file.c 47.7 KB
Newer Older
1
2
/*
  FUSE: Filesystem in Userspace
Miklos Szeredi's avatar
Miklos Szeredi committed
3
  Copyright (C) 2001-2008  Miklos Szeredi <miklos@szeredi.hu>
4
5
6
7
8
9
10
11
12
13

  This program can be distributed under the terms of the GNU GPL.
  See the file COPYING.
*/

#include "fuse_i.h"

#include <linux/pagemap.h>
#include <linux/slab.h>
#include <linux/kernel.h>
Alexey Dobriyan's avatar
Alexey Dobriyan committed
14
#include <linux/sched.h>
15

16
static const struct file_operations fuse_direct_io_file_operations;
17

18
19
static int fuse_send_open(struct inode *inode, struct file *file, int isdir,
			  struct fuse_open_out *outargp)
20
21
22
{
	struct fuse_conn *fc = get_fuse_conn(inode);
	struct fuse_open_in inarg;
23
24
25
	struct fuse_req *req;
	int err;

26
27
28
	req = fuse_get_req(fc);
	if (IS_ERR(req))
		return PTR_ERR(req);
29
30

	memset(&inarg, 0, sizeof(inarg));
31
32
33
	inarg.flags = file->f_flags & ~(O_CREAT | O_EXCL | O_NOCTTY);
	if (!fc->atomic_o_trunc)
		inarg.flags &= ~O_TRUNC;
34
35
36
37
38
39
40
41
	req->in.h.opcode = isdir ? FUSE_OPENDIR : FUSE_OPEN;
	req->in.h.nodeid = get_node_id(inode);
	req->in.numargs = 1;
	req->in.args[0].size = sizeof(inarg);
	req->in.args[0].value = &inarg;
	req->out.numargs = 1;
	req->out.args[0].size = sizeof(*outargp);
	req->out.args[0].value = outargp;
42
	fuse_request_send(fc, req);
43
44
45
46
47
48
	err = req->out.h.error;
	fuse_put_request(fc, req);

	return err;
}

Tejun Heo's avatar
Tejun Heo committed
49
struct fuse_file *fuse_file_alloc(struct fuse_conn *fc)
50
51
{
	struct fuse_file *ff;
Tejun Heo's avatar
Tejun Heo committed
52

53
	ff = kmalloc(sizeof(struct fuse_file), GFP_KERNEL);
Tejun Heo's avatar
Tejun Heo committed
54
55
56
57
58
59
60
	if (unlikely(!ff))
		return NULL;

	ff->reserved_req = fuse_request_alloc();
	if (unlikely(!ff->reserved_req)) {
		kfree(ff);
		return NULL;
61
	}
Tejun Heo's avatar
Tejun Heo committed
62
63
64
65
66
67
68
69
70
71

	INIT_LIST_HEAD(&ff->write_entry);
	atomic_set(&ff->count, 0);
	RB_CLEAR_NODE(&ff->polled_node);
	init_waitqueue_head(&ff->poll_wait);

	spin_lock(&fc->lock);
	ff->kh = ++fc->khctr;
	spin_unlock(&fc->lock);

72
73
74
75
76
	return ff;
}

void fuse_file_free(struct fuse_file *ff)
{
77
	fuse_request_free(ff->reserved_req);
78
79
80
	kfree(ff);
}

81
82
83
84
85
86
static struct fuse_file *fuse_file_get(struct fuse_file *ff)
{
	atomic_inc(&ff->count);
	return ff;
}

Miklos Szeredi's avatar
Miklos Szeredi committed
87
88
static void fuse_release_end(struct fuse_conn *fc, struct fuse_req *req)
{
89
90
	dput(req->misc.release.dentry);
	mntput(req->misc.release.vfsmount);
Miklos Szeredi's avatar
Miklos Szeredi committed
91
92
}

93
94
95
96
static void fuse_file_put(struct fuse_file *ff)
{
	if (atomic_dec_and_test(&ff->count)) {
		struct fuse_req *req = ff->reserved_req;
97
98
		struct inode *inode = req->misc.release.dentry->d_inode;
		struct fuse_conn *fc = get_fuse_conn(inode);
Miklos Szeredi's avatar
Miklos Szeredi committed
99
		req->end = fuse_release_end;
100
		fuse_request_send_background(fc, req);
101
102
103
104
		kfree(ff);
	}
}

105
106
107
108
109
110
void fuse_finish_open(struct inode *inode, struct file *file,
		      struct fuse_file *ff, struct fuse_open_out *outarg)
{
	if (outarg->open_flags & FOPEN_DIRECT_IO)
		file->f_op = &fuse_direct_io_file_operations;
	if (!(outarg->open_flags & FOPEN_KEEP_CACHE))
Miklos Szeredi's avatar
Miklos Szeredi committed
111
		invalidate_inode_pages2(inode->i_mapping);
Tejun Heo's avatar
Tejun Heo committed
112
113
	if (outarg->open_flags & FOPEN_NONSEEKABLE)
		nonseekable_open(inode, file);
114
	ff->fh = outarg->fh;
115
	file->private_data = fuse_file_get(ff);
116
117
118
119
}

int fuse_open_common(struct inode *inode, struct file *file, int isdir)
{
Tejun Heo's avatar
Tejun Heo committed
120
	struct fuse_conn *fc = get_fuse_conn(inode);
121
122
123
124
	struct fuse_open_out outarg;
	struct fuse_file *ff;
	int err;

Miklos Szeredi's avatar
Miklos Szeredi committed
125
126
127
128
	/* VFS checks this, but only _after_ ->open() */
	if (file->f_flags & O_DIRECT)
		return -EINVAL;

129
130
131
132
	err = generic_file_open(inode, file);
	if (err)
		return err;

Tejun Heo's avatar
Tejun Heo committed
133
	ff = fuse_file_alloc(fc);
134
	if (!ff)
135
		return -ENOMEM;
136

137
138
139
140
141
142
143
	err = fuse_send_open(inode, file, isdir, &outarg);
	if (err)
		fuse_file_free(ff);
	else {
		if (isdir)
			outarg.open_flags &= ~FOPEN_DIRECT_IO;
		fuse_finish_open(inode, file, ff, &outarg);
144
145
146
147
148
	}

	return err;
}

149
void fuse_release_fill(struct fuse_file *ff, u64 nodeid, int flags, int opcode)
150
{
151
	struct fuse_req *req = ff->reserved_req;
152
	struct fuse_release_in *inarg = &req->misc.release.in;
153
154

	inarg->fh = ff->fh;
155
	inarg->flags = flags;
156
	req->in.h.opcode = opcode;
157
	req->in.h.nodeid = nodeid;
158
159
160
	req->in.numargs = 1;
	req->in.args[0].size = sizeof(struct fuse_release_in);
	req->in.args[0].value = inarg;
161
162
163
164
}

int fuse_release_common(struct inode *inode, struct file *file, int isdir)
{
Tejun Heo's avatar
Tejun Heo committed
165
166
167
	struct fuse_conn *fc;
	struct fuse_file *ff;
	struct fuse_req *req;
168

Tejun Heo's avatar
Tejun Heo committed
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
	ff = file->private_data;
	if (unlikely(!ff))
		return 0;	/* return value is ignored by VFS */

	fc = get_fuse_conn(inode);
	req = ff->reserved_req;

	fuse_release_fill(ff, get_node_id(inode), file->f_flags,
			  isdir ? FUSE_RELEASEDIR : FUSE_RELEASE);

	/* Hold vfsmount and dentry until release is finished */
	req->misc.release.vfsmount = mntget(file->f_path.mnt);
	req->misc.release.dentry = dget(file->f_path.dentry);

	spin_lock(&fc->lock);
	list_del(&ff->write_entry);
	if (!RB_EMPTY_NODE(&ff->polled_node))
		rb_erase(&ff->polled_node, &fc->polled_files);
	spin_unlock(&fc->lock);

	wake_up_interruptible_sync(&ff->poll_wait);
	/*
	 * Normally this will send the RELEASE request, however if
	 * some asynchronous READ or WRITE requests are outstanding,
	 * the sending will be delayed.
	 */
	fuse_file_put(ff);
196
197
198
	return 0;
}

199
200
201
202
203
204
205
206
207
208
static int fuse_open(struct inode *inode, struct file *file)
{
	return fuse_open_common(inode, file, 0);
}

static int fuse_release(struct inode *inode, struct file *file)
{
	return fuse_release_common(inode, file, 0);
}

209
/*
210
211
 * Scramble the ID space with XTEA, so that the value of the files_struct
 * pointer is not exposed to userspace.
212
 */
213
u64 fuse_lock_owner_id(struct fuse_conn *fc, fl_owner_t id)
214
{
215
216
217
218
219
220
221
222
223
224
225
226
227
228
	u32 *k = fc->scramble_key;
	u64 v = (unsigned long) id;
	u32 v0 = v;
	u32 v1 = v >> 32;
	u32 sum = 0;
	int i;

	for (i = 0; i < 32; i++) {
		v0 += ((v1 << 4 ^ v1 >> 5) + v1) ^ (sum + k[sum & 3]);
		sum += 0x9E3779B9;
		v1 += ((v0 << 4 ^ v0 >> 5) + v0) ^ (sum + k[sum>>11 & 3]);
	}

	return (u64) v0 + ((u64) v1 << 32);
229
230
}

Miklos Szeredi's avatar
Miklos Szeredi committed
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
/*
 * Check if page is under writeback
 *
 * This is currently done by walking the list of writepage requests
 * for the inode, which can be pretty inefficient.
 */
static bool fuse_page_is_writeback(struct inode *inode, pgoff_t index)
{
	struct fuse_conn *fc = get_fuse_conn(inode);
	struct fuse_inode *fi = get_fuse_inode(inode);
	struct fuse_req *req;
	bool found = false;

	spin_lock(&fc->lock);
	list_for_each_entry(req, &fi->writepages, writepages_entry) {
		pgoff_t curr_index;

		BUG_ON(req->inode != inode);
		curr_index = req->misc.write.in.offset >> PAGE_CACHE_SHIFT;
		if (curr_index == index) {
			found = true;
			break;
		}
	}
	spin_unlock(&fc->lock);

	return found;
}

/*
 * Wait for page writeback to be completed.
 *
 * Since fuse doesn't rely on the VM writeback tracking, this has to
 * use some other means.
 */
static int fuse_wait_on_page_writeback(struct inode *inode, pgoff_t index)
{
	struct fuse_inode *fi = get_fuse_inode(inode);

	wait_event(fi->page_waitq, !fuse_page_is_writeback(inode, index));
	return 0;
}

274
static int fuse_flush(struct file *file, fl_owner_t id)
275
{
276
	struct inode *inode = file->f_path.dentry->d_inode;
277
278
279
280
281
282
	struct fuse_conn *fc = get_fuse_conn(inode);
	struct fuse_file *ff = file->private_data;
	struct fuse_req *req;
	struct fuse_flush_in inarg;
	int err;

283
284
285
	if (is_bad_inode(inode))
		return -EIO;

286
287
288
	if (fc->no_flush)
		return 0;

289
	req = fuse_get_req_nofail(fc, file);
290
291
	memset(&inarg, 0, sizeof(inarg));
	inarg.fh = ff->fh;
292
	inarg.lock_owner = fuse_lock_owner_id(fc, id);
293
294
295
296
297
	req->in.h.opcode = FUSE_FLUSH;
	req->in.h.nodeid = get_node_id(inode);
	req->in.numargs = 1;
	req->in.args[0].size = sizeof(inarg);
	req->in.args[0].value = &inarg;
298
	req->force = 1;
299
	fuse_request_send(fc, req);
300
301
302
303
304
305
306
307
308
	err = req->out.h.error;
	fuse_put_request(fc, req);
	if (err == -ENOSYS) {
		fc->no_flush = 1;
		err = 0;
	}
	return err;
}

Miklos Szeredi's avatar
Miklos Szeredi committed
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
/*
 * Wait for all pending writepages on the inode to finish.
 *
 * This is currently done by blocking further writes with FUSE_NOWRITE
 * and waiting for all sent writes to complete.
 *
 * This must be called under i_mutex, otherwise the FUSE_NOWRITE usage
 * could conflict with truncation.
 */
static void fuse_sync_writes(struct inode *inode)
{
	fuse_set_nowrite(inode);
	fuse_release_nowrite(inode);
}

324
325
int fuse_fsync_common(struct file *file, struct dentry *de, int datasync,
		      int isdir)
326
327
328
329
330
331
332
333
{
	struct inode *inode = de->d_inode;
	struct fuse_conn *fc = get_fuse_conn(inode);
	struct fuse_file *ff = file->private_data;
	struct fuse_req *req;
	struct fuse_fsync_in inarg;
	int err;

334
335
336
	if (is_bad_inode(inode))
		return -EIO;

337
	if ((!isdir && fc->no_fsync) || (isdir && fc->no_fsyncdir))
338
339
		return 0;

Miklos Szeredi's avatar
Miklos Szeredi committed
340
341
342
343
344
345
346
347
348
349
350
	/*
	 * Start writeback against all dirty pages of the inode, then
	 * wait for all outstanding writes, before sending the FSYNC
	 * request.
	 */
	err = write_inode_now(inode, 0);
	if (err)
		return err;

	fuse_sync_writes(inode);

351
352
353
	req = fuse_get_req(fc);
	if (IS_ERR(req))
		return PTR_ERR(req);
354
355
356
357

	memset(&inarg, 0, sizeof(inarg));
	inarg.fh = ff->fh;
	inarg.fsync_flags = datasync ? 1 : 0;
358
	req->in.h.opcode = isdir ? FUSE_FSYNCDIR : FUSE_FSYNC;
359
360
361
362
	req->in.h.nodeid = get_node_id(inode);
	req->in.numargs = 1;
	req->in.args[0].size = sizeof(inarg);
	req->in.args[0].value = &inarg;
363
	fuse_request_send(fc, req);
364
365
366
	err = req->out.h.error;
	fuse_put_request(fc, req);
	if (err == -ENOSYS) {
367
368
369
370
		if (isdir)
			fc->no_fsyncdir = 1;
		else
			fc->no_fsync = 1;
371
372
373
374
375
		err = 0;
	}
	return err;
}

376
377
378
379
380
static int fuse_fsync(struct file *file, struct dentry *de, int datasync)
{
	return fuse_fsync_common(file, de, datasync, 0);
}

381
void fuse_read_fill(struct fuse_req *req, struct file *file,
382
		    struct inode *inode, loff_t pos, size_t count, int opcode)
383
{
384
	struct fuse_read_in *inarg = &req->misc.read.in;
385
	struct fuse_file *ff = file->private_data;
386

387
388
389
	inarg->fh = ff->fh;
	inarg->offset = pos;
	inarg->size = count;
390
	inarg->flags = file->f_flags;
391
	req->in.h.opcode = opcode;
392
393
394
	req->in.h.nodeid = get_node_id(inode);
	req->in.numargs = 1;
	req->in.args[0].size = sizeof(struct fuse_read_in);
395
	req->in.args[0].value = inarg;
396
397
398
399
400
	req->out.argvar = 1;
	req->out.numargs = 1;
	req->out.args[0].size = count;
}

401
static size_t fuse_send_read(struct fuse_req *req, struct file *file,
402
403
			     struct inode *inode, loff_t pos, size_t count,
			     fl_owner_t owner)
404
{
405
	struct fuse_conn *fc = get_fuse_conn(inode);
406

407
	fuse_read_fill(req, file, inode, pos, count, FUSE_READ);
408
	if (owner != NULL) {
409
		struct fuse_read_in *inarg = &req->misc.read.in;
410
411
412
413

		inarg->read_flags |= FUSE_READ_LOCKOWNER;
		inarg->lock_owner = fuse_lock_owner_id(fc, owner);
	}
414
	fuse_request_send(fc, req);
415
	return req->out.args[0].size;
416
417
}

418
419
420
421
422
423
424
425
426
427
428
429
430
431
static void fuse_read_update_size(struct inode *inode, loff_t size,
				  u64 attr_ver)
{
	struct fuse_conn *fc = get_fuse_conn(inode);
	struct fuse_inode *fi = get_fuse_inode(inode);

	spin_lock(&fc->lock);
	if (attr_ver == fi->attr_version && size < inode->i_size) {
		fi->attr_version = ++fc->attr_version;
		i_size_write(inode, size);
	}
	spin_unlock(&fc->lock);
}

432
433
434
435
static int fuse_readpage(struct file *file, struct page *page)
{
	struct inode *inode = page->mapping->host;
	struct fuse_conn *fc = get_fuse_conn(inode);
436
	struct fuse_req *req;
437
438
439
440
	size_t num_read;
	loff_t pos = page_offset(page);
	size_t count = PAGE_CACHE_SIZE;
	u64 attr_ver;
441
442
443
444
445
446
	int err;

	err = -EIO;
	if (is_bad_inode(inode))
		goto out;

Miklos Szeredi's avatar
Miklos Szeredi committed
447
448
449
450
451
452
453
	/*
	 * Page writeback can extend beyond the liftime of the
	 * page-cache page, so make sure we read a properly synced
	 * page.
	 */
	fuse_wait_on_page_writeback(inode, page->index);

454
455
456
	req = fuse_get_req(fc);
	err = PTR_ERR(req);
	if (IS_ERR(req))
457
458
		goto out;

459
460
	attr_ver = fuse_get_attr_version(fc);

461
	req->out.page_zeroing = 1;
462
	req->out.argpages = 1;
463
464
	req->num_pages = 1;
	req->pages[0] = page;
465
	num_read = fuse_send_read(req, file, inode, pos, count, NULL);
466
467
	err = req->out.h.error;
	fuse_put_request(fc, req);
468
469
470
471
472
473
474
475

	if (!err) {
		/*
		 * Short read means EOF.  If file size is larger, truncate it
		 */
		if (num_read < count)
			fuse_read_update_size(inode, pos + num_read, attr_ver);

476
		SetPageUptodate(page);
477
478
	}

479
	fuse_invalidate_attr(inode); /* atime changed */
480
481
482
483
484
 out:
	unlock_page(page);
	return err;
}

485
static void fuse_readpages_end(struct fuse_conn *fc, struct fuse_req *req)
486
{
487
	int i;
488
489
490
	size_t count = req->misc.read.in.size;
	size_t num_read = req->out.args[0].size;
	struct inode *inode = req->pages[0]->mapping->host;
491

492
493
494
495
496
497
498
499
500
	/*
	 * Short read means EOF.  If file size is larger, truncate it
	 */
	if (!req->out.h.error && num_read < count) {
		loff_t pos = page_offset(req->pages[0]) + num_read;
		fuse_read_update_size(inode, pos, req->misc.read.attr_ver);
	}

	fuse_invalidate_attr(inode); /* atime changed */
501

502
503
504
505
	for (i = 0; i < req->num_pages; i++) {
		struct page *page = req->pages[i];
		if (!req->out.h.error)
			SetPageUptodate(page);
506
507
		else
			SetPageError(page);
508
509
		unlock_page(page);
	}
510
511
	if (req->ff)
		fuse_file_put(req->ff);
512
513
}

514
static void fuse_send_readpages(struct fuse_req *req, struct file *file,
515
516
517
518
519
				struct inode *inode)
{
	struct fuse_conn *fc = get_fuse_conn(inode);
	loff_t pos = page_offset(req->pages[0]);
	size_t count = req->num_pages << PAGE_CACHE_SHIFT;
520
521

	req->out.argpages = 1;
522
	req->out.page_zeroing = 1;
523
	fuse_read_fill(req, file, inode, pos, count, FUSE_READ);
524
	req->misc.read.attr_ver = fuse_get_attr_version(fc);
525
	if (fc->async_read) {
526
		struct fuse_file *ff = file->private_data;
527
		req->ff = fuse_file_get(ff);
528
		req->end = fuse_readpages_end;
529
		fuse_request_send_background(fc, req);
530
	} else {
531
		fuse_request_send(fc, req);
532
		fuse_readpages_end(fc, req);
533
		fuse_put_request(fc, req);
534
	}
535
536
}

537
struct fuse_fill_data {
538
	struct fuse_req *req;
539
	struct file *file;
540
541
542
543
544
	struct inode *inode;
};

static int fuse_readpages_fill(void *_data, struct page *page)
{
545
	struct fuse_fill_data *data = _data;
546
547
548
549
	struct fuse_req *req = data->req;
	struct inode *inode = data->inode;
	struct fuse_conn *fc = get_fuse_conn(inode);

Miklos Szeredi's avatar
Miklos Szeredi committed
550
551
	fuse_wait_on_page_writeback(inode, page->index);

552
553
554
555
	if (req->num_pages &&
	    (req->num_pages == FUSE_MAX_PAGES_PER_REQ ||
	     (req->num_pages + 1) * PAGE_CACHE_SIZE > fc->max_read ||
	     req->pages[req->num_pages - 1]->index + 1 != page->index)) {
556
		fuse_send_readpages(req, data->file, inode);
557
558
		data->req = req = fuse_get_req(fc);
		if (IS_ERR(req)) {
559
			unlock_page(page);
560
			return PTR_ERR(req);
561
562
563
		}
	}
	req->pages[req->num_pages] = page;
Miklos Szeredi's avatar
Miklos Szeredi committed
564
	req->num_pages++;
565
566
567
568
569
570
571
572
	return 0;
}

static int fuse_readpages(struct file *file, struct address_space *mapping,
			  struct list_head *pages, unsigned nr_pages)
{
	struct inode *inode = mapping->host;
	struct fuse_conn *fc = get_fuse_conn(inode);
573
	struct fuse_fill_data data;
574
	int err;
575

576
	err = -EIO;
577
	if (is_bad_inode(inode))
578
		goto out;
579

580
	data.file = file;
581
	data.inode = inode;
582
	data.req = fuse_get_req(fc);
583
	err = PTR_ERR(data.req);
584
	if (IS_ERR(data.req))
585
		goto out;
586
587

	err = read_cache_pages(mapping, pages, fuse_readpages_fill, &data);
588
589
	if (!err) {
		if (data.req->num_pages)
590
			fuse_send_readpages(data.req, file, inode);
591
592
593
		else
			fuse_put_request(fc, data.req);
	}
594
out:
595
	return err;
596
597
}

Miklos Szeredi's avatar
Miklos Szeredi committed
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
static ssize_t fuse_file_aio_read(struct kiocb *iocb, const struct iovec *iov,
				  unsigned long nr_segs, loff_t pos)
{
	struct inode *inode = iocb->ki_filp->f_mapping->host;

	if (pos + iov_length(iov, nr_segs) > i_size_read(inode)) {
		int err;
		/*
		 * If trying to read past EOF, make sure the i_size
		 * attribute is up-to-date.
		 */
		err = fuse_update_attributes(inode, NULL, iocb->ki_filp, NULL);
		if (err)
			return err;
	}

	return generic_file_aio_read(iocb, iov, nr_segs, pos);
}

617
static void fuse_write_fill(struct fuse_req *req, struct file *file,
Miklos Szeredi's avatar
Miklos Szeredi committed
618
619
			    struct fuse_file *ff, struct inode *inode,
			    loff_t pos, size_t count, int writepage)
620
{
621
	struct fuse_conn *fc = get_fuse_conn(inode);
622
623
	struct fuse_write_in *inarg = &req->misc.write.in;
	struct fuse_write_out *outarg = &req->misc.write.out;
624

625
626
627
628
629
	memset(inarg, 0, sizeof(struct fuse_write_in));
	inarg->fh = ff->fh;
	inarg->offset = pos;
	inarg->size = count;
	inarg->write_flags = writepage ? FUSE_WRITE_CACHE : 0;
Miklos Szeredi's avatar
Miklos Szeredi committed
630
	inarg->flags = file ? file->f_flags : 0;
631
632
633
	req->in.h.opcode = FUSE_WRITE;
	req->in.h.nodeid = get_node_id(inode);
	req->in.numargs = 2;
634
635
636
637
	if (fc->minor < 9)
		req->in.args[0].size = FUSE_COMPAT_WRITE_IN_SIZE;
	else
		req->in.args[0].size = sizeof(struct fuse_write_in);
638
	req->in.args[0].value = inarg;
639
640
641
	req->in.args[1].size = count;
	req->out.numargs = 1;
	req->out.args[0].size = sizeof(struct fuse_write_out);
642
643
644
645
	req->out.args[0].value = outarg;
}

static size_t fuse_send_write(struct fuse_req *req, struct file *file,
646
647
			      struct inode *inode, loff_t pos, size_t count,
			      fl_owner_t owner)
648
649
{
	struct fuse_conn *fc = get_fuse_conn(inode);
Miklos Szeredi's avatar
Miklos Szeredi committed
650
	fuse_write_fill(req, file, file->private_data, inode, pos, count, 0);
651
652
653
654
655
	if (owner != NULL) {
		struct fuse_write_in *inarg = &req->misc.write.in;
		inarg->write_flags |= FUSE_WRITE_LOCKOWNER;
		inarg->lock_owner = fuse_lock_owner_id(fc, owner);
	}
656
	fuse_request_send(fc, req);
657
	return req->misc.write.out.size;
658
659
}

Nick Piggin's avatar
Nick Piggin committed
660
661
662
static int fuse_write_begin(struct file *file, struct address_space *mapping,
			loff_t pos, unsigned len, unsigned flags,
			struct page **pagep, void **fsdata)
663
{
Nick Piggin's avatar
Nick Piggin committed
664
665
	pgoff_t index = pos >> PAGE_CACHE_SHIFT;

666
	*pagep = grab_cache_page_write_begin(mapping, index, flags);
Nick Piggin's avatar
Nick Piggin committed
667
668
	if (!*pagep)
		return -ENOMEM;
669
670
671
	return 0;
}

672
673
674
675
676
677
678
679
680
681
682
683
static void fuse_write_update_size(struct inode *inode, loff_t pos)
{
	struct fuse_conn *fc = get_fuse_conn(inode);
	struct fuse_inode *fi = get_fuse_inode(inode);

	spin_lock(&fc->lock);
	fi->attr_version = ++fc->attr_version;
	if (pos > inode->i_size)
		i_size_write(inode, pos);
	spin_unlock(&fc->lock);
}

Nick Piggin's avatar
Nick Piggin committed
684
685
static int fuse_buffered_write(struct file *file, struct inode *inode,
			       loff_t pos, unsigned count, struct page *page)
686
687
{
	int err;
688
	size_t nres;
689
	struct fuse_conn *fc = get_fuse_conn(inode);
Nick Piggin's avatar
Nick Piggin committed
690
	unsigned offset = pos & (PAGE_CACHE_SIZE - 1);
691
692
693
694
695
	struct fuse_req *req;

	if (is_bad_inode(inode))
		return -EIO;

Miklos Szeredi's avatar
Miklos Szeredi committed
696
697
698
699
700
701
	/*
	 * Make sure writepages on the same page are not mixed up with
	 * plain writes.
	 */
	fuse_wait_on_page_writeback(inode, page->index);

702
703
704
	req = fuse_get_req(fc);
	if (IS_ERR(req))
		return PTR_ERR(req);
705

706
	req->in.argpages = 1;
707
708
709
	req->num_pages = 1;
	req->pages[0] = page;
	req->page_offset = offset;
710
	nres = fuse_send_write(req, file, inode, pos, count, NULL);
711
712
	err = req->out.h.error;
	fuse_put_request(fc, req);
Nick Piggin's avatar
Nick Piggin committed
713
	if (!err && !nres)
714
715
		err = -EIO;
	if (!err) {
Nick Piggin's avatar
Nick Piggin committed
716
		pos += nres;
717
		fuse_write_update_size(inode, pos);
Nick Piggin's avatar
Nick Piggin committed
718
		if (count == PAGE_CACHE_SIZE)
719
			SetPageUptodate(page);
720
721
	}
	fuse_invalidate_attr(inode);
Nick Piggin's avatar
Nick Piggin committed
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
	return err ? err : nres;
}

static int fuse_write_end(struct file *file, struct address_space *mapping,
			loff_t pos, unsigned len, unsigned copied,
			struct page *page, void *fsdata)
{
	struct inode *inode = mapping->host;
	int res = 0;

	if (copied)
		res = fuse_buffered_write(file, inode, pos, copied, page);

	unlock_page(page);
	page_cache_release(page);
	return res;
738
739
}

Nick Piggin's avatar
Nick Piggin committed
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
static size_t fuse_send_write_pages(struct fuse_req *req, struct file *file,
				    struct inode *inode, loff_t pos,
				    size_t count)
{
	size_t res;
	unsigned offset;
	unsigned i;

	for (i = 0; i < req->num_pages; i++)
		fuse_wait_on_page_writeback(inode, req->pages[i]->index);

	res = fuse_send_write(req, file, inode, pos, count, NULL);

	offset = req->page_offset;
	count = res;
	for (i = 0; i < req->num_pages; i++) {
		struct page *page = req->pages[i];

		if (!req->out.h.error && !offset && count >= PAGE_CACHE_SIZE)
			SetPageUptodate(page);

		if (count > PAGE_CACHE_SIZE - offset)
			count -= PAGE_CACHE_SIZE - offset;
		else
			count = 0;
		offset = 0;

		unlock_page(page);
		page_cache_release(page);
	}

	return res;
}

static ssize_t fuse_fill_write_pages(struct fuse_req *req,
			       struct address_space *mapping,
			       struct iov_iter *ii, loff_t pos)
{
	struct fuse_conn *fc = get_fuse_conn(mapping->host);
	unsigned offset = pos & (PAGE_CACHE_SIZE - 1);
	size_t count = 0;
	int err;

783
	req->in.argpages = 1;
Nick Piggin's avatar
Nick Piggin committed
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
	req->page_offset = offset;

	do {
		size_t tmp;
		struct page *page;
		pgoff_t index = pos >> PAGE_CACHE_SHIFT;
		size_t bytes = min_t(size_t, PAGE_CACHE_SIZE - offset,
				     iov_iter_count(ii));

		bytes = min_t(size_t, bytes, fc->max_write - count);

 again:
		err = -EFAULT;
		if (iov_iter_fault_in_readable(ii, bytes))
			break;

		err = -ENOMEM;
801
		page = grab_cache_page_write_begin(mapping, index, 0);
Nick Piggin's avatar
Nick Piggin committed
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
		if (!page)
			break;

		pagefault_disable();
		tmp = iov_iter_copy_from_user_atomic(page, ii, offset, bytes);
		pagefault_enable();
		flush_dcache_page(page);

		if (!tmp) {
			unlock_page(page);
			page_cache_release(page);
			bytes = min(bytes, iov_iter_single_seg_count(ii));
			goto again;
		}

		err = 0;
		req->pages[req->num_pages] = page;
		req->num_pages++;

		iov_iter_advance(ii, tmp);
		count += tmp;
		pos += tmp;
		offset += tmp;
		if (offset == PAGE_CACHE_SIZE)
			offset = 0;

828
829
		if (!fc->big_writes)
			break;
Nick Piggin's avatar
Nick Piggin committed
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
	} while (iov_iter_count(ii) && count < fc->max_write &&
		 req->num_pages < FUSE_MAX_PAGES_PER_REQ && offset == 0);

	return count > 0 ? count : err;
}

static ssize_t fuse_perform_write(struct file *file,
				  struct address_space *mapping,
				  struct iov_iter *ii, loff_t pos)
{
	struct inode *inode = mapping->host;
	struct fuse_conn *fc = get_fuse_conn(inode);
	int err = 0;
	ssize_t res = 0;

	if (is_bad_inode(inode))
		return -EIO;

	do {
		struct fuse_req *req;
		ssize_t count;

		req = fuse_get_req(fc);
		if (IS_ERR(req)) {
			err = PTR_ERR(req);
			break;
		}

		count = fuse_fill_write_pages(req, mapping, ii, pos);
		if (count <= 0) {
			err = count;
		} else {
			size_t num_written;

			num_written = fuse_send_write_pages(req, file, inode,
							    pos, count);
			err = req->out.h.error;
			if (!err) {
				res += num_written;
				pos += num_written;

				/* break out of the loop on short write */
				if (num_written != count)
					err = -EIO;
			}
		}
		fuse_put_request(fc, req);
	} while (!err && iov_iter_count(ii));

	if (res > 0)
		fuse_write_update_size(inode, pos);

	fuse_invalidate_attr(inode);

	return res > 0 ? res : err;
}

static ssize_t fuse_file_aio_write(struct kiocb *iocb, const struct iovec *iov,
				   unsigned long nr_segs, loff_t pos)
{
	struct file *file = iocb->ki_filp;
	struct address_space *mapping = file->f_mapping;
	size_t count = 0;
	ssize_t written = 0;
	struct inode *inode = mapping->host;
	ssize_t err;
	struct iov_iter i;

	WARN_ON(iocb->ki_pos != pos);

	err = generic_segment_checks(iov, &nr_segs, &count, VERIFY_READ);
	if (err)
		return err;

	mutex_lock(&inode->i_mutex);
	vfs_check_frozen(inode->i_sb, SB_FREEZE_WRITE);

	/* We can write back this queue in page reclaim */
	current->backing_dev_info = mapping->backing_dev_info;

	err = generic_write_checks(file, &pos, &count, S_ISBLK(inode->i_mode));
	if (err)
		goto out;

	if (count == 0)
		goto out;

917
	err = file_remove_suid(file);
Nick Piggin's avatar
Nick Piggin committed
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
	if (err)
		goto out;

	file_update_time(file);

	iov_iter_init(&i, iov, nr_segs, count, 0);
	written = fuse_perform_write(file, mapping, &i, pos);
	if (written >= 0)
		iocb->ki_pos = pos + written;

out:
	current->backing_dev_info = NULL;
	mutex_unlock(&inode->i_mutex);

	return written ? written : err;
}

Miklos Szeredi's avatar
Miklos Szeredi committed
935
936
937
938
939
940
941
942
943
944
945
946
947
static void fuse_release_user_pages(struct fuse_req *req, int write)
{
	unsigned i;

	for (i = 0; i < req->num_pages; i++) {
		struct page *page = req->pages[i];
		if (write)
			set_page_dirty_lock(page);
		put_page(page);
	}
}

static int fuse_get_user_pages(struct fuse_req *req, const char __user *buf,
948
			       size_t *nbytesp, int write)
Miklos Szeredi's avatar
Miklos Szeredi committed
949
{
950
	size_t nbytes = *nbytesp;
Miklos Szeredi's avatar
Miklos Szeredi committed
951
952
953
954
	unsigned long user_addr = (unsigned long) buf;
	unsigned offset = user_addr & ~PAGE_MASK;
	int npages;

955
956
957
958
959
960
961
962
963
	/* Special case for kernel I/O: can copy directly into the buffer */
	if (segment_eq(get_fs(), KERNEL_DS)) {
		if (write)
			req->in.args[1].value = (void *) user_addr;
		else
			req->out.args[0].value = (void *) user_addr;

		return 0;
	}
Miklos Szeredi's avatar
Miklos Szeredi committed
964

965
	nbytes = min_t(size_t, nbytes, FUSE_MAX_PAGES_PER_REQ << PAGE_SHIFT);
Miklos Szeredi's avatar
Miklos Szeredi committed
966
	npages = (nbytes + offset + PAGE_SIZE - 1) >> PAGE_SHIFT;
967
	npages = clamp(npages, 1, FUSE_MAX_PAGES_PER_REQ);
Miklos Szeredi's avatar
Miklos Szeredi committed
968
	down_read(&current->mm->mmap_sem);
969
	npages = get_user_pages(current, current->mm, user_addr, npages, !write,
Miklos Szeredi's avatar
Miklos Szeredi committed
970
971
972
973
974
975
976
				0, req->pages, NULL);
	up_read(&current->mm->mmap_sem);
	if (npages < 0)
		return npages;

	req->num_pages = npages;
	req->page_offset = offset;
977
978
979
980
981
982
983
984
985

	if (write)
		req->in.argpages = 1;
	else
		req->out.argpages = 1;

	nbytes = (req->num_pages << PAGE_SHIFT) - req->page_offset;
	*nbytesp = min(*nbytesp, nbytes);

Miklos Szeredi's avatar
Miklos Szeredi committed
986
987
988
989
990
991
	return 0;
}

static ssize_t fuse_direct_io(struct file *file, const char __user *buf,
			      size_t count, loff_t *ppos, int write)
{
992
	struct inode *inode = file->f_path.dentry->d_inode;
Miklos Szeredi's avatar
Miklos Szeredi committed
993
994
995
996
	struct fuse_conn *fc = get_fuse_conn(inode);
	size_t nmax = write ? fc->max_write : fc->max_read;
	loff_t pos = *ppos;
	ssize_t res = 0;
997
998
999
1000
1001
	struct fuse_req *req;

	if (is_bad_inode(inode))
		return -EIO;

1002
1003
1004
	req = fuse_get_req(fc);
	if (IS_ERR(req))
		return PTR_ERR(req);
Miklos Szeredi's avatar
Miklos Szeredi committed
1005
1006
1007

	while (count) {
		size_t nres;
1008
1009
		size_t nbytes = min(count, nmax);
		int err = fuse_get_user_pages(req, buf, &nbytes, write);
Miklos Szeredi's avatar
Miklos Szeredi committed
1010
1011
1012
1013
		if (err) {
			res = err;
			break;
		}
1014

Miklos Szeredi's avatar
Miklos Szeredi committed
1015
		if (write)
1016
1017
			nres = fuse_send_write(req, file, inode, pos, nbytes,
					       current->files);
Miklos Szeredi's avatar
Miklos Szeredi committed
1018
		else
1019
1020
			nres = fuse_send_read(req, file, inode, pos, nbytes,
					      current->files);
Miklos Szeredi's avatar
Miklos Szeredi committed
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
		fuse_release_user_pages(req, !write);
		if (req->out.h.error) {
			if (!res)
				res = req->out.h.error;
			break;
		} else if (nres > nbytes) {
			res = -EIO;
			break;
		}
		count -= nres;
		res += nres;
		pos += nres;
		buf += nres;
		if (nres != nbytes)
			break;
1036
1037
1038
1039
1040
1041
		if (count) {
			fuse_put_request(fc, req);
			req = fuse_get_req(fc);
			if (IS_ERR(req))
				break;
		}
Miklos Szeredi's avatar
Miklos Szeredi committed
1042
1043
1044
	}
	fuse_put_request(fc, req);
	if (res > 0) {
1045
1046
		if (write)
			fuse_write_update_size(inode, pos);
Miklos Szeredi's avatar
Miklos Szeredi committed
1047
		*ppos = pos;
1048
1049
	}
	fuse_invalidate_attr(inode);
Miklos Szeredi's avatar
Miklos Szeredi committed
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062

	return res;
}

static ssize_t fuse_direct_read(struct file *file, char __user *buf,
				     size_t count, loff_t *ppos)
{
	return fuse_direct_io(file, buf, count, ppos, 0);
}

static ssize_t fuse_direct_write(struct file *file, const char __user *buf,
				 size_t count, loff_t *ppos)
{
1063
	struct inode *inode = file->f_path.dentry->d_inode;
Miklos Szeredi's avatar
Miklos Szeredi committed
1064
1065
	ssize_t res;
	/* Don't allow parallel writes to the same file */
1066
	mutex_lock(&inode->i_mutex);
1067
1068
1069
	res = generic_write_checks(file, ppos, &count, 0);
	if (!res)
		res = fuse_direct_io(file, buf, count, ppos, 1);
1070
	mutex_unlock(&inode->i_mutex);
Miklos Szeredi's avatar
Miklos Szeredi committed
1071
1072
1073
	return res;
}

Miklos Szeredi's avatar
Miklos Szeredi committed
1074
static void fuse_writepage_free(struct fuse_conn *fc, struct fuse_req *req)
1075
{
Miklos Szeredi's avatar
Miklos Szeredi committed
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
	__free_page(req->pages[0]);
	fuse_file_put(req->ff);
}

static void fuse_writepage_finish(struct fuse_conn *fc, struct fuse_req *req)
{
	struct inode *inode = req->inode;
	struct fuse_inode *fi = get_fuse_inode(inode);
	struct backing_dev_info *bdi = inode->i_mapping->backing_dev_info;

	list_del(&req->writepages_entry);
	dec_bdi_stat(bdi, BDI_WRITEBACK);
	dec_zone_page_state(req->pages[0], NR_WRITEBACK_TEMP);
	bdi_writeout_inc(bdi);
	wake_up(&fi->page_waitq);
}

/* Called under fc->lock, may release and reacquire it */
static void fuse_send_writepage(struct fuse_conn *fc, struct fuse_req *req)
1095
1096
__releases(&fc->lock)
__acquires(&fc->lock)
Miklos Szeredi's avatar
Miklos Szeredi committed
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
{
	struct fuse_inode *fi = get_fuse_inode(req->inode);
	loff_t size = i_size_read(req->inode);
	struct fuse_write_in *inarg = &req->misc.write.in;

	if (!fc->connected)
		goto out_free;

	if (inarg->offset + PAGE_CACHE_SIZE <= size) {
		inarg->size = PAGE_CACHE_SIZE;
	} else if (inarg->offset < size) {
		inarg->size = size & (PAGE_CACHE_SIZE - 1);
	} else {
		/* Got truncated off completely */
		goto out_free;
1112
	}
Miklos Szeredi's avatar
Miklos Szeredi committed
1113
1114
1115

	req->in.args[1].size = inarg->size;
	fi->writectr++;
1116
	fuse_request_send_background_locked(fc, req);
Miklos Szeredi's avatar
Miklos Szeredi committed
1117
1118
1119
1120
1121
1122
	return;

 out_free:
	fuse_writepage_finish(fc, req);
	spin_unlock(&fc->lock);
	fuse_writepage_free(fc, req);
1123
	fuse_put_request(fc, req);
Miklos Szeredi's avatar
Miklos Szeredi committed
1124
	spin_lock(&fc->lock);
1125
1126
}

Miklos Szeredi's avatar
Miklos Szeredi committed
1127
1128
1129
1130
1131
1132
1133
/*
 * If fi->writectr is positive (no truncate or fsync going on) send
 * all queued writepage requests.
 *
 * Called with fc->lock
 */
void fuse_flush_writepages(struct inode *inode)
1134
1135
__releases(&fc->lock)
__acquires(&fc->lock)
1136
{
Miklos Szeredi's avatar
Miklos Szeredi committed
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
	struct fuse_conn *fc = get_fuse_conn(inode);
	struct fuse_inode *fi = get_fuse_inode(inode);
	struct fuse_req *req;

	while (fi->writectr >= 0 && !list_empty(&fi->queued_writes)) {
		req = list_entry(fi->queued_writes.next, struct fuse_req, list);
		list_del_init(&req->list);
		fuse_send_writepage(fc, req);
	}
}

static void fuse_writepage_end(struct fuse_conn *fc, struct fuse_req *req)
{
	struct inode *inode = req->inode;
	struct fuse_inode *fi = get_fuse_inode(inode);

	mapping_set_error(inode->i_mapping, req->out.h.error);
	spin_lock(&fc->lock);
	fi->writectr--;
	fuse_writepage_finish(fc, req);
	spin_unlock(&fc->lock);
	fuse_writepage_free(fc, req);
}

static int fuse_writepage_locked(struct page *page)
{
	struct address_space *mapping = page->mapping;
	struct inode *inode = mapping->host;
	struct fuse_conn *fc = get_fuse_conn(inode);
	struct fuse_inode *fi = get_fuse_inode(inode);
	struct fuse_req *req;
	struct fuse_file *ff;
	struct page *tmp_page;

	set_page_writeback(page);

	req = fuse_request_alloc_nofs();
	if (!req)
		goto err;

	tmp_page = alloc_page(GFP_NOFS | __GFP_HIGHMEM);
	if (!tmp_page)
		goto err_free;

	spin_lock(&fc->lock);
	BUG_ON(list_empty(&fi->write_files));
	ff = list_entry(fi->write_files.next, struct fuse_file, write_entry);
	req->ff = fuse_file_get(ff);
	spin_unlock(&fc->lock);

	fuse_write_fill(req, NULL, ff, inode, page_offset(page), 0, 1);

	copy_highpage(tmp_page, page);
1190
	req->in.argpages = 1;
Miklos Szeredi's avatar
Miklos Szeredi committed
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
	req->num_pages = 1;
	req->pages[0] = tmp_page;
	req->page_offset = 0;
	req->end = fuse_writepage_end;
	req->inode = inode;

	inc_bdi_stat(mapping->backing_dev_info, BDI_WRITEBACK);
	inc_zone_page_state(tmp_page, NR_WRITEBACK_TEMP);
	end_page_writeback(page);

	spin_lock(&fc->lock);
	list_add(&req->writepages_entry, &fi->writepages);
	list_add_tail(&req->list, &fi->queued_writes);
	fuse_flush_writepages(inode);
	spin_unlock(&fc->lock);

	return 0;

err_free:
	fuse_request_free(req);
err:
	end_page_writeback(page);
	return -ENOMEM;
}

static int fuse_writepage(struct page *page, struct writeback_control *wbc)
{
	int err;

	err = fuse_writepage_locked(page);
	unlock_page(page);

	return err;
}

static int fuse_launder_page(struct page *page)
{
	int err = 0;
	if (clear_page_dirty_for_io(page)) {
		struct inode *inode = page->mapping->host;
		err = fuse_writepage_locked(page);
		if (!err)
			fuse_wait_on_page_writeback(inode, page->index);
	}
	return err;
}

/*
 * Write back dirty pages now, because there may not be any suitable
 * open files later
 */
static void fuse_vma_close(struct vm_area_struct *vma)
{
	filemap_write_and_wait(vma->vm_file->f_mapping);
}

/*
 * Wait for writeback against this page to complete before allowing it
 * to be marked dirty again, and hence written back again, possibly
 * before the previous writepage completed.
 *
 * Block here, instead of in ->writepage(), so that the userspace fs
 * can only block processes actually operating on the filesystem.
 *
 * Otherwise unprivileged userspace fs would be able to block
 * unrelated:
 *
 * - page migration
 * - sync(2)
 * - try_to_free_pages() with order > PAGE_ALLOC_COSTLY_ORDER
 */
1262
static int fuse_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
Miklos Szeredi's avatar
Miklos Szeredi committed
1263
{
1264
	struct page *page = vmf->page;
Miklos Szeredi's avatar
Miklos Szeredi committed
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
	/*
	 * Don't use page->mapping as it may become NULL from a
	 * concurrent truncate.
	 */
	struct inode *inode = vma->vm_file->f_mapping->host;

	fuse_wait_on_page_writeback(inode, page->index);
	return 0;
}

static struct vm_operations_struct fuse_file_vm_ops = {
	.close		= fuse_vma_close,
	.fault		= filemap_fault,
	.page_mkwrite	= fuse_page_mkwrite,
};

static int fuse_file_mmap(struct file *file, struct vm_area_struct *vma)
{
	if ((vma->vm_flags & VM_SHARED) && (vma->vm_flags & VM_MAYWRITE)) {
		struct inode *inode = file->f_dentry->d_inode;
		struct fuse_conn *fc = get_fuse_conn(inode);
		struct fuse_inode *fi = get_fuse_inode(inode);
		struct fuse_file *ff = file->private_data;
		/*
		 * file may be written through mmap, so chain it onto the
		 * inodes's write_file list
		 */
		spin_lock(&fc->lock);
		if (list_empty(&ff->write_entry))
			list_add(&ff->write_entry, &fi->write_files);
		spin_unlock(&fc->lock);
	}
	file_accessed(file);
	vma->vm_ops = &fuse_file_vm_ops;
1299
1300
1301
	return 0;
}

1302
1303
1304
1305
1306
1307
static int fuse_direct_mmap(struct file *file, struct vm_area_struct *vma)
{
	/* Can't provide the coherency needed for MAP_SHARED */
	if (vma->vm_flags & VM_MAYSHARE)
		return -ENODEV;

1308
1309
	invalidate_inode_pages2(file->f_mapping);

1310
1311
1312
	return generic_file_mmap(file, vma);
}

1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
static int convert_fuse_file_lock(const struct fuse_file_lock *ffl,
				  struct file_lock *fl)
{
	switch (ffl->type) {
	case F_UNLCK:
		break;

	case F_RDLCK:
	case F_WRLCK:
		if (ffl->start > OFFSET_MAX || ffl->end > OFFSET_MAX ||
		    ffl->end < ffl->start)
			return -EIO;

		fl->fl_start = ffl->start;
		fl->fl_end = ffl->end;
		fl->fl_pid = ffl->pid;
		break;

	default:
		return -EIO;
	}
	fl->fl_type = ffl->type;
	return 0;
}

static void fuse_lk_fill(struct fuse_req *req, struct file *file,
1339
1340
			 const struct file_lock *fl, int opcode, pid_t pid,
			 int flock)
1341
{
1342
	struct inode *inode = file->f_path.dentry->d_inode;
1343
	struct fuse_conn *fc = get_fuse_conn(inode);
1344
1345
1346
1347
	struct fuse_file *ff = file->private_data;
	struct fuse_lk_in *arg = &req->misc.lk_in;

	arg->fh = ff->fh;
1348
	arg->owner = fuse_lock_owner_id(fc, fl->fl_owner);
1349
1350
1351
1352
	arg->lk.start = fl->fl_start;
	arg->lk.end = fl->fl_end;
	arg->lk.type = fl->fl_type;
	arg->lk.pid = pid;
1353
1354
	if (flock)
		arg->lk_flags |= FUSE_LK_FLOCK;
1355
1356
1357
1358
1359
1360
1361
1362
1363
	req->in.h.opcode = opcode;
	req->in.h.nodeid = get_node_id(inode);
	req->in.numargs = 1;
	req->in.args[0].size = sizeof(*arg);
	req->in.args[0].value = arg;
}

static int fuse_getlk(struct file *file, struct file_lock *fl)
{
1364
	struct inode *inode = file->f_path.dentry->d_inode;
1365
1366
1367
1368
1369
1370
1371
1372
1373
	struct fuse_conn *fc = get_fuse_conn(inode);
	struct fuse_req *req;
	struct fuse_lk_out outarg;
	int err;

	req = fuse_get_req(fc);
	if (IS_ERR(req))
		return PTR_ERR(req);

1374
	fuse_lk_fill(req, file, fl, FUSE_GETLK, 0, 0);
1375
1376
1377
	req->out.numargs = 1;
	req->out.args[0].size = sizeof(outarg);
	req->out.args[0].value = &outarg;
1378
	fuse_request_send(fc, req);
1379
1380
1381
1382
1383
1384
1385
1386
	err = req->out.h.error;
	fuse_put_request(fc, req);
	if (!err)
		err = convert_fuse_file_lock(&outarg.lk, fl);

	return err;
}

1387
static int fuse_setlk(struct file *file, struct file_lock *fl, int flock)
1388
{