blob: 76efe5b71d7d8f744739a129f4b5333b898acf5c [file] [log] [blame]
Linus Torvalds1da177e2005-04-16 15:20:36 -07001 The text below describes the locking rules for VFS-related methods.
2It is (believed to be) up-to-date. *Please*, if you change anything in
3prototypes or locking protocols - update this file. And update the relevant
4instances in the tree, don't leave that to maintainers of filesystems/devices/
5etc. At the very least, put the list of dubious cases in the end of this file.
6Don't turn it into log - maintainers of out-of-the-tree code are supposed to
7be able to use diff(1).
8 Thing currently missing here: socket operations. Alexey?
9
10--------------------------- dentry_operations --------------------------
11prototypes:
12 int (*d_revalidate)(struct dentry *, int);
13 int (*d_hash) (struct dentry *, struct qstr *);
14 int (*d_compare) (struct dentry *, struct qstr *, struct qstr *);
15 int (*d_delete)(struct dentry *);
16 void (*d_release)(struct dentry *);
17 void (*d_iput)(struct dentry *, struct inode *);
Eric Dumazetc23fbb62007-05-08 00:26:18 -070018 char *(*d_dname)((struct dentry *dentry, char *buffer, int buflen);
Linus Torvalds1da177e2005-04-16 15:20:36 -070019
20locking rules:
21 none have BKL
22 dcache_lock rename_lock ->d_lock may block
23d_revalidate: no no no yes
24d_hash no no no yes
25d_compare: no yes no no
26d_delete: yes no yes no
27d_release: no no no yes
28d_iput: no no no yes
Eric Dumazetc23fbb62007-05-08 00:26:18 -070029d_dname: no no no no
Linus Torvalds1da177e2005-04-16 15:20:36 -070030
31--------------------------- inode_operations ---------------------------
32prototypes:
33 int (*create) (struct inode *,struct dentry *,int, struct nameidata *);
34 struct dentry * (*lookup) (struct inode *,struct dentry *, struct nameid
35ata *);
36 int (*link) (struct dentry *,struct inode *,struct dentry *);
37 int (*unlink) (struct inode *,struct dentry *);
38 int (*symlink) (struct inode *,struct dentry *,const char *);
39 int (*mkdir) (struct inode *,struct dentry *,int);
40 int (*rmdir) (struct inode *,struct dentry *);
41 int (*mknod) (struct inode *,struct dentry *,int,dev_t);
42 int (*rename) (struct inode *, struct dentry *,
43 struct inode *, struct dentry *);
44 int (*readlink) (struct dentry *, char __user *,int);
45 int (*follow_link) (struct dentry *, struct nameidata *);
46 void (*truncate) (struct inode *);
47 int (*permission) (struct inode *, int, struct nameidata *);
48 int (*setattr) (struct dentry *, struct iattr *);
49 int (*getattr) (struct vfsmount *, struct dentry *, struct kstat *);
50 int (*setxattr) (struct dentry *, const char *,const void *,size_t,int);
51 ssize_t (*getxattr) (struct dentry *, const char *, void *, size_t);
52 ssize_t (*listxattr) (struct dentry *, char *, size_t);
53 int (*removexattr) (struct dentry *, const char *);
54
55locking rules:
56 all may block, none have BKL
Artem Bityutskiya7bc02f2007-05-09 07:53:16 +020057 i_mutex(inode)
Linus Torvalds1da177e2005-04-16 15:20:36 -070058lookup: yes
59create: yes
60link: yes (both)
61mknod: yes
62symlink: yes
63mkdir: yes
64unlink: yes (both)
65rmdir: yes (both) (see below)
66rename: yes (all) (see below)
67readlink: no
68follow_link: no
69truncate: yes (see below)
70setattr: yes
71permission: no
72getattr: no
73setxattr: yes
74getxattr: no
75listxattr: no
76removexattr: yes
Artem Bityutskiya7bc02f2007-05-09 07:53:16 +020077 Additionally, ->rmdir(), ->unlink() and ->rename() have ->i_mutex on
Linus Torvalds1da177e2005-04-16 15:20:36 -070078victim.
79 cross-directory ->rename() has (per-superblock) ->s_vfs_rename_sem.
80 ->truncate() is never called directly - it's a callback, not a
81method. It's called by vmtruncate() - library function normally used by
82->setattr(). Locking information above applies to that call (i.e. is
83inherited from ->setattr() - vmtruncate() is used when ATTR_SIZE had been
84passed).
85
86See Documentation/filesystems/directory-locking for more detailed discussion
87of the locking scheme for directory operations.
88
89--------------------------- super_operations ---------------------------
90prototypes:
91 struct inode *(*alloc_inode)(struct super_block *sb);
92 void (*destroy_inode)(struct inode *);
Linus Torvalds1da177e2005-04-16 15:20:36 -070093 void (*dirty_inode) (struct inode *);
94 int (*write_inode) (struct inode *, int);
Linus Torvalds1da177e2005-04-16 15:20:36 -070095 void (*drop_inode) (struct inode *);
96 void (*delete_inode) (struct inode *);
97 void (*put_super) (struct super_block *);
98 void (*write_super) (struct super_block *);
99 int (*sync_fs)(struct super_block *sb, int wait);
Takashi Satoc4be0c12009-01-09 16:40:58 -0800100 int (*freeze_fs) (struct super_block *);
101 int (*unfreeze_fs) (struct super_block *);
David Howells726c3342006-06-23 02:02:58 -0700102 int (*statfs) (struct dentry *, struct kstatfs *);
Linus Torvalds1da177e2005-04-16 15:20:36 -0700103 int (*remount_fs) (struct super_block *, int *, char *);
104 void (*clear_inode) (struct inode *);
105 void (*umount_begin) (struct super_block *);
106 int (*show_options)(struct seq_file *, struct vfsmount *);
107 ssize_t (*quota_read)(struct super_block *, int, char *, size_t, loff_t);
108 ssize_t (*quota_write)(struct super_block *, int, const char *, size_t, loff_t);
109
110locking rules:
111 All may block.
112 BKL s_lock s_umount
113alloc_inode: no no no
114destroy_inode: no
Linus Torvalds1da177e2005-04-16 15:20:36 -0700115dirty_inode: no (must not sleep)
116write_inode: no
Linus Torvalds1da177e2005-04-16 15:20:36 -0700117drop_inode: no !!!inode_lock!!!
118delete_inode: no
119put_super: yes yes no
120write_super: no yes read
121sync_fs: no no read
Takashi Satoc4be0c12009-01-09 16:40:58 -0800122freeze_fs: ?
123unfreeze_fs: ?
Linus Torvalds1da177e2005-04-16 15:20:36 -0700124statfs: no no no
Vasily Averin70888bd2006-12-06 20:37:23 -0800125remount_fs: yes yes maybe (see below)
Linus Torvalds1da177e2005-04-16 15:20:36 -0700126clear_inode: no
127umount_begin: yes no no
128show_options: no (vfsmount->sem)
129quota_read: no no no (see below)
130quota_write: no no no (see below)
131
Linus Torvalds1da177e2005-04-16 15:20:36 -0700132->remount_fs() will have the s_umount lock if it's already mounted.
133When called from get_sb_single, it does NOT have the s_umount lock.
134->quota_read() and ->quota_write() functions are both guaranteed to
135be the only ones operating on the quota file by the quota code (via
136dqio_sem) (unless an admin really wants to screw up something and
137writes to quota files with quotas on). For other details about locking
138see also dquot_operations section.
139
140--------------------------- file_system_type ---------------------------
141prototypes:
Jonathan Corbet5d8b2eb2006-07-10 04:44:07 -0700142 int (*get_sb) (struct file_system_type *, int,
143 const char *, void *, struct vfsmount *);
Linus Torvalds1da177e2005-04-16 15:20:36 -0700144 void (*kill_sb) (struct super_block *);
145locking rules:
146 may block BKL
Christoph Hellwigadaae722008-09-09 20:02:01 +0200147get_sb yes no
148kill_sb yes no
Linus Torvalds1da177e2005-04-16 15:20:36 -0700149
David Howells454e2392006-06-23 02:02:57 -0700150->get_sb() returns error or 0 with locked superblock attached to the vfsmount
151(exclusive on ->s_umount).
Linus Torvalds1da177e2005-04-16 15:20:36 -0700152->kill_sb() takes a write-locked superblock, does all shutdown work on it,
153unlocks and drops the reference.
154
155--------------------------- address_space_operations --------------------------
156prototypes:
157 int (*writepage)(struct page *page, struct writeback_control *wbc);
158 int (*readpage)(struct file *, struct page *);
159 int (*sync_page)(struct page *);
160 int (*writepages)(struct address_space *, struct writeback_control *);
161 int (*set_page_dirty)(struct page *page);
162 int (*readpages)(struct file *filp, struct address_space *mapping,
163 struct list_head *pages, unsigned nr_pages);
Nick Piggin4e02ed42008-10-29 14:00:55 -0700164 int (*write_begin)(struct file *, struct address_space *mapping,
165 loff_t pos, unsigned len, unsigned flags,
166 struct page **pagep, void **fsdata);
167 int (*write_end)(struct file *, struct address_space *mapping,
168 loff_t pos, unsigned len, unsigned copied,
169 struct page *page, void *fsdata);
Linus Torvalds1da177e2005-04-16 15:20:36 -0700170 sector_t (*bmap)(struct address_space *, sector_t);
171 int (*invalidatepage) (struct page *, unsigned long);
172 int (*releasepage) (struct page *, int);
173 int (*direct_IO)(int, struct kiocb *, const struct iovec *iov,
174 loff_t offset, unsigned long nr_segs);
Trond Myklebuste3db7692007-01-10 23:15:39 -0800175 int (*launder_page) (struct page *);
Linus Torvalds1da177e2005-04-16 15:20:36 -0700176
177locking rules:
178 All except set_page_dirty may block
179
Nick Pigginafddba42007-10-16 01:25:01 -0700180 BKL PageLocked(page) i_sem
Linus Torvalds1da177e2005-04-16 15:20:36 -0700181writepage: no yes, unlocks (see below)
182readpage: no yes, unlocks
183sync_page: no maybe
184writepages: no
185set_page_dirty no no
186readpages: no
Nick Pigginafddba42007-10-16 01:25:01 -0700187write_begin: no locks the page yes
188write_end: no yes, unlocks yes
189perform_write: no n/a yes
Linus Torvalds1da177e2005-04-16 15:20:36 -0700190bmap: yes
191invalidatepage: no yes
192releasepage: no yes
193direct_IO: no
Trond Myklebuste3db7692007-01-10 23:15:39 -0800194launder_page: no yes
Linus Torvalds1da177e2005-04-16 15:20:36 -0700195
Nick Piggin4e02ed42008-10-29 14:00:55 -0700196 ->write_begin(), ->write_end(), ->sync_page() and ->readpage()
Linus Torvalds1da177e2005-04-16 15:20:36 -0700197may be called from the request handler (/dev/loop).
198
199 ->readpage() unlocks the page, either synchronously or via I/O
200completion.
201
202 ->readpages() populates the pagecache with the passed pages and starts
203I/O against them. They come unlocked upon I/O completion.
204
205 ->writepage() is used for two purposes: for "memory cleansing" and for
206"sync". These are quite different operations and the behaviour may differ
207depending upon the mode.
208
209If writepage is called for sync (wbc->sync_mode != WBC_SYNC_NONE) then
210it *must* start I/O against the page, even if that would involve
211blocking on in-progress I/O.
212
213If writepage is called for memory cleansing (sync_mode ==
214WBC_SYNC_NONE) then its role is to get as much writeout underway as
215possible. So writepage should try to avoid blocking against
216currently-in-progress I/O.
217
218If the filesystem is not called for "sync" and it determines that it
219would need to block against in-progress I/O to be able to start new I/O
220against the page the filesystem should redirty the page with
221redirty_page_for_writepage(), then unlock the page and return zero.
222This may also be done to avoid internal deadlocks, but rarely.
223
Robert P. J. Day3a4fa0a2007-10-19 23:10:43 +0200224If the filesystem is called for sync then it must wait on any
Linus Torvalds1da177e2005-04-16 15:20:36 -0700225in-progress I/O and then start new I/O.
226
Nikita Danilov20546062005-05-01 08:58:37 -0700227The filesystem should unlock the page synchronously, before returning to the
228caller, unless ->writepage() returns special WRITEPAGE_ACTIVATE
229value. WRITEPAGE_ACTIVATE means that page cannot really be written out
230currently, and VM should stop calling ->writepage() on this page for some
231time. VM does this by moving page to the head of the active list, hence the
232name.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700233
234Unless the filesystem is going to redirty_page_for_writepage(), unlock the page
235and return zero, writepage *must* run set_page_writeback() against the page,
236followed by unlocking it. Once set_page_writeback() has been run against the
237page, write I/O can be submitted and the write I/O completion handler must run
238end_page_writeback() once the I/O is complete. If no I/O is submitted, the
239filesystem must run end_page_writeback() against the page before returning from
240writepage.
241
242That is: after 2.5.12, pages which are under writeout are *not* locked. Note,
243if the filesystem needs the page to be locked during writeout, that is ok, too,
244the page is allowed to be unlocked at any point in time between the calls to
245set_page_writeback() and end_page_writeback().
246
247Note, failure to run either redirty_page_for_writepage() or the combination of
248set_page_writeback()/end_page_writeback() on a page submitted to writepage
249will leave the page itself marked clean but it will be tagged as dirty in the
250radix tree. This incoherency can lead to all sorts of hard-to-debug problems
251in the filesystem like having dirty inodes at umount and losing written data.
252
253 ->sync_page() locking rules are not well-defined - usually it is called
254with lock on page, but that is not guaranteed. Considering the currently
255existing instances of this method ->sync_page() itself doesn't look
256well-defined...
257
258 ->writepages() is used for periodic writeback and for syscall-initiated
259sync operations. The address_space should start I/O against at least
260*nr_to_write pages. *nr_to_write must be decremented for each page which is
261written. The address_space implementation may write more (or less) pages
262than *nr_to_write asks for, but it should try to be reasonably close. If
263nr_to_write is NULL, all dirty pages must be written.
264
265writepages should _only_ write pages which are present on
266mapping->io_pages.
267
268 ->set_page_dirty() is called from various places in the kernel
269when the target page is marked as needing writeback. It may be called
270under spinlock (it cannot block) and is sometimes called with the page
271not locked.
272
273 ->bmap() is currently used by legacy ioctl() (FIBMAP) provided by some
274filesystems and by the swapper. The latter will eventually go away. All
275instances do not actually need the BKL. Please, keep it that way and don't
276breed new callers.
277
278 ->invalidatepage() is called when the filesystem must attempt to drop
279some or all of the buffers from the page when it is being truncated. It
280returns zero on success. If ->invalidatepage is zero, the kernel uses
281block_invalidatepage() instead.
282
283 ->releasepage() is called when the kernel is about to try to drop the
284buffers from the page in preparation for freeing it. It returns zero to
285indicate that the buffers are (or may be) freeable. If ->releasepage is zero,
286the kernel assumes that the fs has no private interest in the buffers.
287
Trond Myklebuste3db7692007-01-10 23:15:39 -0800288 ->launder_page() may be called prior to releasing a page if
289it is still found to be dirty. It returns zero if the page was successfully
290cleaned, or an error value if not. Note that in order to prevent the page
291getting mapped back in and redirtied, it needs to be kept locked
292across the entire operation.
293
Linus Torvalds1da177e2005-04-16 15:20:36 -0700294 Note: currently almost all instances of address_space methods are
295using BKL for internal serialization and that's one of the worst sources
296of contention. Normally they are calling library functions (in fs/buffer.c)
297and pass foo_get_block() as a callback (on local block-based filesystems,
298indeed). BKL is not needed for library stuff and is usually taken by
299foo_get_block(). It's an overkill, since block bitmaps can be protected by
300internal fs locking and real critical areas are much smaller than the areas
301filesystems protect now.
302
303----------------------- file_lock_operations ------------------------------
304prototypes:
305 void (*fl_insert)(struct file_lock *); /* lock insertion callback */
306 void (*fl_remove)(struct file_lock *); /* lock removal callback */
307 void (*fl_copy_lock)(struct file_lock *, struct file_lock *);
308 void (*fl_release_private)(struct file_lock *);
309
310
311locking rules:
312 BKL may block
313fl_insert: yes no
314fl_remove: yes no
315fl_copy_lock: yes no
316fl_release_private: yes yes
317
318----------------------- lock_manager_operations ---------------------------
319prototypes:
320 int (*fl_compare_owner)(struct file_lock *, struct file_lock *);
321 void (*fl_notify)(struct file_lock *); /* unblock callback */
322 void (*fl_copy_lock)(struct file_lock *, struct file_lock *);
323 void (*fl_release_private)(struct file_lock *);
324 void (*fl_break)(struct file_lock *); /* break_lease callback */
325
326locking rules:
327 BKL may block
328fl_compare_owner: yes no
329fl_notify: yes no
330fl_copy_lock: yes no
331fl_release_private: yes yes
332fl_break: yes no
333
334 Currently only NFSD and NLM provide instances of this class. None of the
335them block. If you have out-of-tree instances - please, show up. Locking
336in that area will change.
337--------------------------- buffer_head -----------------------------------
338prototypes:
339 void (*b_end_io)(struct buffer_head *bh, int uptodate);
340
341locking rules:
342 called from interrupts. In other words, extreme care is needed here.
343bh is locked, but that's all warranties we have here. Currently only RAID1,
344highmem, fs/buffer.c, and fs/ntfs/aops.c are providing these. Block devices
345call this method upon the IO completion.
346
347--------------------------- block_device_operations -----------------------
348prototypes:
349 int (*open) (struct inode *, struct file *);
350 int (*release) (struct inode *, struct file *);
351 int (*ioctl) (struct inode *, struct file *, unsigned, unsigned long);
352 int (*media_changed) (struct gendisk *);
353 int (*revalidate_disk) (struct gendisk *);
354
355locking rules:
356 BKL bd_sem
357open: yes yes
358release: yes yes
359ioctl: yes no
360media_changed: no no
361revalidate_disk: no no
362
363The last two are called only from check_disk_change().
364
365--------------------------- file_operations -------------------------------
366prototypes:
367 loff_t (*llseek) (struct file *, loff_t, int);
368 ssize_t (*read) (struct file *, char __user *, size_t, loff_t *);
Linus Torvalds1da177e2005-04-16 15:20:36 -0700369 ssize_t (*write) (struct file *, const char __user *, size_t, loff_t *);
Badari Pulavarty027445c2006-09-30 23:28:46 -0700370 ssize_t (*aio_read) (struct kiocb *, const struct iovec *, unsigned long, loff_t);
371 ssize_t (*aio_write) (struct kiocb *, const struct iovec *, unsigned long, loff_t);
Linus Torvalds1da177e2005-04-16 15:20:36 -0700372 int (*readdir) (struct file *, void *, filldir_t);
373 unsigned int (*poll) (struct file *, struct poll_table_struct *);
374 int (*ioctl) (struct inode *, struct file *, unsigned int,
375 unsigned long);
376 long (*unlocked_ioctl) (struct file *, unsigned int, unsigned long);
377 long (*compat_ioctl) (struct file *, unsigned int, unsigned long);
378 int (*mmap) (struct file *, struct vm_area_struct *);
379 int (*open) (struct inode *, struct file *);
380 int (*flush) (struct file *);
381 int (*release) (struct inode *, struct file *);
382 int (*fsync) (struct file *, struct dentry *, int datasync);
383 int (*aio_fsync) (struct kiocb *, int datasync);
384 int (*fasync) (int, struct file *, int);
385 int (*lock) (struct file *, int, struct file_lock *);
386 ssize_t (*readv) (struct file *, const struct iovec *, unsigned long,
387 loff_t *);
388 ssize_t (*writev) (struct file *, const struct iovec *, unsigned long,
389 loff_t *);
390 ssize_t (*sendfile) (struct file *, loff_t *, size_t, read_actor_t,
391 void __user *);
392 ssize_t (*sendpage) (struct file *, struct page *, int, size_t,
393 loff_t *, int);
394 unsigned long (*get_unmapped_area)(struct file *, unsigned long,
395 unsigned long, unsigned long, unsigned long);
396 int (*check_flags)(int);
Linus Torvalds1da177e2005-04-16 15:20:36 -0700397};
398
399locking rules:
Tejun Heo5f820f62009-01-06 14:40:59 -0800400 All may block.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700401 BKL
402llseek: no (see below)
403read: no
404aio_read: no
405write: no
406aio_write: no
407readdir: no
408poll: no
409ioctl: yes (see below)
410unlocked_ioctl: no (see below)
411compat_ioctl: no
412mmap: no
Christoph Hellwigadaae722008-09-09 20:02:01 +0200413open: no
Linus Torvalds1da177e2005-04-16 15:20:36 -0700414flush: no
415release: no
416fsync: no (see below)
417aio_fsync: no
Christoph Hellwigadaae722008-09-09 20:02:01 +0200418fasync: no
Linus Torvalds1da177e2005-04-16 15:20:36 -0700419lock: yes
420readv: no
421writev: no
422sendfile: no
423sendpage: no
424get_unmapped_area: no
425check_flags: no
Linus Torvalds1da177e2005-04-16 15:20:36 -0700426
427->llseek() locking has moved from llseek to the individual llseek
428implementations. If your fs is not using generic_file_llseek, you
429need to acquire and release the appropriate locks in your ->llseek().
430For many filesystems, it is probably safe to acquire the inode
431semaphore. Note some filesystems (i.e. remote ones) provide no
432protection for i_size so you will need to use the BKL.
433
Linus Torvalds1da177e2005-04-16 15:20:36 -0700434Note: ext2_release() was *the* source of contention on fs-intensive
435loads and dropping BKL on ->release() helps to get rid of that (we still
436grab BKL for cases when we close a file that had been opened r/w, but that
437can and should be done using the internal locking with smaller critical areas).
438Current worst offender is ext2_get_block()...
439
Jonathan Corbet76398422009-02-01 14:26:59 -0700440->fasync() is called without BKL protection, and is responsible for
441maintaining the FASYNC bit in filp->f_flags. Most instances call
442fasync_helper(), which does that maintenance, so it's not normally
443something one needs to worry about. Return values > 0 will be mapped to
444zero in the VFS layer.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700445
446->readdir() and ->ioctl() on directories must be changed. Ideally we would
447move ->readdir() to inode_operations and use a separate method for directory
448->ioctl() or kill the latter completely. One of the problems is that for
449anything that resembles union-mount we won't have a struct file for all
450components. And there are other reasons why the current interface is a mess...
451
452->ioctl() on regular files is superceded by the ->unlocked_ioctl() that
453doesn't take the BKL.
454
455->read on directories probably must go away - we should just enforce -EISDIR
456in sys_read() and friends.
457
Artem Bityutskiya7bc02f2007-05-09 07:53:16 +0200458->fsync() has i_mutex on inode.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700459
460--------------------------- dquot_operations -------------------------------
461prototypes:
462 int (*initialize) (struct inode *, int);
463 int (*drop) (struct inode *);
464 int (*alloc_space) (struct inode *, qsize_t, int);
465 int (*alloc_inode) (const struct inode *, unsigned long);
466 int (*free_space) (struct inode *, qsize_t);
467 int (*free_inode) (const struct inode *, unsigned long);
468 int (*transfer) (struct inode *, struct iattr *);
469 int (*write_dquot) (struct dquot *);
470 int (*acquire_dquot) (struct dquot *);
471 int (*release_dquot) (struct dquot *);
472 int (*mark_dirty) (struct dquot *);
473 int (*write_info) (struct super_block *, int);
474
475These operations are intended to be more or less wrapping functions that ensure
476a proper locking wrt the filesystem and call the generic quota operations.
477
478What filesystem should expect from the generic quota functions:
479
480 FS recursion Held locks when called
481initialize: yes maybe dqonoff_sem
482drop: yes -
483alloc_space: ->mark_dirty() -
484alloc_inode: ->mark_dirty() -
485free_space: ->mark_dirty() -
486free_inode: ->mark_dirty() -
487transfer: yes -
488write_dquot: yes dqonoff_sem or dqptr_sem
489acquire_dquot: yes dqonoff_sem or dqptr_sem
490release_dquot: yes dqonoff_sem or dqptr_sem
491mark_dirty: no -
492write_info: yes dqonoff_sem
493
494FS recursion means calling ->quota_read() and ->quota_write() from superblock
495operations.
496
497->alloc_space(), ->alloc_inode(), ->free_space(), ->free_inode() are called
498only directly by the filesystem and do not call any fs functions only
499the ->mark_dirty() operation.
500
501More details about quota locking can be found in fs/dquot.c.
502
503--------------------------- vm_operations_struct -----------------------------
504prototypes:
505 void (*open)(struct vm_area_struct*);
506 void (*close)(struct vm_area_struct*);
Nick Piggind0217ac2007-07-19 01:47:03 -0700507 int (*fault)(struct vm_area_struct*, struct vm_fault *);
Nick Pigginc2ec1752009-03-31 15:23:21 -0700508 int (*page_mkwrite)(struct vm_area_struct *, struct vm_fault *);
Rik van Riel28b2ee22008-07-23 21:27:05 -0700509 int (*access)(struct vm_area_struct *, unsigned long, void*, int, int);
Linus Torvalds1da177e2005-04-16 15:20:36 -0700510
511locking rules:
Mark Fashehed2f2f92007-07-19 01:47:01 -0700512 BKL mmap_sem PageLocked(page)
Linus Torvalds1da177e2005-04-16 15:20:36 -0700513open: no yes
514close: no yes
Nick Piggin54cb8822007-07-19 01:46:59 -0700515fault: no yes
Mark Fashehed2f2f92007-07-19 01:47:01 -0700516page_mkwrite: no yes no
Rik van Riel28b2ee22008-07-23 21:27:05 -0700517access: no yes
Mark Fashehed2f2f92007-07-19 01:47:01 -0700518
519 ->page_mkwrite() is called when a previously read-only page is
520about to become writeable. The file system is responsible for
521protecting against truncate races. Once appropriate action has been
522taking to lock out truncate, the page range should be verified to be
523within i_size. The page mapping should also be checked that it is not
524NULL.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700525
Rik van Riel28b2ee22008-07-23 21:27:05 -0700526 ->access() is called when get_user_pages() fails in
527acces_process_vm(), typically used to debug a process through
528/proc/pid/mem or ptrace. This function is needed only for
529VM_IO | VM_PFNMAP VMAs.
530
Linus Torvalds1da177e2005-04-16 15:20:36 -0700531================================================================================
532 Dubious stuff
533
534(if you break something or notice that it is broken and do not fix it yourself
535- at least put it here)
536
537ipc/shm.c::shm_delete() - may need BKL.
538->read() and ->write() in many drivers are (probably) missing BKL.