unlink) (struct inode *, 494 int (*unlink) (struct inode *,struct dentry *);
497 int (*symlink) (struct mnt_idm 495 int (*symlink) (struct mnt_idmap *, struct inode *,struct dentry *,const char *);
498 int (*mkdir) (struct mnt_idmap 496 int (*mkdir) (struct mnt_idmap *, struct inode *,struct dentry *,umode_t);
499 int (*rmdir) (struct inode *,s 497 int (*rmdir) (struct inode *,struct dentry *);
500 int (*mknod) (struct mnt_idmap 498 int (*mknod) (struct mnt_idmap *, struct inode *,struct dentry *,umode_t,dev_t);
501 int (*rename) (struct mnt_idma 499 int (*rename) (struct mnt_idmap *, struct inode *, struct dentry *,
502 struct inode *, 500 struct inode *, struct dentry *, unsigned int);
503 int (*readlink) (struct dentry 501 int (*readlink) (struct dentry *, char __user *,int);
504 const char *(*get_link) (struc 502 const char *(*get_link) (struct dentry *, struct inode *,
505 struc 503 struct delayed_call *);
506 int (*permission) (struct mnt_ 504 int (*permission) (struct mnt_idmap *, struct inode *, int);
507 struct posix_acl * (*get_inode 505 struct posix_acl * (*get_inode_acl)(struct inode *, int, bool);
508 int (*setattr) (struct mnt_idm 506 int (*setattr) (struct mnt_idmap *, struct dentry *, struct iattr *);
509 int (*getattr) (struct mnt_idm 507 int (*getattr) (struct mnt_idmap *, const struct path *, struct kstat *, u32, unsigned int);
510 ssize_t (*listxattr) (struct d 508 ssize_t (*listxattr) (struct dentry *, char *, size_t);
511 void (*update_time)(struct ino 509 void (*update_time)(struct inode *, struct timespec *, int);
512 int (*atomic_open)(struct inod 510 int (*atomic_open)(struct inode *, struct dentry *, struct file *,
513 unsigned op 511 unsigned open_flag, umode_t create_mode);
514 int (*tmpfile) (struct mnt_idm 512 int (*tmpfile) (struct mnt_idmap *, struct inode *, struct file *, umode_t);
515 struct posix_acl * (*get_acl)( 513 struct posix_acl * (*get_acl)(struct mnt_idmap *, struct dentry *, int);
516 int (*set_acl)(struct mnt_idma 514 int (*set_acl)(struct mnt_idmap *, struct dentry *, struct posix_acl *, int);
517 int (*fileattr_set)(struct mnt 515 int (*fileattr_set)(struct mnt_idmap *idmap,
518 struct den 516 struct dentry *dentry, struct fileattr *fa);
519 int (*fileattr_get)(struct den 517 int (*fileattr_get)(struct dentry *dentry, struct fileattr *fa);
520 struct offset_ctx *(*get_offse <<
521 }; 518 };
522 519
523 Again, all methods are called without any lock 520 Again, all methods are called without any locks being held, unless
524 otherwise noted. 521 otherwise noted.
525 522
526 ``create`` 523 ``create``
527 called by the open(2) and creat(2) sys 524 called by the open(2) and creat(2) system calls. Only required
528 if you want to support regular files. 525 if you want to support regular files. The dentry you get should
529 not have an inode (i.e. it should be a 526 not have an inode (i.e. it should be a negative dentry). Here
530 you will probably call d_instantiate() 527 you will probably call d_instantiate() with the dentry and the
531 newly created inode 528 newly created inode
532 529
533 ``lookup`` 530 ``lookup``
534 called when the VFS needs to look up a 531 called when the VFS needs to look up an inode in a parent
535 directory. The name to look for is fo 532 directory. The name to look for is found in the dentry. This
536 method must call d_add() to insert the 533 method must call d_add() to insert the found inode into the
537 dentry. The "i_count" field in the in 534 dentry. The "i_count" field in the inode structure should be
538 incremented. If the named inode does 535 incremented. If the named inode does not exist a NULL inode
539 should be inserted into the dentry (th 536 should be inserted into the dentry (this is called a negative
540 dentry). Returning an error code from 537 dentry). Returning an error code from this routine must only be
541 done on a real error, otherwise creati 538 done on a real error, otherwise creating inodes with system
542 calls like create(2), mknod(2), mkdir( 539 calls like create(2), mknod(2), mkdir(2) and so on will fail.
543 If you wish to overload the dentry met 540 If you wish to overload the dentry methods then you should
544 initialise the "d_dop" field in the de 541 initialise the "d_dop" field in the dentry; this is a pointer to
545 a struct "dentry_operations". This me 542 a struct "dentry_operations". This method is called with the
546 directory inode semaphore held 543 directory inode semaphore held
547 544
548 ``link`` 545 ``link``
549 called by the link(2) system call. On 546 called by the link(2) system call. Only required if you want to
550 support hard links. You will probably 547 support hard links. You will probably need to call
551 d_instantiate() just as you would in t 548 d_instantiate() just as you would in the create() method
552 549
553 ``unlink`` 550 ``unlink``
554 called by the unlink(2) system call. 551 called by the unlink(2) system call. Only required if you want
555 to support deleting inodes 552 to support deleting inodes
556 553
557 ``symlink`` 554 ``symlink``
558 called by the symlink(2) system call. 555 called by the symlink(2) system call. Only required if you want
559 to support symlinks. You will probabl 556 to support symlinks. You will probably need to call
560 d_instantiate() just as you would in t 557 d_instantiate() just as you would in the create() method
561 558
562 ``mkdir`` 559 ``mkdir``
563 called by the mkdir(2) system call. O 560 called by the mkdir(2) system call. Only required if you want
564 to support creating subdirectories. Y 561 to support creating subdirectories. You will probably need to
565 call d_instantiate() just as you would 562 call d_instantiate() just as you would in the create() method
566 563
567 ``rmdir`` 564 ``rmdir``
568 called by the rmdir(2) system call. O 565 called by the rmdir(2) system call. Only required if you want
569 to support deleting subdirectories 566 to support deleting subdirectories
570 567
571 ``mknod`` 568 ``mknod``
572 called by the mknod(2) system call to 569 called by the mknod(2) system call to create a device (char,
573 block) inode or a named pipe (FIFO) or 570 block) inode or a named pipe (FIFO) or socket. Only required if
574 you want to support creating these typ 571 you want to support creating these types of inodes. You will
575 probably need to call d_instantiate() 572 probably need to call d_instantiate() just as you would in the
576 create() method 573 create() method
577 574
578 ``rename`` 575 ``rename``
579 called by the rename(2) system call to 576 called by the rename(2) system call to rename the object to have
580 the parent and name given by the secon 577 the parent and name given by the second inode and dentry.
581 578
582 The filesystem must return -EINVAL for 579 The filesystem must return -EINVAL for any unsupported or
583 unknown flags. Currently the followin 580 unknown flags. Currently the following flags are implemented:
584 (1) RENAME_NOREPLACE: this flag indica 581 (1) RENAME_NOREPLACE: this flag indicates that if the target of
585 the rename exists the rename should fa 582 the rename exists the rename should fail with -EEXIST instead of
586 replacing the target. The VFS already 583 replacing the target. The VFS already checks for existence, so
587 for local filesystems the RENAME_NOREP 584 for local filesystems the RENAME_NOREPLACE implementation is
588 equivalent to plain rename. 585 equivalent to plain rename.
589 (2) RENAME_EXCHANGE: exchange source a 586 (2) RENAME_EXCHANGE: exchange source and target. Both must
590 exist; this is checked by the VFS. Un 587 exist; this is checked by the VFS. Unlike plain rename, source
591 and target may be of different type. 588 and target may be of different type.
592 589
593 ``get_link`` 590 ``get_link``
594 called by the VFS to follow a symbolic 591 called by the VFS to follow a symbolic link to the inode it
595 points to. Only required if you want 592 points to. Only required if you want to support symbolic links.
596 This method returns the symlink body t 593 This method returns the symlink body to traverse (and possibly
597 resets the current position with nd_ju 594 resets the current position with nd_jump_link()). If the body
598 won't go away until the inode is gone, 595 won't go away until the inode is gone, nothing else is needed;
599 if it needs to be otherwise pinned, ar 596 if it needs to be otherwise pinned, arrange for its release by
600 having get_link(..., ..., done) do set 597 having get_link(..., ..., done) do set_delayed_call(done,
601 destructor, argument). In that case d 598 destructor, argument). In that case destructor(argument) will
602 be called once VFS is done with the bo 599 be called once VFS is done with the body you've returned. May
603 be called in RCU mode; that is indicat 600 be called in RCU mode; that is indicated by NULL dentry
604 argument. If request can't be handled 601 argument. If request can't be handled without leaving RCU mode,
605 have it return ERR_PTR(-ECHILD). 602 have it return ERR_PTR(-ECHILD).
606 603
607 If the filesystem stores the symlink t 604 If the filesystem stores the symlink target in ->i_link, the
608 VFS may use it directly without callin 605 VFS may use it directly without calling ->get_link(); however,
609 ->get_link() must still be provided. 606 ->get_link() must still be provided. ->i_link must not be
610 freed until after an RCU grace period. 607 freed until after an RCU grace period. Writing to ->i_link
611 post-iget() time requires a 'release' 608 post-iget() time requires a 'release' memory barrier.
612 609
613 ``readlink`` 610 ``readlink``
614 this is now just an override for use b 611 this is now just an override for use by readlink(2) for the
615 cases when ->get_link uses nd_jump_lin 612 cases when ->get_link uses nd_jump_link() or object is not in
616 fact a symlink. Normally filesystems 613 fact a symlink. Normally filesystems should only implement
617 ->get_link for symlinks and readlink(2 614 ->get_link for symlinks and readlink(2) will automatically use
618 that. 615 that.
619 616
620 ``permission`` 617 ``permission``
621 called by the VFS to check for access 618 called by the VFS to check for access rights on a POSIX-like
622 filesystem. 619 filesystem.
623 620
624 May be called in rcu-walk mode (mask & 621 May be called in rcu-walk mode (mask & MAY_NOT_BLOCK). If in
625 rcu-walk mode, the filesystem must che 622 rcu-walk mode, the filesystem must check the permission without
626 blocking or storing to the inode. 623 blocking or storing to the inode.
627 624
628 If a situation is encountered that rcu 625 If a situation is encountered that rcu-walk cannot handle,
629 return 626 return
630 -ECHILD and it will be called again in 627 -ECHILD and it will be called again in ref-walk mode.
631 628
632 ``setattr`` 629 ``setattr``
633 called by the VFS to set attributes fo 630 called by the VFS to set attributes for a file. This method is
634 called by chmod(2) and related system 631 called by chmod(2) and related system calls.
635 632
636 ``getattr`` 633 ``getattr``
637 called by the VFS to get attributes of 634 called by the VFS to get attributes of a file. This method is
638 called by stat(2) and related system c 635 called by stat(2) and related system calls.
639 636
640 ``listxattr`` 637 ``listxattr``
641 called by the VFS to list all extended 638 called by the VFS to list all extended attributes for a given
642 file. This method is called by the li 639 file. This method is called by the listxattr(2) system call.
643 640
644 ``update_time`` 641 ``update_time``
645 called by the VFS to update a specific 642 called by the VFS to update a specific time or the i_version of
646 an inode. If this is not defined the 643 an inode. If this is not defined the VFS will update the inode
647 itself and call mark_inode_dirty_sync. 644 itself and call mark_inode_dirty_sync.
648 645
649 ``atomic_open`` 646 ``atomic_open``
650 called on the last component of an ope 647 called on the last component of an open. Using this optional
651 method the filesystem can look up, pos 648 method the filesystem can look up, possibly create and open the
652 file in one atomic operation. If it w 649 file in one atomic operation. If it wants to leave actual
653 opening to the caller (e.g. if the fil 650 opening to the caller (e.g. if the file turned out to be a
654 symlink, device, or just something fil 651 symlink, device, or just something filesystem won't do atomic
655 open for), it may signal this by retur 652 open for), it may signal this by returning finish_no_open(file,
656 dentry). This method is only called i 653 dentry). This method is only called if the last component is
657 negative or needs lookup. Cached posi 654 negative or needs lookup. Cached positive dentries are still
658 handled by f_op->open(). If the file 655 handled by f_op->open(). If the file was created, FMODE_CREATED
659 flag should be set in file->f_mode. I 656 flag should be set in file->f_mode. In case of O_EXCL the
660 method must only succeed if the file d 657 method must only succeed if the file didn't exist and hence
661 FMODE_CREATED shall always be set on s 658 FMODE_CREATED shall always be set on success.
662 659
663 ``tmpfile`` 660 ``tmpfile``
664 called in the end of O_TMPFILE open(). 661 called in the end of O_TMPFILE open(). Optional, equivalent to
665 atomically creating, opening and unlin 662 atomically creating, opening and unlinking a file in given
666 directory. On success needs to return 663 directory. On success needs to return with the file already
667 open; this can be done by calling fini 664 open; this can be done by calling finish_open_simple() right at
668 the end. 665 the end.
669 666
670 ``fileattr_get`` 667 ``fileattr_get``
671 called on ioctl(FS_IOC_GETFLAGS) and i 668 called on ioctl(FS_IOC_GETFLAGS) and ioctl(FS_IOC_FSGETXATTR) to
672 retrieve miscellaneous file flags and 669 retrieve miscellaneous file flags and attributes. Also called
673 before the relevant SET operation to c 670 before the relevant SET operation to check what is being changed
674 (in this case with i_rwsem locked excl 671 (in this case with i_rwsem locked exclusive). If unset, then
675 fall back to f_op->ioctl(). 672 fall back to f_op->ioctl().
676 673
677 ``fileattr_set`` 674 ``fileattr_set``
678 called on ioctl(FS_IOC_SETFLAGS) and i 675 called on ioctl(FS_IOC_SETFLAGS) and ioctl(FS_IOC_FSSETXATTR) to
679 change miscellaneous file flags and at 676 change miscellaneous file flags and attributes. Callers hold
680 i_rwsem exclusive. If unset, then fal 677 i_rwsem exclusive. If unset, then fall back to f_op->ioctl().
681 ``get_offset_ctx`` !! 678
682 called to get the offset context for a <<
683 filesystem must define this operation <<
684 simple_offset_dir_operations. <<
685 679
686 The Address Space Object 680 The Address Space Object
687 ======================== 681 ========================
688 682
689 The address space object is used to group and 683 The address space object is used to group and manage pages in the page
690 cache. It can be used to keep track of the pa 684 cache. It can be used to keep track of the pages in a file (or anything
691 else) and also track the mapping of sections o 685 else) and also track the mapping of sections of the file into process
692 address spaces. 686 address spaces.
693 687
694 There are a number of distinct yet related ser 688 There are a number of distinct yet related services that an
695 address-space can provide. These include comm 689 address-space can provide. These include communicating memory pressure,
696 page lookup by address, and keeping track of p 690 page lookup by address, and keeping track of pages tagged as Dirty or
697 Writeback. 691 Writeback.
698 692
699 The first can be used independently to the oth 693 The first can be used independently to the others. The VM can try to
700 either write dirty pages in order to clean the 694 either write dirty pages in order to clean them, or release clean pages
701 in order to reuse them. To do this it can cal 695 in order to reuse them. To do this it can call the ->writepage method
702 on dirty pages, and ->release_folio on clean f 696 on dirty pages, and ->release_folio on clean folios with the private
703 flag set. Clean pages without PagePrivate and 697 flag set. Clean pages without PagePrivate and with no external references
704 will be released without notice being given to 698 will be released without notice being given to the address_space.
705 699
706 To achieve this functionality, pages need to b 700 To achieve this functionality, pages need to be placed on an LRU with
707 lru_cache_add and mark_page_active needs to be 701 lru_cache_add and mark_page_active needs to be called whenever the page
708 is used. 702 is used.
709 703
710 Pages are normally kept in a radix tree index 704 Pages are normally kept in a radix tree index by ->index. This tree
711 maintains information about the PG_Dirty and P 705 maintains information about the PG_Dirty and PG_Writeback status of each
712 page, so that pages with either of these flags 706 page, so that pages with either of these flags can be found quickly.
713 707
714 The Dirty tag is primarily used by mpage_write 708 The Dirty tag is primarily used by mpage_writepages - the default
715 ->writepages method. It uses the tag to find 709 ->writepages method. It uses the tag to find dirty pages to call
716 ->writepage on. If mpage_writepages is not us 710 ->writepage on. If mpage_writepages is not used (i.e. the address
717 provides its own ->writepages) , the PAGECACHE 711 provides its own ->writepages) , the PAGECACHE_TAG_DIRTY tag is almost
718 unused. write_inode_now and sync_inode do use 712 unused. write_inode_now and sync_inode do use it (through
719 __sync_single_inode) to check if ->writepages 713 __sync_single_inode) to check if ->writepages has been successful in
720 writing out the whole address_space. 714 writing out the whole address_space.
721 715
722 The Writeback tag is used by filemap*wait* and 716 The Writeback tag is used by filemap*wait* and sync_page* functions, via
723 filemap_fdatawait_range, to wait for all write 717 filemap_fdatawait_range, to wait for all writeback to complete.
724 718
725 An address_space handler may attach extra info 719 An address_space handler may attach extra information to a page,
726 typically using the 'private' field in the 'st 720 typically using the 'private' field in the 'struct page'. If such
727 information is attached, the PG_Private flag s 721 information is attached, the PG_Private flag should be set. This will
728 cause various VM routines to make extra calls 722 cause various VM routines to make extra calls into the address_space
729 handler to deal with that data. 723 handler to deal with that data.
730 724
731 An address space acts as an intermediate betwe 725 An address space acts as an intermediate between storage and
732 application. Data is read into the address sp 726 application. Data is read into the address space a whole page at a
733 time, and provided to the application either b 727 time, and provided to the application either by copying of the page, or
734 by memory-mapping the page. Data is written i 728 by memory-mapping the page. Data is written into the address space by
735 the application, and then written-back to stor 729 the application, and then written-back to storage typically in whole
736 pages, however the address_space has finer con 730 pages, however the address_space has finer control of write sizes.
737 731
738 The read process essentially only requires 're 732 The read process essentially only requires 'read_folio'. The write
739 process is more complicated and uses write_beg 733 process is more complicated and uses write_begin/write_end or
740 dirty_folio to write data into the address_spa 734 dirty_folio to write data into the address_space, and writepage and
741 writepages to writeback data to storage. 735 writepages to writeback data to storage.
742 736
743 Adding and removing pages to/from an address_s 737 Adding and removing pages to/from an address_space is protected by the
744 inode's i_mutex. 738 inode's i_mutex.
745 739
746 When data is written to a page, the PG_Dirty f 740 When data is written to a page, the PG_Dirty flag should be set. It
747 typically remains set until writepage asks for 741 typically remains set until writepage asks for it to be written. This
748 should clear PG_Dirty and set PG_Writeback. I 742 should clear PG_Dirty and set PG_Writeback. It can be actually written
749 at any point after PG_Dirty is clear. Once it 743 at any point after PG_Dirty is clear. Once it is known to be safe,
750 PG_Writeback is cleared. 744 PG_Writeback is cleared.
751 745
752 Writeback makes use of a writeback_control str 746 Writeback makes use of a writeback_control structure to direct the
753 operations. This gives the writepage and writ 747 operations. This gives the writepage and writepages operations some
754 information about the nature of and reason for 748 information about the nature of and reason for the writeback request,
755 and the constraints under which it is being do 749 and the constraints under which it is being done. It is also used to
756 return information back to the caller about th 750 return information back to the caller about the result of a writepage or
757 writepages request. 751 writepages request.
758 752
759 753
760 Handling errors during writeback 754 Handling errors during writeback
761 -------------------------------- 755 --------------------------------
762 756
763 Most applications that do buffered I/O will pe 757 Most applications that do buffered I/O will periodically call a file
764 synchronization call (fsync, fdatasync, msync 758 synchronization call (fsync, fdatasync, msync or sync_file_range) to
765 ensure that data written has made it to the ba 759 ensure that data written has made it to the backing store. When there
766 is an error during writeback, they expect that 760 is an error during writeback, they expect that error to be reported when
767 a file sync request is made. After an error h 761 a file sync request is made. After an error has been reported on one
768 request, subsequent requests on the same file 762 request, subsequent requests on the same file descriptor should return
769 0, unless further writeback errors have occurr 763 0, unless further writeback errors have occurred since the previous file
770 synchronization. !! 764 syncronization.
771 765
772 Ideally, the kernel would report errors only o 766 Ideally, the kernel would report errors only on file descriptions on
773 which writes were done that subsequently faile 767 which writes were done that subsequently failed to be written back. The
774 generic pagecache infrastructure does not trac 768 generic pagecache infrastructure does not track the file descriptions
775 that have dirtied each individual page however 769 that have dirtied each individual page however, so determining which
776 file descriptors should get back an error is n 770 file descriptors should get back an error is not possible.
777 771
778 Instead, the generic writeback error tracking 772 Instead, the generic writeback error tracking infrastructure in the
779 kernel settles for reporting errors to fsync o 773 kernel settles for reporting errors to fsync on all file descriptions
780 that were open at the time that the error occu 774 that were open at the time that the error occurred. In a situation with
781 multiple writers, all of them will get back an 775 multiple writers, all of them will get back an error on a subsequent
782 fsync, even if all of the writes done through 776 fsync, even if all of the writes done through that particular file
783 descriptor succeeded (or even if there were no 777 descriptor succeeded (or even if there were no writes on that file
784 descriptor at all). 778 descriptor at all).
785 779
786 Filesystems that wish to use this infrastructu 780 Filesystems that wish to use this infrastructure should call
787 mapping_set_error to record the error in the a 781 mapping_set_error to record the error in the address_space when it
788 occurs. Then, after writing back data from th 782 occurs. Then, after writing back data from the pagecache in their
789 file->fsync operation, they should call file_c 783 file->fsync operation, they should call file_check_and_advance_wb_err to
790 ensure that the struct file's error cursor has 784 ensure that the struct file's error cursor has advanced to the correct
791 point in the stream of errors emitted by the b 785 point in the stream of errors emitted by the backing device(s).
792 786
793 787
794 struct address_space_operations 788 struct address_space_operations
795 ------------------------------- 789 -------------------------------
796 790
797 This describes how the VFS can manipulate mapp 791 This describes how the VFS can manipulate mapping of a file to page
798 cache in your filesystem. The following membe 792 cache in your filesystem. The following members are defined:
799 793
800 .. code-block:: c 794 .. code-block:: c
801 795
802 struct address_space_operations { 796 struct address_space_operations {
803 int (*writepage)(struct page * 797 int (*writepage)(struct page *page, struct writeback_control *wbc);
804 int (*read_folio)(struct file 798 int (*read_folio)(struct file *, struct folio *);
805 int (*writepages)(struct addre 799 int (*writepages)(struct address_space *, struct writeback_control *);
806 bool (*dirty_folio)(struct add 800 bool (*dirty_folio)(struct address_space *, struct folio *);
807 void (*readahead)(struct reada 801 void (*readahead)(struct readahead_control *);
808 int (*write_begin)(struct file 802 int (*write_begin)(struct file *, struct address_space *mapping,
809 loff_t pos, 803 loff_t pos, unsigned len,
810 struct page ** 804 struct page **pagep, void **fsdata);
811 int (*write_end)(struct file * 805 int (*write_end)(struct file *, struct address_space *mapping,
812 loff_t pos, u 806 loff_t pos, unsigned len, unsigned copied,
813 struct folio !! 807 struct page *page, void *fsdata);
814 sector_t (*bmap)(struct addres 808 sector_t (*bmap)(struct address_space *, sector_t);
815 void (*invalidate_folio) (stru 809 void (*invalidate_folio) (struct folio *, size_t start, size_t len);
816 bool (*release_folio)(struct f 810 bool (*release_folio)(struct folio *, gfp_t);
817 void (*free_folio)(struct foli 811 void (*free_folio)(struct folio *);
818 ssize_t (*direct_IO)(struct ki 812 ssize_t (*direct_IO)(struct kiocb *, struct iov_iter *iter);
819 int (*migrate_folio)(struct ma 813 int (*migrate_folio)(struct mapping *, struct folio *dst,
820 struct folio * 814 struct folio *src, enum migrate_mode);
821 int (*launder_folio) (struct f 815 int (*launder_folio) (struct folio *);
822 816
823 bool (*is_partially_uptodate) 817 bool (*is_partially_uptodate) (struct folio *, size_t from,
824 818 size_t count);
825 void (*is_dirty_writeback)(str 819 void (*is_dirty_writeback)(struct folio *, bool *, bool *);
826 int (*error_remove_folio)(stru !! 820 int (*error_remove_page) (struct mapping *mapping, struct page *page);
827 int (*swap_activate)(struct sw 821 int (*swap_activate)(struct swap_info_struct *sis, struct file *f, sector_t *span)
828 int (*swap_deactivate)(struct 822 int (*swap_deactivate)(struct file *);
829 int (*swap_rw)(struct kiocb *i 823 int (*swap_rw)(struct kiocb *iocb, struct iov_iter *iter);
830 }; 824 };
831 825
832 ``writepage`` 826 ``writepage``
833 called by the VM to write a dirty page 827 called by the VM to write a dirty page to backing store. This
834 may happen for data integrity reasons 828 may happen for data integrity reasons (i.e. 'sync'), or to free
835 up memory (flush). The difference can 829 up memory (flush). The difference can be seen in
836 wbc->sync_mode. The PG_Dirty flag has 830 wbc->sync_mode. The PG_Dirty flag has been cleared and
837 PageLocked is true. writepage should 831 PageLocked is true. writepage should start writeout, should set
838 PG_Writeback, and should make sure the 832 PG_Writeback, and should make sure the page is unlocked, either
839 synchronously or asynchronously when t 833 synchronously or asynchronously when the write operation
840 completes. 834 completes.
841 835
842 If wbc->sync_mode is WB_SYNC_NONE, ->w 836 If wbc->sync_mode is WB_SYNC_NONE, ->writepage doesn't have to
843 try too hard if there are problems, an 837 try too hard if there are problems, and may choose to write out
844 other pages from the mapping if that i 838 other pages from the mapping if that is easier (e.g. due to
845 internal dependencies). If it chooses 839 internal dependencies). If it chooses not to start writeout, it
846 should return AOP_WRITEPAGE_ACTIVATE s 840 should return AOP_WRITEPAGE_ACTIVATE so that the VM will not
847 keep calling ->writepage on that page. 841 keep calling ->writepage on that page.
848 842
849 See the file "Locking" for more detail 843 See the file "Locking" for more details.
850 844
851 ``read_folio`` 845 ``read_folio``
852 Called by the page cache to read a fol 846 Called by the page cache to read a folio from the backing store.
853 The 'file' argument supplies authentic 847 The 'file' argument supplies authentication information to network
854 filesystems, and is generally not used 848 filesystems, and is generally not used by block based filesystems.
855 It may be NULL if the caller does not 849 It may be NULL if the caller does not have an open file (eg if
856 the kernel is performing a read for it 850 the kernel is performing a read for itself rather than on behalf
857 of a userspace process with an open fi 851 of a userspace process with an open file).
858 852
859 If the mapping does not support large 853 If the mapping does not support large folios, the folio will
860 contain a single page. The folio will 854 contain a single page. The folio will be locked when read_folio
861 is called. If the read completes succ 855 is called. If the read completes successfully, the folio should
862 be marked uptodate. The filesystem sh 856 be marked uptodate. The filesystem should unlock the folio
863 once the read has completed, whether i 857 once the read has completed, whether it was successful or not.
864 The filesystem does not need to modify 858 The filesystem does not need to modify the refcount on the folio;
865 the page cache holds a reference count 859 the page cache holds a reference count and that will not be
866 released until the folio is unlocked. 860 released until the folio is unlocked.
867 861
868 Filesystems may implement ->read_folio 862 Filesystems may implement ->read_folio() synchronously.
869 In normal operation, folios are read t 863 In normal operation, folios are read through the ->readahead()
870 method. Only if this fails, or if the 864 method. Only if this fails, or if the caller needs to wait for
871 the read to complete will the page cac 865 the read to complete will the page cache call ->read_folio().
872 Filesystems should not attempt to perf 866 Filesystems should not attempt to perform their own readahead
873 in the ->read_folio() operation. 867 in the ->read_folio() operation.
874 868
875 If the filesystem cannot perform the r 869 If the filesystem cannot perform the read at this time, it can
876 unlock the folio, do whatever action i 870 unlock the folio, do whatever action it needs to ensure that the
877 read will succeed in the future and re 871 read will succeed in the future and return AOP_TRUNCATED_PAGE.
878 In this case, the caller should look u 872 In this case, the caller should look up the folio, lock it,
879 and call ->read_folio again. 873 and call ->read_folio again.
880 874
881 Callers may invoke the ->read_folio() 875 Callers may invoke the ->read_folio() method directly, but using
882 read_mapping_folio() will take care of 876 read_mapping_folio() will take care of locking, waiting for the
883 read to complete and handle cases such 877 read to complete and handle cases such as AOP_TRUNCATED_PAGE.
884 878
885 ``writepages`` 879 ``writepages``
886 called by the VM to write out pages as 880 called by the VM to write out pages associated with the
887 address_space object. If wbc->sync_mo 881 address_space object. If wbc->sync_mode is WB_SYNC_ALL, then
888 the writeback_control will specify a r 882 the writeback_control will specify a range of pages that must be
889 written out. If it is WB_SYNC_NONE, t 883 written out. If it is WB_SYNC_NONE, then a nr_to_write is
890 given and that many pages should be wr 884 given and that many pages should be written if possible. If no
891 ->writepages is given, then mpage_writ 885 ->writepages is given, then mpage_writepages is used instead.
892 This will choose pages from the addres 886 This will choose pages from the address space that are tagged as
893 DIRTY and will pass them to ->writepag 887 DIRTY and will pass them to ->writepage.
894 888
895 ``dirty_folio`` 889 ``dirty_folio``
896 called by the VM to mark a folio as di 890 called by the VM to mark a folio as dirty. This is particularly
897 needed if an address space attaches pr 891 needed if an address space attaches private data to a folio, and
898 that data needs to be updated when a f 892 that data needs to be updated when a folio is dirtied. This is
899 called, for example, when a memory map 893 called, for example, when a memory mapped page gets modified.
900 If defined, it should set the folio di 894 If defined, it should set the folio dirty flag, and the
901 PAGECACHE_TAG_DIRTY search mark in i_p 895 PAGECACHE_TAG_DIRTY search mark in i_pages.
902 896
903 ``readahead`` 897 ``readahead``
904 Called by the VM to read pages associa 898 Called by the VM to read pages associated with the address_space
905 object. The pages are consecutive in 899 object. The pages are consecutive in the page cache and are
906 locked. The implementation should dec 900 locked. The implementation should decrement the page refcount
907 after starting I/O on each page. Usua 901 after starting I/O on each page. Usually the page will be
908 unlocked by the I/O completion handler 902 unlocked by the I/O completion handler. The set of pages are
909 divided into some sync pages followed 903 divided into some sync pages followed by some async pages,
910 rac->ra->async_size gives the number o 904 rac->ra->async_size gives the number of async pages. The
911 filesystem should attempt to read all 905 filesystem should attempt to read all sync pages but may decide
912 to stop once it reaches the async page 906 to stop once it reaches the async pages. If it does decide to
913 stop attempting I/O, it can simply ret 907 stop attempting I/O, it can simply return. The caller will
914 remove the remaining pages from the ad 908 remove the remaining pages from the address space, unlock them
915 and decrement the page refcount. Set 909 and decrement the page refcount. Set PageUptodate if the I/O
916 completes successfully. !! 910 completes successfully. Setting PageError on any page will be
>> 911 ignored; simply unlock the page if an I/O error occurs.
917 912
918 ``write_begin`` 913 ``write_begin``
919 Called by the generic buffered write c 914 Called by the generic buffered write code to ask the filesystem
920 to prepare to write len bytes at the g 915 to prepare to write len bytes at the given offset in the file.
921 The address_space should check that th 916 The address_space should check that the write will be able to
922 complete, by allocating space if neces 917 complete, by allocating space if necessary and doing any other
923 internal housekeeping. If the write w 918 internal housekeeping. If the write will update parts of any
924 basic-blocks on storage, then those bl 919 basic-blocks on storage, then those blocks should be pre-read
925 (if they haven't been read already) so 920 (if they haven't been read already) so that the updated blocks
926 can be written out properly. 921 can be written out properly.
927 922
928 The filesystem must return the locked !! 923 The filesystem must return the locked pagecache page for the
929 specified offset, in ``*foliop``, for !! 924 specified offset, in ``*pagep``, for the caller to write into.
930 925
931 It must be able to cope with short wri 926 It must be able to cope with short writes (where the length
932 passed to write_begin is greater than 927 passed to write_begin is greater than the number of bytes copied
933 into the folio). !! 928 into the page).
934 929
935 A void * may be returned in fsdata, wh 930 A void * may be returned in fsdata, which then gets passed into
936 write_end. 931 write_end.
937 932
938 Returns 0 on success; < 0 on failure ( 933 Returns 0 on success; < 0 on failure (which is the error code),
939 in which case write_end is not called. 934 in which case write_end is not called.
940 935
941 ``write_end`` 936 ``write_end``
942 After a successful write_begin, and da 937 After a successful write_begin, and data copy, write_end must be
943 called. len is the original len passe 938 called. len is the original len passed to write_begin, and
944 copied is the amount that was able to 939 copied is the amount that was able to be copied.
945 940
946 The filesystem must take care of unloc !! 941 The filesystem must take care of unlocking the page and
947 decrementing its refcount, and updatin !! 942 releasing it refcount, and updating i_size.
948 943
949 Returns < 0 on failure, otherwise the 944 Returns < 0 on failure, otherwise the number of bytes (<=
950 'copied') that were able to be copied 945 'copied') that were able to be copied into pagecache.
951 946
952 ``bmap`` 947 ``bmap``
953 called by the VFS to map a logical blo 948 called by the VFS to map a logical block offset within object to
954 physical block number. This method is 949 physical block number. This method is used by the FIBMAP ioctl
955 and for working with swap-files. To b 950 and for working with swap-files. To be able to swap to a file,
956 the file must have a stable mapping to 951 the file must have a stable mapping to a block device. The swap
957 system does not go through the filesys 952 system does not go through the filesystem but instead uses bmap
958 to find out where the blocks in the fi 953 to find out where the blocks in the file are and uses those
959 addresses directly. 954 addresses directly.
960 955
961 ``invalidate_folio`` 956 ``invalidate_folio``
962 If a folio has private data, then inva 957 If a folio has private data, then invalidate_folio will be
963 called when part or all of the folio i 958 called when part or all of the folio is to be removed from the
964 address space. This generally corresp 959 address space. This generally corresponds to either a
965 truncation, punch hole or a complete i 960 truncation, punch hole or a complete invalidation of the address
966 space (in the latter case 'offset' wil 961 space (in the latter case 'offset' will always be 0 and 'length'
967 will be folio_size()). Any private da 962 will be folio_size()). Any private data associated with the folio
968 should be updated to reflect this trun 963 should be updated to reflect this truncation. If offset is 0
969 and length is folio_size(), then the p 964 and length is folio_size(), then the private data should be
970 released, because the folio must be ab 965 released, because the folio must be able to be completely
971 discarded. This may be done by callin 966 discarded. This may be done by calling the ->release_folio
972 function, but in this case the release 967 function, but in this case the release MUST succeed.
973 968
974 ``release_folio`` 969 ``release_folio``
975 release_folio is called on folios with 970 release_folio is called on folios with private data to tell the
976 filesystem that the folio is about to 971 filesystem that the folio is about to be freed. ->release_folio
977 should remove any private data from th 972 should remove any private data from the folio and clear the
978 private flag. If release_folio() fail 973 private flag. If release_folio() fails, it should return false.
979 release_folio() is used in two distinc 974 release_folio() is used in two distinct though related cases.
980 The first is when the VM wants to free 975 The first is when the VM wants to free a clean folio with no
981 active users. If ->release_folio succ 976 active users. If ->release_folio succeeds, the folio will be
982 removed from the address_space and be 977 removed from the address_space and be freed.
983 978
984 The second case is when a request has 979 The second case is when a request has been made to invalidate
985 some or all folios in an address_space 980 some or all folios in an address_space. This can happen
986 through the fadvise(POSIX_FADV_DONTNEE 981 through the fadvise(POSIX_FADV_DONTNEED) system call or by the
987 filesystem explicitly requesting it as 982 filesystem explicitly requesting it as nfs and 9p do (when they
988 believe the cache may be out of date w 983 believe the cache may be out of date with storage) by calling
989 invalidate_inode_pages2(). If the fil 984 invalidate_inode_pages2(). If the filesystem makes such a call,
990 and needs to be certain that all folio 985 and needs to be certain that all folios are invalidated, then
991 its release_folio will need to ensure 986 its release_folio will need to ensure this. Possibly it can
992 clear the uptodate flag if it cannot f 987 clear the uptodate flag if it cannot free private data yet.
993 988
994 ``free_folio`` 989 ``free_folio``
995 free_folio is called once the folio is 990 free_folio is called once the folio is no longer visible in the
996 page cache in order to allow the clean 991 page cache in order to allow the cleanup of any private data.
997 Since it may be called by the memory r 992 Since it may be called by the memory reclaimer, it should not
998 assume that the original address_space 993 assume that the original address_space mapping still exists, and
999 it should not block. 994 it should not block.
1000 995
1001 ``direct_IO`` 996 ``direct_IO``
1002 called by the generic read/write rout 997 called by the generic read/write routines to perform direct_IO -
1003 that is IO requests which bypass the 998 that is IO requests which bypass the page cache and transfer
1004 data directly between the storage and 999 data directly between the storage and the application's address
1005 space. 1000 space.
1006 1001
1007 ``migrate_folio`` 1002 ``migrate_folio``
1008 This is used to compact the physical 1003 This is used to compact the physical memory usage. If the VM
1009 wants to relocate a folio (maybe from 1004 wants to relocate a folio (maybe from a memory device that is
1010 signalling imminent failure) it will 1005 signalling imminent failure) it will pass a new folio and an old
1011 folio to this function. migrate_foli 1006 folio to this function. migrate_folio should transfer any private
1012 data across and update any references 1007 data across and update any references that it has to the folio.
1013 1008
1014 ``launder_folio`` 1009 ``launder_folio``
1015 Called before freeing a folio - it wr 1010 Called before freeing a folio - it writes back the dirty folio.
1016 To prevent redirtying the folio, it i 1011 To prevent redirtying the folio, it is kept locked during the
1017 whole operation. 1012 whole operation.
1018 1013
1019 ``is_partially_uptodate`` 1014 ``is_partially_uptodate``
1020 Called by the VM when reading a file 1015 Called by the VM when reading a file through the pagecache when
1021 the underlying blocksize is smaller t 1016 the underlying blocksize is smaller than the size of the folio.
1022 If the required block is up to date t 1017 If the required block is up to date then the read can complete
1023 without needing I/O to bring the whol 1018 without needing I/O to bring the whole page up to date.
1024 1019
1025 ``is_dirty_writeback`` 1020 ``is_dirty_writeback``
1026 Called by the VM when attempting to r 1021 Called by the VM when attempting to reclaim a folio. The VM uses
1027 dirty and writeback information to de 1022 dirty and writeback information to determine if it needs to
1028 stall to allow flushers a chance to c 1023 stall to allow flushers a chance to complete some IO.
1029 Ordinarily it can use folio_test_dirt 1024 Ordinarily it can use folio_test_dirty and folio_test_writeback but
1030 some filesystems have more complex st 1025 some filesystems have more complex state (unstable folios in NFS
1031 prevent reclaim) or do not set those 1026 prevent reclaim) or do not set those flags due to locking
1032 problems. This callback allows a fil 1027 problems. This callback allows a filesystem to indicate to the
1033 VM if a folio should be treated as di 1028 VM if a folio should be treated as dirty or writeback for the
1034 purposes of stalling. 1029 purposes of stalling.
1035 1030
1036 ``error_remove_folio`` !! 1031 ``error_remove_page``
1037 normally set to generic_error_remove_ !! 1032 normally set to generic_error_remove_page if truncation is ok
1038 for this address space. Used for mem 1033 for this address space. Used for memory failure handling.
1039 Setting this implies you deal with pa 1034 Setting this implies you deal with pages going away under you,
1040 unless you have them locked or refere 1035 unless you have them locked or reference counts increased.
1041 1036
1042 ``swap_activate`` 1037 ``swap_activate``
1043 1038
1044 Called to prepare the given file for 1039 Called to prepare the given file for swap. It should perform
1045 any validation and preparation necess 1040 any validation and preparation necessary to ensure that writes
1046 can be performed with minimal memory 1041 can be performed with minimal memory allocation. It should call
1047 add_swap_extent(), or the helper ioma 1042 add_swap_extent(), or the helper iomap_swapfile_activate(), and
1048 return the number of extents added. 1043 return the number of extents added. If IO should be submitted
1049 through ->swap_rw(), it should set SW 1044 through ->swap_rw(), it should set SWP_FS_OPS, otherwise IO will
1050 be submitted directly to the block de 1045 be submitted directly to the block device ``sis->bdev``.
1051 1046
1052 ``swap_deactivate`` 1047 ``swap_deactivate``
1053 Called during swapoff on files where 1048 Called during swapoff on files where swap_activate was
1054 successful. 1049 successful.
1055 1050
1056 ``swap_rw`` 1051 ``swap_rw``
1057 Called to read or write swap pages wh 1052 Called to read or write swap pages when SWP_FS_OPS is set.
1058 1053
1059 The File Object 1054 The File Object
1060 =============== 1055 ===============
1061 1056
1062 A file object represents a file opened by a p 1057 A file object represents a file opened by a process. This is also known
1063 as an "open file description" in POSIX parlan 1058 as an "open file description" in POSIX parlance.
1064 1059
1065 1060
1066 struct file_operations 1061 struct file_operations
1067 ---------------------- 1062 ----------------------
1068 1063
1069 This describes how the VFS can manipulate an 1064 This describes how the VFS can manipulate an open file. As of kernel
1070 4.18, the following members are defined: 1065 4.18, the following members are defined:
1071 1066
1072 .. code-block:: c 1067 .. code-block:: c
1073 1068
1074 struct file_operations { 1069 struct file_operations {
1075 struct module *owner; 1070 struct module *owner;
1076 loff_t (*llseek) (struct file 1071 loff_t (*llseek) (struct file *, loff_t, int);
1077 ssize_t (*read) (struct file 1072 ssize_t (*read) (struct file *, char __user *, size_t, loff_t *);
1078 ssize_t (*write) (struct file 1073 ssize_t (*write) (struct file *, const char __user *, size_t, loff_t *);
1079 ssize_t (*read_iter) (struct 1074 ssize_t (*read_iter) (struct kiocb *, struct iov_iter *);
1080 ssize_t (*write_iter) (struct 1075 ssize_t (*write_iter) (struct kiocb *, struct iov_iter *);
1081 int (*iopoll)(struct kiocb *k 1076 int (*iopoll)(struct kiocb *kiocb, bool spin);
>> 1077 int (*iterate) (struct file *, struct dir_context *);
1082 int (*iterate_shared) (struct 1078 int (*iterate_shared) (struct file *, struct dir_context *);
1083 __poll_t (*poll) (struct file 1079 __poll_t (*poll) (struct file *, struct poll_table_struct *);
1084 long (*unlocked_ioctl) (struc 1080 long (*unlocked_ioctl) (struct file *, unsigned int, unsigned long);
1085 long (*compat_ioctl) (struct 1081 long (*compat_ioctl) (struct file *, unsigned int, unsigned long);
1086 int (*mmap) (struct file *, s 1082 int (*mmap) (struct file *, struct vm_area_struct *);
1087 int (*open) (struct inode *, 1083 int (*open) (struct inode *, struct file *);
1088 int (*flush) (struct file *, 1084 int (*flush) (struct file *, fl_owner_t id);
1089 int (*release) (struct inode 1085 int (*release) (struct inode *, struct file *);
1090 int (*fsync) (struct file *, 1086 int (*fsync) (struct file *, loff_t, loff_t, int datasync);
1091 int (*fasync) (int, struct fi 1087 int (*fasync) (int, struct file *, int);
1092 int (*lock) (struct file *, i 1088 int (*lock) (struct file *, int, struct file_lock *);
>> 1089 ssize_t (*sendpage) (struct file *, struct page *, int, size_t, loff_t *, int);
1093 unsigned long (*get_unmapped_ 1090 unsigned long (*get_unmapped_area)(struct file *, unsigned long, unsigned long, unsigned long, unsigned long);
1094 int (*check_flags)(int); 1091 int (*check_flags)(int);
1095 int (*flock) (struct file *, 1092 int (*flock) (struct file *, int, struct file_lock *);
1096 ssize_t (*splice_write)(struc 1093 ssize_t (*splice_write)(struct pipe_inode_info *, struct file *, loff_t *, size_t, unsigned int);
1097 ssize_t (*splice_read)(struct 1094 ssize_t (*splice_read)(struct file *, loff_t *, struct pipe_inode_info *, size_t, unsigned int);
1098 int (*setlease)(struct file * 1095 int (*setlease)(struct file *, long, struct file_lock **, void **);
1099 long (*fallocate)(struct file 1096 long (*fallocate)(struct file *file, int mode, loff_t offset,
1100 loff_t len) 1097 loff_t len);
1101 void (*show_fdinfo)(struct se 1098 void (*show_fdinfo)(struct seq_file *m, struct file *f);
1102 #ifndef CONFIG_MMU 1099 #ifndef CONFIG_MMU
1103 unsigned (*mmap_capabilities) 1100 unsigned (*mmap_capabilities)(struct file *);
1104 #endif 1101 #endif
1105 ssize_t (*copy_file_range)(st 1102 ssize_t (*copy_file_range)(struct file *, loff_t, struct file *, loff_t, size_t, unsigned int);
1106 loff_t (*remap_file_range)(st 1103 loff_t (*remap_file_range)(struct file *file_in, loff_t pos_in,
1107 st 1104 struct file *file_out, loff_t pos_out,
1108 lo 1105 loff_t len, unsigned int remap_flags);
1109 int (*fadvise)(struct file *, 1106 int (*fadvise)(struct file *, loff_t, loff_t, int);
1110 }; 1107 };
1111 1108
1112 Again, all methods are called without any loc 1109 Again, all methods are called without any locks being held, unless
1113 otherwise noted. 1110 otherwise noted.
1114 1111
1115 ``llseek`` 1112 ``llseek``
1116 called when the VFS needs to move the 1113 called when the VFS needs to move the file position index
1117 1114
1118 ``read`` 1115 ``read``
1119 called by read(2) and related system 1116 called by read(2) and related system calls
1120 1117
1121 ``read_iter`` 1118 ``read_iter``
1122 possibly asynchronous read with iov_i 1119 possibly asynchronous read with iov_iter as destination
1123 1120
1124 ``write`` 1121 ``write``
1125 called by write(2) and related system 1122 called by write(2) and related system calls
1126 1123
1127 ``write_iter`` 1124 ``write_iter``
1128 possibly asynchronous write with iov_ 1125 possibly asynchronous write with iov_iter as source
1129 1126
1130 ``iopoll`` 1127 ``iopoll``
1131 called when aio wants to poll for com 1128 called when aio wants to poll for completions on HIPRI iocbs
1132 1129
1133 ``iterate_shared`` !! 1130 ``iterate``
1134 called when the VFS needs to read the 1131 called when the VFS needs to read the directory contents
1135 1132
>> 1133 ``iterate_shared``
>> 1134 called when the VFS needs to read the directory contents when
>> 1135 filesystem supports concurrent dir iterators
>> 1136
1136 ``poll`` 1137 ``poll``
1137 called by the VFS when a process want 1138 called by the VFS when a process wants to check if there is
1138 activity on this file and (optionally 1139 activity on this file and (optionally) go to sleep until there
1139 is activity. Called by the select(2) 1140 is activity. Called by the select(2) and poll(2) system calls
1140 1141
1141 ``unlocked_ioctl`` 1142 ``unlocked_ioctl``
1142 called by the ioctl(2) system call. 1143 called by the ioctl(2) system call.
1143 1144
1144 ``compat_ioctl`` 1145 ``compat_ioctl``
1145 called by the ioctl(2) system call wh 1146 called by the ioctl(2) system call when 32 bit system calls are
1146 used on 64 bit kernels. 1147 used on 64 bit kernels.
1147 1148
1148 ``mmap`` 1149 ``mmap``
1149 called by the mmap(2) system call 1150 called by the mmap(2) system call
1150 1151
1151 ``open`` 1152 ``open``
1152 called by the VFS when an inode shoul 1153 called by the VFS when an inode should be opened. When the VFS
1153 opens a file, it creates a new "struc 1154 opens a file, it creates a new "struct file". It then calls the
1154 open method for the newly allocated f 1155 open method for the newly allocated file structure. You might
1155 think that the open method really bel 1156 think that the open method really belongs in "struct
1156 inode_operations", and you may be rig 1157 inode_operations", and you may be right. I think it's done the
1157 way it is because it makes filesystem 1158 way it is because it makes filesystems simpler to implement.
1158 The open() method is a good place to 1159 The open() method is a good place to initialize the
1159 "private_data" member in the file str 1160 "private_data" member in the file structure if you want to point
1160 to a device structure 1161 to a device structure
1161 1162
1162 ``flush`` 1163 ``flush``
1163 called by the close(2) system call to 1164 called by the close(2) system call to flush a file
1164 1165
1165 ``release`` 1166 ``release``
1166 called when the last reference to an 1167 called when the last reference to an open file is closed
1167 1168
1168 ``fsync`` 1169 ``fsync``
1169 called by the fsync(2) system call. 1170 called by the fsync(2) system call. Also see the section above
1170 entitled "Handling errors during writ 1171 entitled "Handling errors during writeback".
1171 1172
1172 ``fasync`` 1173 ``fasync``
1173 called by the fcntl(2) system call wh 1174 called by the fcntl(2) system call when asynchronous
1174 (non-blocking) mode is enabled for a 1175 (non-blocking) mode is enabled for a file
1175 1176
1176 ``lock`` 1177 ``lock``
1177 called by the fcntl(2) system call fo 1178 called by the fcntl(2) system call for F_GETLK, F_SETLK, and
1178 F_SETLKW commands 1179 F_SETLKW commands
1179 1180
1180 ``get_unmapped_area`` 1181 ``get_unmapped_area``
1181 called by the mmap(2) system call 1182 called by the mmap(2) system call
1182 1183
1183 ``check_flags`` 1184 ``check_flags``
1184 called by the fcntl(2) system call fo 1185 called by the fcntl(2) system call for F_SETFL command
1185 1186
1186 ``flock`` 1187 ``flock``
1187 called by the flock(2) system call 1188 called by the flock(2) system call
1188 1189
1189 ``splice_write`` 1190 ``splice_write``
1190 called by the VFS to splice data from 1191 called by the VFS to splice data from a pipe to a file. This
1191 method is used by the splice(2) syste 1192 method is used by the splice(2) system call
1192 1193
1193 ``splice_read`` 1194 ``splice_read``
1194 called by the VFS to splice data from 1195 called by the VFS to splice data from file to a pipe. This
1195 method is used by the splice(2) syste 1196 method is used by the splice(2) system call
1196 1197
1197 ``setlease`` 1198 ``setlease``
1198 called by the VFS to set or release a 1199 called by the VFS to set or release a file lock lease. setlease
1199 implementations should call generic_s 1200 implementations should call generic_setlease to record or remove
1200 the lease in the inode after setting 1201 the lease in the inode after setting it.
1201 1202
1202 ``fallocate`` 1203 ``fallocate``
1203 called by the VFS to preallocate bloc 1204 called by the VFS to preallocate blocks or punch a hole.
1204 1205
1205 ``copy_file_range`` 1206 ``copy_file_range``
1206 called by the copy_file_range(2) syst 1207 called by the copy_file_range(2) system call.
1207 1208
1208 ``remap_file_range`` 1209 ``remap_file_range``
1209 called by the ioctl(2) system call fo 1210 called by the ioctl(2) system call for FICLONERANGE and FICLONE
1210 and FIDEDUPERANGE commands to remap f 1211 and FIDEDUPERANGE commands to remap file ranges. An
1211 implementation should remap len bytes 1212 implementation should remap len bytes at pos_in of the source
1212 file into the dest file at pos_out. 1213 file into the dest file at pos_out. Implementations must handle
1213 callers passing in len == 0; this mea 1214 callers passing in len == 0; this means "remap to the end of the
1214 source file". The return value shoul 1215 source file". The return value should the number of bytes
1215 remapped, or the usual negative error 1216 remapped, or the usual negative error code if errors occurred
1216 before any bytes were remapped. The 1217 before any bytes were remapped. The remap_flags parameter
1217 accepts REMAP_FILE_* flags. If REMAP 1218 accepts REMAP_FILE_* flags. If REMAP_FILE_DEDUP is set then the
1218 implementation must only remap if the 1219 implementation must only remap if the requested file ranges have
1219 identical contents. If REMAP_FILE_CA 1220 identical contents. If REMAP_FILE_CAN_SHORTEN is set, the caller is
1220 ok with the implementation shortening 1221 ok with the implementation shortening the request length to
1221 satisfy alignment or EOF requirements 1222 satisfy alignment or EOF requirements (or any other reason).
1222 1223
1223 ``fadvise`` 1224 ``fadvise``
1224 possibly called by the fadvise64() sy 1225 possibly called by the fadvise64() system call.
1225 1226
1226 Note that the file operations are implemented 1227 Note that the file operations are implemented by the specific
1227 filesystem in which the inode resides. When 1228 filesystem in which the inode resides. When opening a device node
1228 (character or block special) most filesystems 1229 (character or block special) most filesystems will call special
1229 support routines in the VFS which will locate 1230 support routines in the VFS which will locate the required device
1230 driver information. These support routines r 1231 driver information. These support routines replace the filesystem file
1231 operations with those for the device driver, 1232 operations with those for the device driver, and then proceed to call
1232 the new open() method for the file. This is 1233 the new open() method for the file. This is how opening a device file
1233 in the filesystem eventually ends up calling 1234 in the filesystem eventually ends up calling the device driver open()
1234 method. 1235 method.
1235 1236
1236 1237
1237 Directory Entry Cache (dcache) 1238 Directory Entry Cache (dcache)
1238 ============================== 1239 ==============================
1239 1240
1240 1241
1241 struct dentry_operations 1242 struct dentry_operations
1242 ------------------------ 1243 ------------------------
1243 1244
1244 This describes how a filesystem can overload 1245 This describes how a filesystem can overload the standard dentry
1245 operations. Dentries and the dcache are the 1246 operations. Dentries and the dcache are the domain of the VFS and the
1246 individual filesystem implementations. Devic 1247 individual filesystem implementations. Device drivers have no business
1247 here. These methods may be set to NULL, as t 1248 here. These methods may be set to NULL, as they are either optional or
1248 the VFS uses a default. As of kernel 2.6.22, 1249 the VFS uses a default. As of kernel 2.6.22, the following members are
1249 defined: 1250 defined:
1250 1251
1251 .. code-block:: c 1252 .. code-block:: c
1252 1253
1253 struct dentry_operations { 1254 struct dentry_operations {
1254 int (*d_revalidate)(struct de 1255 int (*d_revalidate)(struct dentry *, unsigned int);
1255 int (*d_weak_revalidate)(stru 1256 int (*d_weak_revalidate)(struct dentry *, unsigned int);
1256 int (*d_hash)(const struct de 1257 int (*d_hash)(const struct dentry *, struct qstr *);
1257 int (*d_compare)(const struct 1258 int (*d_compare)(const struct dentry *,
1258 unsigned int 1259 unsigned int, const char *, const struct qstr *);
1259 int (*d_delete)(const struct 1260 int (*d_delete)(const struct dentry *);
1260 int (*d_init)(struct dentry * 1261 int (*d_init)(struct dentry *);
1261 void (*d_release)(struct dent 1262 void (*d_release)(struct dentry *);
1262 void (*d_iput)(struct dentry 1263 void (*d_iput)(struct dentry *, struct inode *);
1263 char *(*d_dname)(struct dentr 1264 char *(*d_dname)(struct dentry *, char *, int);
1264 struct vfsmount *(*d_automoun 1265 struct vfsmount *(*d_automount)(struct path *);
1265 int (*d_manage)(const struct 1266 int (*d_manage)(const struct path *, bool);
1266 struct dentry *(*d_real)(stru !! 1267 struct dentry *(*d_real)(struct dentry *, const struct inode *);
1267 }; 1268 };
1268 1269
1269 ``d_revalidate`` 1270 ``d_revalidate``
1270 called when the VFS needs to revalida 1271 called when the VFS needs to revalidate a dentry. This is
1271 called whenever a name look-up finds 1272 called whenever a name look-up finds a dentry in the dcache.
1272 Most local filesystems leave this as 1273 Most local filesystems leave this as NULL, because all their
1273 dentries in the dcache are valid. Ne 1274 dentries in the dcache are valid. Network filesystems are
1274 different since things can change on 1275 different since things can change on the server without the
1275 client necessarily being aware of it. 1276 client necessarily being aware of it.
1276 1277
1277 This function should return a positiv 1278 This function should return a positive value if the dentry is
1278 still valid, and zero or a negative e 1279 still valid, and zero or a negative error code if it isn't.
1279 1280
1280 d_revalidate may be called in rcu-wal 1281 d_revalidate may be called in rcu-walk mode (flags &
1281 LOOKUP_RCU). If in rcu-walk mode, th 1282 LOOKUP_RCU). If in rcu-walk mode, the filesystem must
1282 revalidate the dentry without blockin 1283 revalidate the dentry without blocking or storing to the dentry,
1283 d_parent and d_inode should not be us 1284 d_parent and d_inode should not be used without care (because
1284 they can change and, in d_inode case, 1285 they can change and, in d_inode case, even become NULL under
1285 us). 1286 us).
1286 1287
1287 If a situation is encountered that rc 1288 If a situation is encountered that rcu-walk cannot handle,
1288 return 1289 return
1289 -ECHILD and it will be called again i 1290 -ECHILD and it will be called again in ref-walk mode.
1290 1291
1291 ``d_weak_revalidate`` 1292 ``d_weak_revalidate``
1292 called when the VFS needs to revalida 1293 called when the VFS needs to revalidate a "jumped" dentry. This
1293 is called when a path-walk ends at de 1294 is called when a path-walk ends at dentry that was not acquired
1294 by doing a lookup in the parent direc 1295 by doing a lookup in the parent directory. This includes "/",
1295 "." and "..", as well as procfs-style 1296 "." and "..", as well as procfs-style symlinks and mountpoint
1296 traversal. 1297 traversal.
1297 1298
1298 In this case, we are less concerned w 1299 In this case, we are less concerned with whether the dentry is
1299 still fully correct, but rather that 1300 still fully correct, but rather that the inode is still valid.
1300 As with d_revalidate, most local file 1301 As with d_revalidate, most local filesystems will set this to
1301 NULL since their dcache entries are a 1302 NULL since their dcache entries are always valid.
1302 1303
1303 This function has the same return cod 1304 This function has the same return code semantics as
1304 d_revalidate. 1305 d_revalidate.
1305 1306
1306 d_weak_revalidate is only called afte 1307 d_weak_revalidate is only called after leaving rcu-walk mode.
1307 1308
1308 ``d_hash`` 1309 ``d_hash``
1309 called when the VFS adds a dentry to 1310 called when the VFS adds a dentry to the hash table. The first
1310 dentry passed to d_hash is the parent 1311 dentry passed to d_hash is the parent directory that the name is
1311 to be hashed into. 1312 to be hashed into.
1312 1313
1313 Same locking and synchronisation rule 1314 Same locking and synchronisation rules as d_compare regarding
1314 what is safe to dereference etc. 1315 what is safe to dereference etc.
1315 1316
1316 ``d_compare`` 1317 ``d_compare``
1317 called to compare a dentry name with 1318 called to compare a dentry name with a given name. The first
1318 dentry is the parent of the dentry to 1319 dentry is the parent of the dentry to be compared, the second is
1319 the child dentry. len and name strin 1320 the child dentry. len and name string are properties of the
1320 dentry to be compared. qstr is the n 1321 dentry to be compared. qstr is the name to compare it with.
1321 1322
1322 Must be constant and idempotent, and 1323 Must be constant and idempotent, and should not take locks if
1323 possible, and should not or store int 1324 possible, and should not or store into the dentry. Should not
1324 dereference pointers outside the dent 1325 dereference pointers outside the dentry without lots of care
1325 (eg. d_parent, d_inode, d_name shoul 1326 (eg. d_parent, d_inode, d_name should not be used).
1326 1327
1327 However, our vfsmount is pinned, and 1328 However, our vfsmount is pinned, and RCU held, so the dentries
1328 and inodes won't disappear, neither w 1329 and inodes won't disappear, neither will our sb or filesystem
1329 module. ->d_sb may be used. 1330 module. ->d_sb may be used.
1330 1331
1331 It is a tricky calling convention bec 1332 It is a tricky calling convention because it needs to be called
1332 under "rcu-walk", ie. without any loc 1333 under "rcu-walk", ie. without any locks or references on things.
1333 1334
1334 ``d_delete`` 1335 ``d_delete``
1335 called when the last reference to a d 1336 called when the last reference to a dentry is dropped and the
1336 dcache is deciding whether or not to 1337 dcache is deciding whether or not to cache it. Return 1 to
1337 delete immediately, or 0 to cache the 1338 delete immediately, or 0 to cache the dentry. Default is NULL
1338 which means to always cache a reachab 1339 which means to always cache a reachable dentry. d_delete must
1339 be constant and idempotent. 1340 be constant and idempotent.
1340 1341
1341 ``d_init`` 1342 ``d_init``
1342 called when a dentry is allocated 1343 called when a dentry is allocated
1343 1344
1344 ``d_release`` 1345 ``d_release``
1345 called when a dentry is really deallo 1346 called when a dentry is really deallocated
1346 1347
1347 ``d_iput`` 1348 ``d_iput``
1348 called when a dentry loses its inode 1349 called when a dentry loses its inode (just prior to its being
1349 deallocated). The default when this 1350 deallocated). The default when this is NULL is that the VFS
1350 calls iput(). If you define this met 1351 calls iput(). If you define this method, you must call iput()
1351 yourself 1352 yourself
1352 1353
1353 ``d_dname`` 1354 ``d_dname``
1354 called when the pathname of a dentry 1355 called when the pathname of a dentry should be generated.
1355 Useful for some pseudo filesystems (s 1356 Useful for some pseudo filesystems (sockfs, pipefs, ...) to
1356 delay pathname generation. (Instead 1357 delay pathname generation. (Instead of doing it when dentry is
1357 created, it's done only when the path 1358 created, it's done only when the path is needed.). Real
1358 filesystems probably dont want to use 1359 filesystems probably dont want to use it, because their dentries
1359 are present in global dcache hash, so 1360 are present in global dcache hash, so their hash should be an
1360 invariant. As no lock is held, d_dna 1361 invariant. As no lock is held, d_dname() should not try to
1361 modify the dentry itself, unless appr 1362 modify the dentry itself, unless appropriate SMP safety is used.
1362 CAUTION : d_path() logic is quite tri 1363 CAUTION : d_path() logic is quite tricky. The correct way to
1363 return for example "Hello" is to put 1364 return for example "Hello" is to put it at the end of the
1364 buffer, and returns a pointer to the 1365 buffer, and returns a pointer to the first char.
1365 dynamic_dname() helper function is pr 1366 dynamic_dname() helper function is provided to take care of
1366 this. 1367 this.
1367 1368
1368 Example : 1369 Example :
1369 1370
1370 .. code-block:: c 1371 .. code-block:: c
1371 1372
1372 static char *pipefs_dname(struct dent 1373 static char *pipefs_dname(struct dentry *dent, char *buffer, int buflen)
1373 { 1374 {
1374 return dynamic_dname(dentry, 1375 return dynamic_dname(dentry, buffer, buflen, "pipe:[%lu]",
1375 dentry->d_ino 1376 dentry->d_inode->i_ino);
1376 } 1377 }
1377 1378
1378 ``d_automount`` 1379 ``d_automount``
1379 called when an automount dentry is to 1380 called when an automount dentry is to be traversed (optional).
1380 This should create a new VFS mount re 1381 This should create a new VFS mount record and return the record
1381 to the caller. The caller is supplie 1382 to the caller. The caller is supplied with a path parameter
1382 giving the automount directory to des 1383 giving the automount directory to describe the automount target
1383 and the parent VFS mount record to pr 1384 and the parent VFS mount record to provide inheritable mount
1384 parameters. NULL should be returned 1385 parameters. NULL should be returned if someone else managed to
1385 make the automount first. If the vfs 1386 make the automount first. If the vfsmount creation failed, then
1386 an error code should be returned. If 1387 an error code should be returned. If -EISDIR is returned, then
1387 the directory will be treated as an o 1388 the directory will be treated as an ordinary directory and
1388 returned to pathwalk to continue walk 1389 returned to pathwalk to continue walking.
1389 1390
1390 If a vfsmount is returned, the caller 1391 If a vfsmount is returned, the caller will attempt to mount it
1391 on the mountpoint and will remove the 1392 on the mountpoint and will remove the vfsmount from its
1392 expiration list in the case of failur 1393 expiration list in the case of failure. The vfsmount should be
1393 returned with 2 refs on it to prevent 1394 returned with 2 refs on it to prevent automatic expiration - the
1394 caller will clean up the additional r 1395 caller will clean up the additional ref.
1395 1396
1396 This function is only used if DCACHE_ 1397 This function is only used if DCACHE_NEED_AUTOMOUNT is set on
1397 the dentry. This is set by __d_insta 1398 the dentry. This is set by __d_instantiate() if S_AUTOMOUNT is
1398 set on the inode being added. 1399 set on the inode being added.
1399 1400
1400 ``d_manage`` 1401 ``d_manage``
1401 called to allow the filesystem to man 1402 called to allow the filesystem to manage the transition from a
1402 dentry (optional). This allows autof 1403 dentry (optional). This allows autofs, for example, to hold up
1403 clients waiting to explore behind a ' 1404 clients waiting to explore behind a 'mountpoint' while letting
1404 the daemon go past and construct the 1405 the daemon go past and construct the subtree there. 0 should be
1405 returned to let the calling process c 1406 returned to let the calling process continue. -EISDIR can be
1406 returned to tell pathwalk to use this 1407 returned to tell pathwalk to use this directory as an ordinary
1407 directory and to ignore anything moun 1408 directory and to ignore anything mounted on it and not to check
1408 the automount flag. Any other error 1409 the automount flag. Any other error code will abort pathwalk
1409 completely. 1410 completely.
1410 1411
1411 If the 'rcu_walk' parameter is true, 1412 If the 'rcu_walk' parameter is true, then the caller is doing a
1412 pathwalk in RCU-walk mode. Sleeping 1413 pathwalk in RCU-walk mode. Sleeping is not permitted in this
1413 mode, and the caller can be asked to 1414 mode, and the caller can be asked to leave it and call again by
1414 returning -ECHILD. -EISDIR may also 1415 returning -ECHILD. -EISDIR may also be returned to tell
1415 pathwalk to ignore d_automount or any 1416 pathwalk to ignore d_automount or any mounts.
1416 1417
1417 This function is only used if DCACHE_ 1418 This function is only used if DCACHE_MANAGE_TRANSIT is set on
1418 the dentry being transited from. 1419 the dentry being transited from.
1419 1420
1420 ``d_real`` 1421 ``d_real``
1421 overlay/union type filesystems implem !! 1422 overlay/union type filesystems implement this method to return
1422 of the underlying dentries of a regul !! 1423 one of the underlying dentries hidden by the overlay. It is
1423 !! 1424 used in two different modes:
1424 The 'type' argument takes the values !! 1425
1425 for returning the real underlying den !! 1426 Called from file_dentry() it returns the real dentry matching
1426 hosting the file's data or metadata r !! 1427 the inode argument. The real dentry may be from a lower layer
>> 1428 already copied up, but still referenced from the file. This
>> 1429 mode is selected with a non-NULL inode argument.
1427 1430
1428 For non-regular files, the 'dentry' a !! 1431 With NULL inode the topmost real underlying dentry is returned.
1429 1432
1430 Each dentry has a pointer to its parent dentr 1433 Each dentry has a pointer to its parent dentry, as well as a hash list
1431 of child dentries. Child dentries are basica 1434 of child dentries. Child dentries are basically like files in a
1432 directory. 1435 directory.
1433 1436
1434 1437
1435 Directory Entry Cache API 1438 Directory Entry Cache API
1436 -------------------------- 1439 --------------------------
1437 1440
1438 There are a number of functions defined which 1441 There are a number of functions defined which permit a filesystem to
1439 manipulate dentries: 1442 manipulate dentries:
1440 1443
1441 ``dget`` 1444 ``dget``
1442 open a new handle for an existing den 1445 open a new handle for an existing dentry (this just increments
1443 the usage count) 1446 the usage count)
1444 1447
1445 ``dput`` 1448 ``dput``
1446 close a handle for a dentry (decremen 1449 close a handle for a dentry (decrements the usage count). If
1447 the usage count drops to 0, and the d 1450 the usage count drops to 0, and the dentry is still in its
1448 parent's hash, the "d_delete" method 1451 parent's hash, the "d_delete" method is called to check whether
1449 it should be cached. If it should no 1452 it should be cached. If it should not be cached, or if the
1450 dentry is not hashed, it is deleted. 1453 dentry is not hashed, it is deleted. Otherwise cached dentries
1451 are put into an LRU list to be reclai 1454 are put into an LRU list to be reclaimed on memory shortage.
1452 1455
1453 ``d_drop`` 1456 ``d_drop``
1454 this unhashes a dentry from its paren 1457 this unhashes a dentry from its parents hash list. A subsequent
1455 call to dput() will deallocate the de 1458 call to dput() will deallocate the dentry if its usage count
1456 drops to 0 1459 drops to 0
1457 1460
1458 ``d_delete`` 1461 ``d_delete``
1459 delete a dentry. If there are no oth 1462 delete a dentry. If there are no other open references to the
1460 dentry then the dentry is turned into 1463 dentry then the dentry is turned into a negative dentry (the
1461 d_iput() method is called). If there 1464 d_iput() method is called). If there are other references, then
1462 d_drop() is called instead 1465 d_drop() is called instead
1463 1466
1464 ``d_add`` 1467 ``d_add``
1465 add a dentry to its parents hash list 1468 add a dentry to its parents hash list and then calls
1466 d_instantiate() 1469 d_instantiate()
1467 1470
1468 ``d_instantiate`` 1471 ``d_instantiate``
1469 add a dentry to the alias hash list f 1472 add a dentry to the alias hash list for the inode and updates
1470 the "d_inode" member. The "i_count" 1473 the "d_inode" member. The "i_count" member in the inode
1471 structure should be set/incremented. 1474 structure should be set/incremented. If the inode pointer is
1472 NULL, the dentry is called a "negativ 1475 NULL, the dentry is called a "negative dentry". This function
1473 is commonly called when an inode is c 1476 is commonly called when an inode is created for an existing
1474 negative dentry 1477 negative dentry
1475 1478
1476 ``d_lookup`` 1479 ``d_lookup``
1477 look up a dentry given its parent and 1480 look up a dentry given its parent and path name component It
1478 looks up the child of that given name 1481 looks up the child of that given name from the dcache hash
1479 table. If it is found, the reference 1482 table. If it is found, the reference count is incremented and
1480 the dentry is returned. The caller m 1483 the dentry is returned. The caller must use dput() to free the
1481 dentry when it finishes using it. 1484 dentry when it finishes using it.
1482 1485
1483 1486
1484 Mount Options 1487 Mount Options
1485 ============= 1488 =============
1486 1489
1487 1490
1488 Parsing options 1491 Parsing options
1489 --------------- 1492 ---------------
1490 1493
1491 On mount and remount the filesystem is passed 1494 On mount and remount the filesystem is passed a string containing a
1492 comma separated list of mount options. The o 1495 comma separated list of mount options. The options can have either of
1493 these forms: 1496 these forms:
1494 1497
1495 option 1498 option
1496 option=value 1499 option=value
1497 1500
1498 The <linux/parser.h> header defines an API th 1501 The <linux/parser.h> header defines an API that helps parse these
1499 options. There are plenty of examples on how 1502 options. There are plenty of examples on how to use it in existing
1500 filesystems. 1503 filesystems.
1501 1504
1502 1505
1503 Showing options 1506 Showing options
1504 --------------- 1507 ---------------
1505 1508
1506 If a filesystem accepts mount options, it mus 1509 If a filesystem accepts mount options, it must define show_options() to
1507 show all the currently active options. The r 1510 show all the currently active options. The rules are:
1508 1511
1509 - options MUST be shown which are not defau 1512 - options MUST be shown which are not default or their values differ
1510 from the default 1513 from the default
1511 1514
1512 - options MAY be shown which are enabled by 1515 - options MAY be shown which are enabled by default or have their
1513 default value 1516 default value
1514 1517
1515 Options used only internally between a mount 1518 Options used only internally between a mount helper and the kernel (such
1516 as file descriptors), or which only have an e 1519 as file descriptors), or which only have an effect during the mounting
1517 (such as ones controlling the creation of a j 1520 (such as ones controlling the creation of a journal) are exempt from the
1518 above rules. 1521 above rules.
1519 1522
1520 The underlying reason for the above rules is 1523 The underlying reason for the above rules is to make sure, that a mount
1521 can be accurately replicated (e.g. umounting 1524 can be accurately replicated (e.g. umounting and mounting again) based
1522 on the information found in /proc/mounts. 1525 on the information found in /proc/mounts.
1523 1526
1524 1527
1525 Resources 1528 Resources
1526 ========= 1529 =========
1527 1530
1528 (Note some of these resources are not up-to-d 1531 (Note some of these resources are not up-to-date with the latest kernel
1529 version.) 1532 version.)
1530 1533
1531 Creating Linux virtual filesystems. 2002 1534 Creating Linux virtual filesystems. 2002
1532 <https://lwn.net/Articles/13325/> 1535 <https://lwn.net/Articles/13325/>
1533 1536
1534 The Linux Virtual File-system Layer by Neil B 1537 The Linux Virtual File-system Layer by Neil Brown. 1999
1535 <http://www.cse.unsw.edu.au/~neilb/oss/li 1538 <http://www.cse.unsw.edu.au/~neilb/oss/linux-commentary/vfs.html>
1536 1539
1537 A tour of the Linux VFS by Michael K. Johnson 1540 A tour of the Linux VFS by Michael K. Johnson. 1996
1538 <https://www.tldp.org/LDP/khg/HyperNews/g 1541 <https://www.tldp.org/LDP/khg/HyperNews/get/fs/vfstour.html>
1539 1542
1540 A small trail through the Linux kernel by And 1543 A small trail through the Linux kernel by Andries Brouwer. 2001
1541 <https://www.win.tue.nl/~aeb/linux/vfs/tr 1544 <https://www.win.tue.nl/~aeb/linux/vfs/trail.html>
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.