file.c 47.1 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
52
53
{
	struct fuse_file *ff;
	ff = kmalloc(sizeof(struct fuse_file), GFP_KERNEL);
	if (ff) {
54
55
		ff->reserved_req = fuse_request_alloc();
		if (!ff->reserved_req) {
56
			kfree(ff);
57
			return NULL;
58
59
60
		} else {
			INIT_LIST_HEAD(&ff->write_entry);
			atomic_set(&ff->count, 0);
Tejun Heo's avatar
Tejun Heo committed
61
62
63
			spin_lock(&fc->lock);
			ff->kh = ++fc->khctr;
			spin_unlock(&fc->lock);
64
		}
Tejun Heo's avatar
Tejun Heo committed
65
66
		RB_CLEAR_NODE(&ff->polled_node);
		init_waitqueue_head(&ff->poll_wait);
67
68
69
70
71
72
	}
	return ff;
}

void fuse_file_free(struct fuse_file *ff)
{
73
	fuse_request_free(ff->reserved_req);
74
75
76
	kfree(ff);
}

77
78
79
80
81
82
static struct fuse_file *fuse_file_get(struct fuse_file *ff)
{
	atomic_inc(&ff->count);
	return ff;
}

Miklos Szeredi's avatar
Miklos Szeredi committed
83
84
static void fuse_release_end(struct fuse_conn *fc, struct fuse_req *req)
{
85
86
	dput(req->misc.release.dentry);
	mntput(req->misc.release.vfsmount);
Miklos Szeredi's avatar
Miklos Szeredi committed
87
88
}

89
90
91
92
static void fuse_file_put(struct fuse_file *ff)
{
	if (atomic_dec_and_test(&ff->count)) {
		struct fuse_req *req = ff->reserved_req;
93
94
		struct inode *inode = req->misc.release.dentry->d_inode;
		struct fuse_conn *fc = get_fuse_conn(inode);
Miklos Szeredi's avatar
Miklos Szeredi committed
95
		req->end = fuse_release_end;
96
		fuse_request_send_background(fc, req);
97
98
99
100
		kfree(ff);
	}
}

101
102
103
104
105
106
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
107
		invalidate_inode_pages2(inode->i_mapping);
Tejun Heo's avatar
Tejun Heo committed
108
109
	if (outarg->open_flags & FOPEN_NONSEEKABLE)
		nonseekable_open(inode, file);
110
	ff->fh = outarg->fh;
111
	file->private_data = fuse_file_get(ff);
112
113
114
115
}

int fuse_open_common(struct inode *inode, struct file *file, int isdir)
{
Tejun Heo's avatar
Tejun Heo committed
116
	struct fuse_conn *fc = get_fuse_conn(inode);
117
118
119
120
	struct fuse_open_out outarg;
	struct fuse_file *ff;
	int err;

Miklos Szeredi's avatar
Miklos Szeredi committed
121
122
123
124
	/* VFS checks this, but only _after_ ->open() */
	if (file->f_flags & O_DIRECT)
		return -EINVAL;

125
126
127
128
	err = generic_file_open(inode, file);
	if (err)
		return err;

Tejun Heo's avatar
Tejun Heo committed
129
	ff = fuse_file_alloc(fc);
130
	if (!ff)
131
		return -ENOMEM;
132

133
134
135
136
137
138
139
	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);
140
141
142
143
144
	}

	return err;
}

145
void fuse_release_fill(struct fuse_file *ff, u64 nodeid, int flags, int opcode)
146
{
147
	struct fuse_req *req = ff->reserved_req;
148
	struct fuse_release_in *inarg = &req->misc.release.in;
149
150

	inarg->fh = ff->fh;
151
	inarg->flags = flags;
152
	req->in.h.opcode = opcode;
153
	req->in.h.nodeid = nodeid;
154
155
156
	req->in.numargs = 1;
	req->in.args[0].size = sizeof(struct fuse_release_in);
	req->in.args[0].value = inarg;
157
158
159
160
161
162
}

int fuse_release_common(struct inode *inode, struct file *file, int isdir)
{
	struct fuse_file *ff = file->private_data;
	if (ff) {
163
		struct fuse_conn *fc = get_fuse_conn(inode);
164
		struct fuse_req *req = ff->reserved_req;
165

166
167
		fuse_release_fill(ff, get_node_id(inode), file->f_flags,
				  isdir ? FUSE_RELEASEDIR : FUSE_RELEASE);
168
169

		/* Hold vfsmount and dentry until release is finished */
170
171
		req->misc.release.vfsmount = mntget(file->f_path.mnt);
		req->misc.release.dentry = dget(file->f_path.dentry);
172
173
174

		spin_lock(&fc->lock);
		list_del(&ff->write_entry);
Tejun Heo's avatar
Tejun Heo committed
175
176
		if (!RB_EMPTY_NODE(&ff->polled_node))
			rb_erase(&ff->polled_node, &fc->polled_files);
177
		spin_unlock(&fc->lock);
Tejun Heo's avatar
Tejun Heo committed
178
179

		wake_up_interruptible_sync(&ff->poll_wait);
180
181
182
183
184
185
		/*
		 * 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);
186
	}
187
188
189
190
191

	/* Return value is ignored by VFS */
	return 0;
}

192
193
194
195
196
197
198
199
200
201
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);
}

202
/*
203
204
 * Scramble the ID space with XTEA, so that the value of the files_struct
 * pointer is not exposed to userspace.
205
 */
206
u64 fuse_lock_owner_id(struct fuse_conn *fc, fl_owner_t id)
207
{
208
209
210
211
212
213
214
215
216
217
218
219
220
221
	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);
222
223
}

Miklos Szeredi's avatar
Miklos Szeredi committed
224
225
226
227
228
229
230
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
/*
 * 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;
}

267
static int fuse_flush(struct file *file, fl_owner_t id)
268
{
269
	struct inode *inode = file->f_path.dentry->d_inode;
270
271
272
273
274
275
	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;

276
277
278
	if (is_bad_inode(inode))
		return -EIO;

279
280
281
	if (fc->no_flush)
		return 0;

282
	req = fuse_get_req_nofail(fc, file);
283
284
	memset(&inarg, 0, sizeof(inarg));
	inarg.fh = ff->fh;
285
	inarg.lock_owner = fuse_lock_owner_id(fc, id);
286
287
288
289
290
	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;
291
	req->force = 1;
292
	fuse_request_send(fc, req);
293
294
295
296
297
298
299
300
301
	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
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
/*
 * 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);
}

317
318
int fuse_fsync_common(struct file *file, struct dentry *de, int datasync,
		      int isdir)
319
320
321
322
323
324
325
326
{
	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;

327
328
329
	if (is_bad_inode(inode))
		return -EIO;

330
	if ((!isdir && fc->no_fsync) || (isdir && fc->no_fsyncdir))
331
332
		return 0;

Miklos Szeredi's avatar
Miklos Szeredi committed
333
334
335
336
337
338
339
340
341
342
343
	/*
	 * 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);

344
345
346
	req = fuse_get_req(fc);
	if (IS_ERR(req))
		return PTR_ERR(req);
347
348
349
350

	memset(&inarg, 0, sizeof(inarg));
	inarg.fh = ff->fh;
	inarg.fsync_flags = datasync ? 1 : 0;
351
	req->in.h.opcode = isdir ? FUSE_FSYNCDIR : FUSE_FSYNC;
352
353
354
355
	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;
356
	fuse_request_send(fc, req);
357
358
359
	err = req->out.h.error;
	fuse_put_request(fc, req);
	if (err == -ENOSYS) {
360
361
362
363
		if (isdir)
			fc->no_fsyncdir = 1;
		else
			fc->no_fsync = 1;
364
365
366
367
368
		err = 0;
	}
	return err;
}

369
370
371
372
373
static int fuse_fsync(struct file *file, struct dentry *de, int datasync)
{
	return fuse_fsync_common(file, de, datasync, 0);
}

374
void fuse_read_fill(struct fuse_req *req, struct file *file,
375
		    struct inode *inode, loff_t pos, size_t count, int opcode)
376
{
377
	struct fuse_read_in *inarg = &req->misc.read.in;
378
	struct fuse_file *ff = file->private_data;
379

380
381
382
	inarg->fh = ff->fh;
	inarg->offset = pos;
	inarg->size = count;
383
	inarg->flags = file->f_flags;
384
	req->in.h.opcode = opcode;
385
386
387
	req->in.h.nodeid = get_node_id(inode);
	req->in.numargs = 1;
	req->in.args[0].size = sizeof(struct fuse_read_in);
388
	req->in.args[0].value = inarg;
389
390
391
392
393
394
	req->out.argpages = 1;
	req->out.argvar = 1;
	req->out.numargs = 1;
	req->out.args[0].size = count;
}

395
static size_t fuse_send_read(struct fuse_req *req, struct file *file,
396
397
			     struct inode *inode, loff_t pos, size_t count,
			     fl_owner_t owner)
398
{
399
	struct fuse_conn *fc = get_fuse_conn(inode);
400

401
	fuse_read_fill(req, file, inode, pos, count, FUSE_READ);
402
	if (owner != NULL) {
403
		struct fuse_read_in *inarg = &req->misc.read.in;
404
405
406
407

		inarg->read_flags |= FUSE_READ_LOCKOWNER;
		inarg->lock_owner = fuse_lock_owner_id(fc, owner);
	}
408
	fuse_request_send(fc, req);
409
	return req->out.args[0].size;
410
411
}

412
413
414
415
416
417
418
419
420
421
422
423
424
425
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);
}

426
427
428
429
static int fuse_readpage(struct file *file, struct page *page)
{
	struct inode *inode = page->mapping->host;
	struct fuse_conn *fc = get_fuse_conn(inode);
430
	struct fuse_req *req;
431
432
433
434
	size_t num_read;
	loff_t pos = page_offset(page);
	size_t count = PAGE_CACHE_SIZE;
	u64 attr_ver;
435
436
437
438
439
440
	int err;

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

Miklos Szeredi's avatar
Miklos Szeredi committed
441
442
443
444
445
446
447
	/*
	 * 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);

448
449
450
	req = fuse_get_req(fc);
	err = PTR_ERR(req);
	if (IS_ERR(req))
451
452
		goto out;

453
454
	attr_ver = fuse_get_attr_version(fc);

455
456
457
	req->out.page_zeroing = 1;
	req->num_pages = 1;
	req->pages[0] = page;
458
	num_read = fuse_send_read(req, file, inode, pos, count, NULL);
459
460
	err = req->out.h.error;
	fuse_put_request(fc, req);
461
462
463
464
465
466
467
468

	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);

469
		SetPageUptodate(page);
470
471
	}

472
	fuse_invalidate_attr(inode); /* atime changed */
473
474
475
476
477
 out:
	unlock_page(page);
	return err;
}

478
static void fuse_readpages_end(struct fuse_conn *fc, struct fuse_req *req)
479
{
480
	int i;
481
482
483
	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;
484

485
486
487
488
489
490
491
492
493
	/*
	 * 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 */
494

495
496
497
498
	for (i = 0; i < req->num_pages; i++) {
		struct page *page = req->pages[i];
		if (!req->out.h.error)
			SetPageUptodate(page);
499
500
		else
			SetPageError(page);
501
502
		unlock_page(page);
	}
503
504
	if (req->ff)
		fuse_file_put(req->ff);
505
506
}

507
static void fuse_send_readpages(struct fuse_req *req, struct file *file,
508
509
510
511
512
513
				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;
	req->out.page_zeroing = 1;
514
	fuse_read_fill(req, file, inode, pos, count, FUSE_READ);
515
	req->misc.read.attr_ver = fuse_get_attr_version(fc);
516
	if (fc->async_read) {
517
		struct fuse_file *ff = file->private_data;
518
		req->ff = fuse_file_get(ff);
519
		req->end = fuse_readpages_end;
520
		fuse_request_send_background(fc, req);
521
	} else {
522
		fuse_request_send(fc, req);
523
		fuse_readpages_end(fc, req);
524
		fuse_put_request(fc, req);
525
	}
526
527
}

528
struct fuse_fill_data {
529
	struct fuse_req *req;
530
	struct file *file;
531
532
533
534
535
	struct inode *inode;
};

static int fuse_readpages_fill(void *_data, struct page *page)
{
536
	struct fuse_fill_data *data = _data;
537
538
539
540
	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
541
542
	fuse_wait_on_page_writeback(inode, page->index);

543
544
545
546
	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)) {
547
		fuse_send_readpages(req, data->file, inode);
548
549
		data->req = req = fuse_get_req(fc);
		if (IS_ERR(req)) {
550
			unlock_page(page);
551
			return PTR_ERR(req);
552
553
554
		}
	}
	req->pages[req->num_pages] = page;
Miklos Szeredi's avatar
Miklos Szeredi committed
555
	req->num_pages++;
556
557
558
559
560
561
562
563
	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);
564
	struct fuse_fill_data data;
565
	int err;
566

567
	err = -EIO;
568
	if (is_bad_inode(inode))
569
		goto out;
570

571
	data.file = file;
572
	data.inode = inode;
573
	data.req = fuse_get_req(fc);
574
	err = PTR_ERR(data.req);
575
	if (IS_ERR(data.req))
576
		goto out;
577
578

	err = read_cache_pages(mapping, pages, fuse_readpages_fill, &data);
579
580
	if (!err) {
		if (data.req->num_pages)
581
			fuse_send_readpages(data.req, file, inode);
582
583
584
		else
			fuse_put_request(fc, data.req);
	}
585
out:
586
	return err;
587
588
}

Miklos Szeredi's avatar
Miklos Szeredi committed
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
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);
}

608
static void fuse_write_fill(struct fuse_req *req, struct file *file,
Miklos Szeredi's avatar
Miklos Szeredi committed
609
610
			    struct fuse_file *ff, struct inode *inode,
			    loff_t pos, size_t count, int writepage)
611
{
612
	struct fuse_conn *fc = get_fuse_conn(inode);
613
614
	struct fuse_write_in *inarg = &req->misc.write.in;
	struct fuse_write_out *outarg = &req->misc.write.out;
615

616
617
618
619
620
	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
621
	inarg->flags = file ? file->f_flags : 0;
622
623
624
625
	req->in.h.opcode = FUSE_WRITE;
	req->in.h.nodeid = get_node_id(inode);
	req->in.argpages = 1;
	req->in.numargs = 2;
626
627
628
629
	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);
630
	req->in.args[0].value = inarg;
631
632
633
	req->in.args[1].size = count;
	req->out.numargs = 1;
	req->out.args[0].size = sizeof(struct fuse_write_out);
634
635
636
637
	req->out.args[0].value = outarg;
}

static size_t fuse_send_write(struct fuse_req *req, struct file *file,
638
639
			      struct inode *inode, loff_t pos, size_t count,
			      fl_owner_t owner)
640
641
{
	struct fuse_conn *fc = get_fuse_conn(inode);
Miklos Szeredi's avatar
Miklos Szeredi committed
642
	fuse_write_fill(req, file, file->private_data, inode, pos, count, 0);
643
644
645
646
647
	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);
	}
648
	fuse_request_send(fc, req);
649
	return req->misc.write.out.size;
650
651
}

Nick Piggin's avatar
Nick Piggin committed
652
653
654
static int fuse_write_begin(struct file *file, struct address_space *mapping,
			loff_t pos, unsigned len, unsigned flags,
			struct page **pagep, void **fsdata)
655
{
Nick Piggin's avatar
Nick Piggin committed
656
657
	pgoff_t index = pos >> PAGE_CACHE_SHIFT;

658
	*pagep = grab_cache_page_write_begin(mapping, index, flags);
Nick Piggin's avatar
Nick Piggin committed
659
660
	if (!*pagep)
		return -ENOMEM;
661
662
663
	return 0;
}

664
665
666
667
668
669
670
671
672
673
674
675
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
676
677
static int fuse_buffered_write(struct file *file, struct inode *inode,
			       loff_t pos, unsigned count, struct page *page)
678
679
{
	int err;
680
	size_t nres;
681
	struct fuse_conn *fc = get_fuse_conn(inode);
Nick Piggin's avatar
Nick Piggin committed
682
	unsigned offset = pos & (PAGE_CACHE_SIZE - 1);
683
684
685
686
687
	struct fuse_req *req;

	if (is_bad_inode(inode))
		return -EIO;

Miklos Szeredi's avatar
Miklos Szeredi committed
688
689
690
691
692
693
	/*
	 * Make sure writepages on the same page are not mixed up with
	 * plain writes.
	 */
	fuse_wait_on_page_writeback(inode, page->index);

694
695
696
	req = fuse_get_req(fc);
	if (IS_ERR(req))
		return PTR_ERR(req);
697
698
699
700

	req->num_pages = 1;
	req->pages[0] = page;
	req->page_offset = offset;
701
	nres = fuse_send_write(req, file, inode, pos, count, NULL);
702
703
	err = req->out.h.error;
	fuse_put_request(fc, req);
Nick Piggin's avatar
Nick Piggin committed
704
	if (!err && !nres)
705
706
		err = -EIO;
	if (!err) {
Nick Piggin's avatar
Nick Piggin committed
707
		pos += nres;
708
		fuse_write_update_size(inode, pos);
Nick Piggin's avatar
Nick Piggin committed
709
		if (count == PAGE_CACHE_SIZE)
710
			SetPageUptodate(page);
711
712
	}
	fuse_invalidate_attr(inode);
Nick Piggin's avatar
Nick Piggin committed
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
	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;
729
730
}

Nick Piggin's avatar
Nick Piggin committed
731
732
733
734
735
736
737
738
739
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
783
784
785
786
787
788
789
790
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;

	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;
791
		page = grab_cache_page_write_begin(mapping, index, 0);
Nick Piggin's avatar
Nick Piggin committed
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
		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;

818
819
		if (!fc->big_writes)
			break;
Nick Piggin's avatar
Nick Piggin committed
820
821
822
823
824
825
826
827
828
829
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
	} 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;

907
	err = file_remove_suid(file);
Nick Piggin's avatar
Nick Piggin committed
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
	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
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
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,
			       unsigned nbytes, int write)
{
	unsigned long user_addr = (unsigned long) buf;
	unsigned offset = user_addr & ~PAGE_MASK;
	int npages;

	/* This doesn't work with nfsd */
	if (!current->mm)
		return -EPERM;

	nbytes = min(nbytes, (unsigned) FUSE_MAX_PAGES_PER_REQ << PAGE_SHIFT);
	npages = (nbytes + offset + PAGE_SIZE - 1) >> PAGE_SHIFT;
950
	npages = clamp(npages, 1, FUSE_MAX_PAGES_PER_REQ);
Miklos Szeredi's avatar
Miklos Szeredi committed
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
	down_read(&current->mm->mmap_sem);
	npages = get_user_pages(current, current->mm, user_addr, npages, write,
				0, req->pages, NULL);
	up_read(&current->mm->mmap_sem);
	if (npages < 0)
		return npages;

	req->num_pages = npages;
	req->page_offset = offset;
	return 0;
}

static ssize_t fuse_direct_io(struct file *file, const char __user *buf,
			      size_t count, loff_t *ppos, int write)
{
966
	struct inode *inode = file->f_path.dentry->d_inode;
Miklos Szeredi's avatar
Miklos Szeredi committed
967
968
969
970
	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;
971
972
973
974
975
	struct fuse_req *req;

	if (is_bad_inode(inode))
		return -EIO;

976
977
978
	req = fuse_get_req(fc);
	if (IS_ERR(req))
		return PTR_ERR(req);
Miklos Szeredi's avatar
Miklos Szeredi committed
979
980
981

	while (count) {
		size_t nres;
982
983
984
		size_t nbytes_limit = min(count, nmax);
		size_t nbytes;
		int err = fuse_get_user_pages(req, buf, nbytes_limit, !write);
Miklos Szeredi's avatar
Miklos Szeredi committed
985
986
987
988
		if (err) {
			res = err;
			break;
		}
989
		nbytes = (req->num_pages << PAGE_SHIFT) - req->page_offset;
990
		nbytes = min(nbytes_limit, nbytes);
Miklos Szeredi's avatar
Miklos Szeredi committed
991
		if (write)
992
993
			nres = fuse_send_write(req, file, inode, pos, nbytes,
					       current->files);
Miklos Szeredi's avatar
Miklos Szeredi committed
994
		else
995
996
			nres = fuse_send_read(req, file, inode, pos, nbytes,
					      current->files);
Miklos Szeredi's avatar
Miklos Szeredi committed
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
		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;
1012
1013
1014
1015
1016
1017
		if (count) {
			fuse_put_request(fc, req);
			req = fuse_get_req(fc);
			if (IS_ERR(req))
				break;
		}
Miklos Szeredi's avatar
Miklos Szeredi committed
1018
1019
1020
	}
	fuse_put_request(fc, req);
	if (res > 0) {
1021
1022
		if (write)
			fuse_write_update_size(inode, pos);
Miklos Szeredi's avatar
Miklos Szeredi committed
1023
		*ppos = pos;
1024
1025
	}
	fuse_invalidate_attr(inode);
Miklos Szeredi's avatar
Miklos Szeredi committed
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038

	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)
{
1039
	struct inode *inode = file->f_path.dentry->d_inode;
Miklos Szeredi's avatar
Miklos Szeredi committed
1040
1041
	ssize_t res;
	/* Don't allow parallel writes to the same file */
1042
	mutex_lock(&inode->i_mutex);
1043
1044
1045
	res = generic_write_checks(file, ppos, &count, 0);
	if (!res)
		res = fuse_direct_io(file, buf, count, ppos, 1);
1046
	mutex_unlock(&inode->i_mutex);
Miklos Szeredi's avatar
Miklos Szeredi committed
1047
1048
1049
	return res;
}

Miklos Szeredi's avatar
Miklos Szeredi committed
1050
static void fuse_writepage_free(struct fuse_conn *fc, struct fuse_req *req)
1051
{
Miklos Szeredi's avatar
Miklos Szeredi committed
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
	__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)
1071
1072
__releases(&fc->lock)
__acquires(&fc->lock)
Miklos Szeredi's avatar
Miklos Szeredi committed
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
{
	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;
1088
	}
Miklos Szeredi's avatar
Miklos Szeredi committed
1089
1090
1091

	req->in.args[1].size = inarg->size;
	fi->writectr++;
1092
	fuse_request_send_background_locked(fc, req);
Miklos Szeredi's avatar
Miklos Szeredi committed
1093
1094
1095
1096
1097
1098
	return;

 out_free:
	fuse_writepage_finish(fc, req);
	spin_unlock(&fc->lock);
	fuse_writepage_free(fc, req);
1099
	fuse_put_request(fc, req);
Miklos Szeredi's avatar
Miklos Szeredi committed
1100
	spin_lock(&fc->lock);
1101
1102
}

Miklos Szeredi's avatar
Miklos Szeredi committed
1103
1104
1105
1106
1107
1108
1109
/*
 * 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)
1110
1111
__releases(&fc->lock)
__acquires(&fc->lock)
1112
{
Miklos Szeredi's avatar
Miklos Szeredi committed
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
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
1190
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
	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);
	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
 */
1237
static int fuse_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
Miklos Szeredi's avatar
Miklos Szeredi committed
1238
{
1239
	struct page *page = vmf->page;
Miklos Szeredi's avatar
Miklos Szeredi committed
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
	/*
	 * 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;
1274
1275
1276
	return 0;
}

1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
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,
1303
1304
			 const struct file_lock *fl, int opcode, pid_t pid,
			 int flock)
1305
{
1306
	struct inode *inode = file->f_path.dentry->d_inode;
1307
	struct fuse_conn *fc = get_fuse_conn(inode);
1308
1309
1310
1311
	struct fuse_file *ff = file->private_data;
	struct fuse_lk_in *arg = &req->misc.lk_in;

	arg->fh = ff->fh;
1312
	arg->owner = fuse_lock_owner_id(fc, fl->fl_owner);
1313
1314
1315
1316
	arg->lk.start = fl->fl_start;
	arg->lk.end = fl->fl_end;
	arg->lk.type = fl->fl_type;
	arg->lk.pid = pid;
1317
1318
	if (flock)
		arg->lk_flags |= FUSE_LK_FLOCK;
1319
1320
1321
1322
1323
1324
1325
1326
1327
	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)
{
1328
	struct inode *inode = file->f_path.dentry->d_inode;
1329
1330
1331
1332
1333
1334
1335
1336
1337
	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);

1338
	fuse_lk_fill(req, file, fl, FUSE_GETLK, 0, 0);
1339
1340
1341
	req->out.numargs = 1;
	req->out.args[0].size = sizeof(outarg);
	req->out.args[0].value = &outarg;
1342
	fuse_request_send(fc, req);
1343
1344
1345
1346
1347
1348
1349
1350
	err = req->out.h.error;
	fuse_put_request(fc, req);
	if (!err)
		err = convert_fuse_file_lock(&outarg.lk, fl);

	return err;
}

1351
static int fuse_setlk(struct file *file, struct file_lock *fl, int flock)
1352
{
1353
	struct inode *inode = file->f_path.dentry->d_inode;
1354
1355
1356
1357
1358
1359
	struct fuse_conn *fc = get_fuse_conn(inode);
	struct fuse_req *req;
	int opcode = (fl->fl_flags & FL_SLEEP) ? FUSE_SETLKW : FUSE_SETLK;
	pid_t pid = fl->fl_type != F_UNLCK ? current->tgid : 0;
	int err;

Miklos Szeredi's avatar
Miklos Szeredi committed
1360
1361
1362
1363
1364
	if (fl->fl_lmops && fl->fl_lmops->fl_grant) {
		/* NLM needs asynchronous locks, which we don't support yet */
		return -ENOLCK;
	}

1365
1366
1367
1368
1369
1370
1371
1372
	/* Unlock on close is handled by the flush method */
	if (fl->fl_flags & FL_CLOSE)
		return 0;

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

1373
	fuse_lk_fill(req, file, fl, opcode, pid, flock);
1374
	fuse_request_send(fc, req);
1375
	err = req->out.h.error;
1376
1377
1378
	/* locking is restartable */
	if (err == -EINTR)
		err = -ERESTARTSYS;
1379
1380
1381
1382
1383
1384
	fuse_put_request(fc, req);
	return err;
}

static int fuse_file_lock(struct file *file, int cmd, struct file_lock *fl)
{
1385
	struct inode *inode = file->f_path.dentry->d_inode;
1386
1387
1388
	struct fuse_conn *fc = get_fuse_conn(inode);
	int err;

Miklos Szeredi's avatar
Miklos Szeredi committed
1389
1390
1391