Commit | Line | Data |
---|---|---|
1da177e4 LT |
1 | The text below describes the locking rules for VFS-related methods. |
2 | It is (believed to be) up-to-date. *Please*, if you change anything in | |
3 | prototypes or locking protocols - update this file. And update the relevant | |
4 | instances in the tree, don't leave that to maintainers of filesystems/devices/ | |
5 | etc. At the very least, put the list of dubious cases in the end of this file. | |
6 | Don't turn it into log - maintainers of out-of-the-tree code are supposed to | |
7 | be able to use diff(1). | |
8 | Thing currently missing here: socket operations. Alexey? | |
9 | ||
10 | --------------------------- dentry_operations -------------------------- | |
11 | prototypes: | |
34286d66 | 12 | int (*d_revalidate)(struct dentry *, struct nameidata *); |
b1e6a015 NP |
13 | int (*d_hash)(const struct dentry *, const struct inode *, |
14 | struct qstr *); | |
621e155a NP |
15 | int (*d_compare)(const struct dentry *, const struct inode *, |
16 | const struct dentry *, const struct inode *, | |
17 | unsigned int, const char *, const struct qstr *); | |
1da177e4 LT |
18 | int (*d_delete)(struct dentry *); |
19 | void (*d_release)(struct dentry *); | |
20 | void (*d_iput)(struct dentry *, struct inode *); | |
c23fbb6b | 21 | char *(*d_dname)((struct dentry *dentry, char *buffer, int buflen); |
9875cf80 | 22 | struct vfsmount *(*d_automount)(struct path *path); |
cc53ce53 | 23 | int (*d_manage)(struct dentry *, bool); |
1da177e4 LT |
24 | |
25 | locking rules: | |
34286d66 NP |
26 | rename_lock ->d_lock may block rcu-walk |
27 | d_revalidate: no no yes (ref-walk) maybe | |
28 | d_hash no no no maybe | |
29 | d_compare: yes no no maybe | |
30 | d_delete: no yes no no | |
31 | d_release: no no yes no | |
32 | d_iput: no no yes no | |
33 | d_dname: no no no no | |
9875cf80 | 34 | d_automount: no no yes no |
ab90911f | 35 | d_manage: no no yes (ref-walk) maybe |
1da177e4 LT |
36 | |
37 | --------------------------- inode_operations --------------------------- | |
38 | prototypes: | |
39 | int (*create) (struct inode *,struct dentry *,int, struct nameidata *); | |
40 | struct dentry * (*lookup) (struct inode *,struct dentry *, struct nameid | |
41 | ata *); | |
42 | int (*link) (struct dentry *,struct inode *,struct dentry *); | |
43 | int (*unlink) (struct inode *,struct dentry *); | |
44 | int (*symlink) (struct inode *,struct dentry *,const char *); | |
45 | int (*mkdir) (struct inode *,struct dentry *,int); | |
46 | int (*rmdir) (struct inode *,struct dentry *); | |
47 | int (*mknod) (struct inode *,struct dentry *,int,dev_t); | |
48 | int (*rename) (struct inode *, struct dentry *, | |
49 | struct inode *, struct dentry *); | |
50 | int (*readlink) (struct dentry *, char __user *,int); | |
b83be6f2 CH |
51 | void * (*follow_link) (struct dentry *, struct nameidata *); |
52 | void (*put_link) (struct dentry *, struct nameidata *, void *); | |
1da177e4 | 53 | void (*truncate) (struct inode *); |
b74c79e9 NP |
54 | int (*permission) (struct inode *, int, unsigned int); |
55 | int (*check_acl)(struct inode *, int, unsigned int); | |
1da177e4 LT |
56 | int (*setattr) (struct dentry *, struct iattr *); |
57 | int (*getattr) (struct vfsmount *, struct dentry *, struct kstat *); | |
58 | int (*setxattr) (struct dentry *, const char *,const void *,size_t,int); | |
59 | ssize_t (*getxattr) (struct dentry *, const char *, void *, size_t); | |
60 | ssize_t (*listxattr) (struct dentry *, char *, size_t); | |
61 | int (*removexattr) (struct dentry *, const char *); | |
b83be6f2 | 62 | void (*truncate_range)(struct inode *, loff_t, loff_t); |
b83be6f2 | 63 | int (*fiemap)(struct inode *, struct fiemap_extent_info *, u64 start, u64 len); |
1da177e4 LT |
64 | |
65 | locking rules: | |
b83be6f2 | 66 | all may block |
a7bc02f4 | 67 | i_mutex(inode) |
1da177e4 LT |
68 | lookup: yes |
69 | create: yes | |
70 | link: yes (both) | |
71 | mknod: yes | |
72 | symlink: yes | |
73 | mkdir: yes | |
74 | unlink: yes (both) | |
75 | rmdir: yes (both) (see below) | |
76 | rename: yes (all) (see below) | |
77 | readlink: no | |
78 | follow_link: no | |
b83be6f2 | 79 | put_link: no |
1da177e4 LT |
80 | truncate: yes (see below) |
81 | setattr: yes | |
b74c79e9 | 82 | permission: no (may not block if called in rcu-walk mode) |
b83be6f2 | 83 | check_acl: no |
1da177e4 LT |
84 | getattr: no |
85 | setxattr: yes | |
86 | getxattr: no | |
87 | listxattr: no | |
88 | removexattr: yes | |
b83be6f2 | 89 | truncate_range: yes |
b83be6f2 | 90 | fiemap: no |
a7bc02f4 | 91 | Additionally, ->rmdir(), ->unlink() and ->rename() have ->i_mutex on |
1da177e4 LT |
92 | victim. |
93 | cross-directory ->rename() has (per-superblock) ->s_vfs_rename_sem. | |
94 | ->truncate() is never called directly - it's a callback, not a | |
b83be6f2 | 95 | method. It's called by vmtruncate() - deprecated library function used by |
1da177e4 LT |
96 | ->setattr(). Locking information above applies to that call (i.e. is |
97 | inherited from ->setattr() - vmtruncate() is used when ATTR_SIZE had been | |
98 | passed). | |
99 | ||
100 | See Documentation/filesystems/directory-locking for more detailed discussion | |
101 | of the locking scheme for directory operations. | |
102 | ||
103 | --------------------------- super_operations --------------------------- | |
104 | prototypes: | |
105 | struct inode *(*alloc_inode)(struct super_block *sb); | |
106 | void (*destroy_inode)(struct inode *); | |
1da177e4 | 107 | void (*dirty_inode) (struct inode *); |
b83be6f2 | 108 | int (*write_inode) (struct inode *, struct writeback_control *wbc); |
336fb3b9 AV |
109 | int (*drop_inode) (struct inode *); |
110 | void (*evict_inode) (struct inode *); | |
1da177e4 LT |
111 | void (*put_super) (struct super_block *); |
112 | void (*write_super) (struct super_block *); | |
113 | int (*sync_fs)(struct super_block *sb, int wait); | |
c4be0c1d TS |
114 | int (*freeze_fs) (struct super_block *); |
115 | int (*unfreeze_fs) (struct super_block *); | |
726c3342 | 116 | int (*statfs) (struct dentry *, struct kstatfs *); |
1da177e4 | 117 | int (*remount_fs) (struct super_block *, int *, char *); |
1da177e4 LT |
118 | void (*umount_begin) (struct super_block *); |
119 | int (*show_options)(struct seq_file *, struct vfsmount *); | |
120 | ssize_t (*quota_read)(struct super_block *, int, char *, size_t, loff_t); | |
121 | ssize_t (*quota_write)(struct super_block *, int, const char *, size_t, loff_t); | |
b83be6f2 | 122 | int (*bdev_try_to_free_page)(struct super_block*, struct page*, gfp_t); |
1da177e4 LT |
123 | |
124 | locking rules: | |
336fb3b9 | 125 | All may block [not true, see below] |
7e325d3a CH |
126 | s_umount |
127 | alloc_inode: | |
128 | destroy_inode: | |
129 | dirty_inode: (must not sleep) | |
130 | write_inode: | |
f283c86a | 131 | drop_inode: !!!inode->i_lock!!! |
336fb3b9 | 132 | evict_inode: |
7e325d3a CH |
133 | put_super: write |
134 | write_super: read | |
135 | sync_fs: read | |
136 | freeze_fs: read | |
137 | unfreeze_fs: read | |
336fb3b9 AV |
138 | statfs: maybe(read) (see below) |
139 | remount_fs: write | |
7e325d3a CH |
140 | umount_begin: no |
141 | show_options: no (namespace_sem) | |
142 | quota_read: no (see below) | |
143 | quota_write: no (see below) | |
b83be6f2 | 144 | bdev_try_to_free_page: no (see below) |
1da177e4 | 145 | |
336fb3b9 AV |
146 | ->statfs() has s_umount (shared) when called by ustat(2) (native or |
147 | compat), but that's an accident of bad API; s_umount is used to pin | |
148 | the superblock down when we only have dev_t given us by userland to | |
149 | identify the superblock. Everything else (statfs(), fstatfs(), etc.) | |
150 | doesn't hold it when calling ->statfs() - superblock is pinned down | |
151 | by resolving the pathname passed to syscall. | |
1da177e4 LT |
152 | ->quota_read() and ->quota_write() functions are both guaranteed to |
153 | be the only ones operating on the quota file by the quota code (via | |
154 | dqio_sem) (unless an admin really wants to screw up something and | |
155 | writes to quota files with quotas on). For other details about locking | |
156 | see also dquot_operations section. | |
b83be6f2 CH |
157 | ->bdev_try_to_free_page is called from the ->releasepage handler of |
158 | the block device inode. See there for more details. | |
1da177e4 LT |
159 | |
160 | --------------------------- file_system_type --------------------------- | |
161 | prototypes: | |
5d8b2ebf JC |
162 | int (*get_sb) (struct file_system_type *, int, |
163 | const char *, void *, struct vfsmount *); | |
b83be6f2 CH |
164 | struct dentry *(*mount) (struct file_system_type *, int, |
165 | const char *, void *); | |
1da177e4 LT |
166 | void (*kill_sb) (struct super_block *); |
167 | locking rules: | |
b83be6f2 | 168 | may block |
b83be6f2 CH |
169 | mount yes |
170 | kill_sb yes | |
1da177e4 | 171 | |
1a102ff9 AV |
172 | ->mount() returns ERR_PTR or the root dentry; its superblock should be locked |
173 | on return. | |
1da177e4 LT |
174 | ->kill_sb() takes a write-locked superblock, does all shutdown work on it, |
175 | unlocks and drops the reference. | |
176 | ||
177 | --------------------------- address_space_operations -------------------------- | |
178 | prototypes: | |
179 | int (*writepage)(struct page *page, struct writeback_control *wbc); | |
180 | int (*readpage)(struct file *, struct page *); | |
181 | int (*sync_page)(struct page *); | |
182 | int (*writepages)(struct address_space *, struct writeback_control *); | |
183 | int (*set_page_dirty)(struct page *page); | |
184 | int (*readpages)(struct file *filp, struct address_space *mapping, | |
185 | struct list_head *pages, unsigned nr_pages); | |
4e02ed4b NP |
186 | int (*write_begin)(struct file *, struct address_space *mapping, |
187 | loff_t pos, unsigned len, unsigned flags, | |
188 | struct page **pagep, void **fsdata); | |
189 | int (*write_end)(struct file *, struct address_space *mapping, | |
190 | loff_t pos, unsigned len, unsigned copied, | |
191 | struct page *page, void *fsdata); | |
1da177e4 LT |
192 | sector_t (*bmap)(struct address_space *, sector_t); |
193 | int (*invalidatepage) (struct page *, unsigned long); | |
194 | int (*releasepage) (struct page *, int); | |
6072d13c | 195 | void (*freepage)(struct page *); |
1da177e4 LT |
196 | int (*direct_IO)(int, struct kiocb *, const struct iovec *iov, |
197 | loff_t offset, unsigned long nr_segs); | |
b83be6f2 CH |
198 | int (*get_xip_mem)(struct address_space *, pgoff_t, int, void **, |
199 | unsigned long *); | |
200 | int (*migratepage)(struct address_space *, struct page *, struct page *); | |
201 | int (*launder_page)(struct page *); | |
202 | int (*is_partially_uptodate)(struct page *, read_descriptor_t *, unsigned long); | |
203 | int (*error_remove_page)(struct address_space *, struct page *); | |
1da177e4 LT |
204 | |
205 | locking rules: | |
6072d13c | 206 | All except set_page_dirty and freepage may block |
1da177e4 | 207 | |
b83be6f2 CH |
208 | PageLocked(page) i_mutex |
209 | writepage: yes, unlocks (see below) | |
210 | readpage: yes, unlocks | |
211 | sync_page: maybe | |
212 | writepages: | |
213 | set_page_dirty no | |
214 | readpages: | |
215 | write_begin: locks the page yes | |
216 | write_end: yes, unlocks yes | |
217 | bmap: | |
218 | invalidatepage: yes | |
219 | releasepage: yes | |
220 | freepage: yes | |
221 | direct_IO: | |
222 | get_xip_mem: maybe | |
223 | migratepage: yes (both) | |
224 | launder_page: yes | |
225 | is_partially_uptodate: yes | |
226 | error_remove_page: yes | |
1da177e4 | 227 | |
4e02ed4b | 228 | ->write_begin(), ->write_end(), ->sync_page() and ->readpage() |
1da177e4 LT |
229 | may be called from the request handler (/dev/loop). |
230 | ||
231 | ->readpage() unlocks the page, either synchronously or via I/O | |
232 | completion. | |
233 | ||
234 | ->readpages() populates the pagecache with the passed pages and starts | |
235 | I/O against them. They come unlocked upon I/O completion. | |
236 | ||
237 | ->writepage() is used for two purposes: for "memory cleansing" and for | |
238 | "sync". These are quite different operations and the behaviour may differ | |
239 | depending upon the mode. | |
240 | ||
241 | If writepage is called for sync (wbc->sync_mode != WBC_SYNC_NONE) then | |
242 | it *must* start I/O against the page, even if that would involve | |
243 | blocking on in-progress I/O. | |
244 | ||
245 | If writepage is called for memory cleansing (sync_mode == | |
246 | WBC_SYNC_NONE) then its role is to get as much writeout underway as | |
247 | possible. So writepage should try to avoid blocking against | |
248 | currently-in-progress I/O. | |
249 | ||
250 | If the filesystem is not called for "sync" and it determines that it | |
251 | would need to block against in-progress I/O to be able to start new I/O | |
252 | against the page the filesystem should redirty the page with | |
253 | redirty_page_for_writepage(), then unlock the page and return zero. | |
254 | This may also be done to avoid internal deadlocks, but rarely. | |
255 | ||
3a4fa0a2 | 256 | If the filesystem is called for sync then it must wait on any |
1da177e4 LT |
257 | in-progress I/O and then start new I/O. |
258 | ||
2054606a ND |
259 | The filesystem should unlock the page synchronously, before returning to the |
260 | caller, unless ->writepage() returns special WRITEPAGE_ACTIVATE | |
261 | value. WRITEPAGE_ACTIVATE means that page cannot really be written out | |
262 | currently, and VM should stop calling ->writepage() on this page for some | |
263 | time. VM does this by moving page to the head of the active list, hence the | |
264 | name. | |
1da177e4 LT |
265 | |
266 | Unless the filesystem is going to redirty_page_for_writepage(), unlock the page | |
267 | and return zero, writepage *must* run set_page_writeback() against the page, | |
268 | followed by unlocking it. Once set_page_writeback() has been run against the | |
269 | page, write I/O can be submitted and the write I/O completion handler must run | |
270 | end_page_writeback() once the I/O is complete. If no I/O is submitted, the | |
271 | filesystem must run end_page_writeback() against the page before returning from | |
272 | writepage. | |
273 | ||
274 | That is: after 2.5.12, pages which are under writeout are *not* locked. Note, | |
275 | if the filesystem needs the page to be locked during writeout, that is ok, too, | |
276 | the page is allowed to be unlocked at any point in time between the calls to | |
277 | set_page_writeback() and end_page_writeback(). | |
278 | ||
279 | Note, failure to run either redirty_page_for_writepage() or the combination of | |
280 | set_page_writeback()/end_page_writeback() on a page submitted to writepage | |
281 | will leave the page itself marked clean but it will be tagged as dirty in the | |
282 | radix tree. This incoherency can lead to all sorts of hard-to-debug problems | |
283 | in the filesystem like having dirty inodes at umount and losing written data. | |
284 | ||
285 | ->sync_page() locking rules are not well-defined - usually it is called | |
286 | with lock on page, but that is not guaranteed. Considering the currently | |
287 | existing instances of this method ->sync_page() itself doesn't look | |
288 | well-defined... | |
289 | ||
290 | ->writepages() is used for periodic writeback and for syscall-initiated | |
291 | sync operations. The address_space should start I/O against at least | |
292 | *nr_to_write pages. *nr_to_write must be decremented for each page which is | |
293 | written. The address_space implementation may write more (or less) pages | |
294 | than *nr_to_write asks for, but it should try to be reasonably close. If | |
295 | nr_to_write is NULL, all dirty pages must be written. | |
296 | ||
297 | writepages should _only_ write pages which are present on | |
298 | mapping->io_pages. | |
299 | ||
300 | ->set_page_dirty() is called from various places in the kernel | |
301 | when the target page is marked as needing writeback. It may be called | |
302 | under spinlock (it cannot block) and is sometimes called with the page | |
303 | not locked. | |
304 | ||
305 | ->bmap() is currently used by legacy ioctl() (FIBMAP) provided by some | |
b83be6f2 CH |
306 | filesystems and by the swapper. The latter will eventually go away. Please, |
307 | keep it that way and don't breed new callers. | |
1da177e4 LT |
308 | |
309 | ->invalidatepage() is called when the filesystem must attempt to drop | |
310 | some or all of the buffers from the page when it is being truncated. It | |
311 | returns zero on success. If ->invalidatepage is zero, the kernel uses | |
312 | block_invalidatepage() instead. | |
313 | ||
314 | ->releasepage() is called when the kernel is about to try to drop the | |
315 | buffers from the page in preparation for freeing it. It returns zero to | |
316 | indicate that the buffers are (or may be) freeable. If ->releasepage is zero, | |
317 | the kernel assumes that the fs has no private interest in the buffers. | |
318 | ||
6072d13c LT |
319 | ->freepage() is called when the kernel is done dropping the page |
320 | from the page cache. | |
321 | ||
e3db7691 TM |
322 | ->launder_page() may be called prior to releasing a page if |
323 | it is still found to be dirty. It returns zero if the page was successfully | |
324 | cleaned, or an error value if not. Note that in order to prevent the page | |
325 | getting mapped back in and redirtied, it needs to be kept locked | |
326 | across the entire operation. | |
327 | ||
1da177e4 LT |
328 | ----------------------- file_lock_operations ------------------------------ |
329 | prototypes: | |
1da177e4 LT |
330 | void (*fl_copy_lock)(struct file_lock *, struct file_lock *); |
331 | void (*fl_release_private)(struct file_lock *); | |
332 | ||
333 | ||
334 | locking rules: | |
b83be6f2 CH |
335 | file_lock_lock may block |
336 | fl_copy_lock: yes no | |
337 | fl_release_private: maybe no | |
1da177e4 LT |
338 | |
339 | ----------------------- lock_manager_operations --------------------------- | |
340 | prototypes: | |
341 | int (*fl_compare_owner)(struct file_lock *, struct file_lock *); | |
342 | void (*fl_notify)(struct file_lock *); /* unblock callback */ | |
b83be6f2 | 343 | int (*fl_grant)(struct file_lock *, struct file_lock *, int); |
1da177e4 LT |
344 | void (*fl_release_private)(struct file_lock *); |
345 | void (*fl_break)(struct file_lock *); /* break_lease callback */ | |
b83be6f2 | 346 | int (*fl_change)(struct file_lock **, int); |
1da177e4 LT |
347 | |
348 | locking rules: | |
b83be6f2 CH |
349 | file_lock_lock may block |
350 | fl_compare_owner: yes no | |
351 | fl_notify: yes no | |
352 | fl_grant: no no | |
353 | fl_release_private: maybe no | |
354 | fl_break: yes no | |
b83be6f2 CH |
355 | fl_change yes no |
356 | ||
1da177e4 LT |
357 | --------------------------- buffer_head ----------------------------------- |
358 | prototypes: | |
359 | void (*b_end_io)(struct buffer_head *bh, int uptodate); | |
360 | ||
361 | locking rules: | |
362 | called from interrupts. In other words, extreme care is needed here. | |
363 | bh is locked, but that's all warranties we have here. Currently only RAID1, | |
364 | highmem, fs/buffer.c, and fs/ntfs/aops.c are providing these. Block devices | |
365 | call this method upon the IO completion. | |
366 | ||
367 | --------------------------- block_device_operations ----------------------- | |
368 | prototypes: | |
e1455d1b CH |
369 | int (*open) (struct block_device *, fmode_t); |
370 | int (*release) (struct gendisk *, fmode_t); | |
371 | int (*ioctl) (struct block_device *, fmode_t, unsigned, unsigned long); | |
372 | int (*compat_ioctl) (struct block_device *, fmode_t, unsigned, unsigned long); | |
373 | int (*direct_access) (struct block_device *, sector_t, void **, unsigned long *); | |
1da177e4 | 374 | int (*media_changed) (struct gendisk *); |
e1455d1b | 375 | void (*unlock_native_capacity) (struct gendisk *); |
1da177e4 | 376 | int (*revalidate_disk) (struct gendisk *); |
e1455d1b CH |
377 | int (*getgeo)(struct block_device *, struct hd_geometry *); |
378 | void (*swap_slot_free_notify) (struct block_device *, unsigned long); | |
1da177e4 LT |
379 | |
380 | locking rules: | |
b83be6f2 CH |
381 | bd_mutex |
382 | open: yes | |
383 | release: yes | |
384 | ioctl: no | |
385 | compat_ioctl: no | |
386 | direct_access: no | |
387 | media_changed: no | |
388 | unlock_native_capacity: no | |
389 | revalidate_disk: no | |
390 | getgeo: no | |
391 | swap_slot_free_notify: no (see below) | |
e1455d1b CH |
392 | |
393 | media_changed, unlock_native_capacity and revalidate_disk are called only from | |
394 | check_disk_change(). | |
395 | ||
396 | swap_slot_free_notify is called with swap_lock and sometimes the page lock | |
397 | held. | |
1da177e4 | 398 | |
1da177e4 LT |
399 | |
400 | --------------------------- file_operations ------------------------------- | |
401 | prototypes: | |
402 | loff_t (*llseek) (struct file *, loff_t, int); | |
403 | ssize_t (*read) (struct file *, char __user *, size_t, loff_t *); | |
1da177e4 | 404 | ssize_t (*write) (struct file *, const char __user *, size_t, loff_t *); |
027445c3 BP |
405 | ssize_t (*aio_read) (struct kiocb *, const struct iovec *, unsigned long, loff_t); |
406 | ssize_t (*aio_write) (struct kiocb *, const struct iovec *, unsigned long, loff_t); | |
1da177e4 LT |
407 | int (*readdir) (struct file *, void *, filldir_t); |
408 | unsigned int (*poll) (struct file *, struct poll_table_struct *); | |
1da177e4 LT |
409 | long (*unlocked_ioctl) (struct file *, unsigned int, unsigned long); |
410 | long (*compat_ioctl) (struct file *, unsigned int, unsigned long); | |
411 | int (*mmap) (struct file *, struct vm_area_struct *); | |
412 | int (*open) (struct inode *, struct file *); | |
413 | int (*flush) (struct file *); | |
414 | int (*release) (struct inode *, struct file *); | |
7ea80859 | 415 | int (*fsync) (struct file *, int datasync); |
1da177e4 LT |
416 | int (*aio_fsync) (struct kiocb *, int datasync); |
417 | int (*fasync) (int, struct file *, int); | |
418 | int (*lock) (struct file *, int, struct file_lock *); | |
419 | ssize_t (*readv) (struct file *, const struct iovec *, unsigned long, | |
420 | loff_t *); | |
421 | ssize_t (*writev) (struct file *, const struct iovec *, unsigned long, | |
422 | loff_t *); | |
423 | ssize_t (*sendfile) (struct file *, loff_t *, size_t, read_actor_t, | |
424 | void __user *); | |
425 | ssize_t (*sendpage) (struct file *, struct page *, int, size_t, | |
426 | loff_t *, int); | |
427 | unsigned long (*get_unmapped_area)(struct file *, unsigned long, | |
428 | unsigned long, unsigned long, unsigned long); | |
429 | int (*check_flags)(int); | |
b83be6f2 CH |
430 | int (*flock) (struct file *, int, struct file_lock *); |
431 | ssize_t (*splice_write)(struct pipe_inode_info *, struct file *, loff_t *, | |
432 | size_t, unsigned int); | |
433 | ssize_t (*splice_read)(struct file *, loff_t *, struct pipe_inode_info *, | |
434 | size_t, unsigned int); | |
435 | int (*setlease)(struct file *, long, struct file_lock **); | |
2fe17c10 | 436 | long (*fallocate)(struct file *, int, loff_t, loff_t); |
1da177e4 LT |
437 | }; |
438 | ||
439 | locking rules: | |
b83be6f2 CH |
440 | All may block except for ->setlease. |
441 | No VFS locks held on entry except for ->fsync and ->setlease. | |
442 | ||
443 | ->fsync() has i_mutex on inode. | |
444 | ||
445 | ->setlease has the file_list_lock held and must not sleep. | |
1da177e4 LT |
446 | |
447 | ->llseek() locking has moved from llseek to the individual llseek | |
448 | implementations. If your fs is not using generic_file_llseek, you | |
449 | need to acquire and release the appropriate locks in your ->llseek(). | |
450 | For many filesystems, it is probably safe to acquire the inode | |
866707fc JB |
451 | mutex or just to use i_size_read() instead. |
452 | Note: this does not protect the file->f_pos against concurrent modifications | |
453 | since this is something the userspace has to take care about. | |
1da177e4 | 454 | |
b83be6f2 CH |
455 | ->fasync() is responsible for maintaining the FASYNC bit in filp->f_flags. |
456 | Most instances call fasync_helper(), which does that maintenance, so it's | |
457 | not normally something one needs to worry about. Return values > 0 will be | |
458 | mapped to zero in the VFS layer. | |
1da177e4 LT |
459 | |
460 | ->readdir() and ->ioctl() on directories must be changed. Ideally we would | |
461 | move ->readdir() to inode_operations and use a separate method for directory | |
462 | ->ioctl() or kill the latter completely. One of the problems is that for | |
463 | anything that resembles union-mount we won't have a struct file for all | |
464 | components. And there are other reasons why the current interface is a mess... | |
465 | ||
1da177e4 LT |
466 | ->read on directories probably must go away - we should just enforce -EISDIR |
467 | in sys_read() and friends. | |
468 | ||
1da177e4 LT |
469 | --------------------------- dquot_operations ------------------------------- |
470 | prototypes: | |
1da177e4 LT |
471 | int (*write_dquot) (struct dquot *); |
472 | int (*acquire_dquot) (struct dquot *); | |
473 | int (*release_dquot) (struct dquot *); | |
474 | int (*mark_dirty) (struct dquot *); | |
475 | int (*write_info) (struct super_block *, int); | |
476 | ||
477 | These operations are intended to be more or less wrapping functions that ensure | |
478 | a proper locking wrt the filesystem and call the generic quota operations. | |
479 | ||
480 | What filesystem should expect from the generic quota functions: | |
481 | ||
482 | FS recursion Held locks when called | |
1da177e4 LT |
483 | write_dquot: yes dqonoff_sem or dqptr_sem |
484 | acquire_dquot: yes dqonoff_sem or dqptr_sem | |
485 | release_dquot: yes dqonoff_sem or dqptr_sem | |
486 | mark_dirty: no - | |
487 | write_info: yes dqonoff_sem | |
488 | ||
489 | FS recursion means calling ->quota_read() and ->quota_write() from superblock | |
490 | operations. | |
491 | ||
1da177e4 LT |
492 | More details about quota locking can be found in fs/dquot.c. |
493 | ||
494 | --------------------------- vm_operations_struct ----------------------------- | |
495 | prototypes: | |
496 | void (*open)(struct vm_area_struct*); | |
497 | void (*close)(struct vm_area_struct*); | |
d0217ac0 | 498 | int (*fault)(struct vm_area_struct*, struct vm_fault *); |
c2ec175c | 499 | int (*page_mkwrite)(struct vm_area_struct *, struct vm_fault *); |
28b2ee20 | 500 | int (*access)(struct vm_area_struct *, unsigned long, void*, int, int); |
1da177e4 LT |
501 | |
502 | locking rules: | |
b83be6f2 CH |
503 | mmap_sem PageLocked(page) |
504 | open: yes | |
505 | close: yes | |
506 | fault: yes can return with page locked | |
507 | page_mkwrite: yes can return with page locked | |
508 | access: yes | |
ed2f2f9b | 509 | |
b827e496 NP |
510 | ->fault() is called when a previously not present pte is about |
511 | to be faulted in. The filesystem must find and return the page associated | |
512 | with the passed in "pgoff" in the vm_fault structure. If it is possible that | |
513 | the page may be truncated and/or invalidated, then the filesystem must lock | |
514 | the page, then ensure it is not already truncated (the page lock will block | |
515 | subsequent truncate), and then return with VM_FAULT_LOCKED, and the page | |
516 | locked. The VM will unlock the page. | |
517 | ||
518 | ->page_mkwrite() is called when a previously read-only pte is | |
519 | about to become writeable. The filesystem again must ensure that there are | |
520 | no truncate/invalidate races, and then return with the page locked. If | |
521 | the page has been truncated, the filesystem should not look up a new page | |
522 | like the ->fault() handler, but simply return with VM_FAULT_NOPAGE, which | |
523 | will cause the VM to retry the fault. | |
1da177e4 | 524 | |
28b2ee20 RR |
525 | ->access() is called when get_user_pages() fails in |
526 | acces_process_vm(), typically used to debug a process through | |
527 | /proc/pid/mem or ptrace. This function is needed only for | |
528 | VM_IO | VM_PFNMAP VMAs. | |
529 | ||
1da177e4 LT |
530 | ================================================================================ |
531 | Dubious stuff | |
532 | ||
533 | (if you break something or notice that it is broken and do not fix it yourself | |
534 | - at least put it here) |