summaryrefslogtreecommitdiffstats
path: root/fs/ceph/quota.c
blob: 9b785f11e95ac75f24b206100f32ecef741b4988 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
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
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
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
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
// SPDX-License-Identifier: GPL-2.0
/*
 * quota.c - CephFS quota
 *
 * Copyright (C) 2017-2018 SUSE
 */

#include <linux/statfs.h>

#include "super.h"
#include "mds_client.h"

void ceph_adjust_quota_realms_count(struct inode *inode, bool inc)
{
	struct ceph_mds_client *mdsc = ceph_sb_to_mdsc(inode->i_sb);
	if (inc)
		atomic64_inc(&mdsc->quotarealms_count);
	else
		atomic64_dec(&mdsc->quotarealms_count);
}

static inline bool ceph_has_realms_with_quotas(struct inode *inode)
{
	struct super_block *sb = inode->i_sb;
	struct ceph_mds_client *mdsc = ceph_sb_to_mdsc(sb);
	struct inode *root = d_inode(sb->s_root);

	if (atomic64_read(&mdsc->quotarealms_count) > 0)
		return true;
	/* if root is the real CephFS root, we don't have quota realms */
	if (root && ceph_ino(root) == CEPH_INO_ROOT)
		return false;
	/* otherwise, we can't know for sure */
	return true;
}

void ceph_handle_quota(struct ceph_mds_client *mdsc,
		       struct ceph_mds_session *session,
		       struct ceph_msg *msg)
{
	struct super_block *sb = mdsc->fsc->sb;
	struct ceph_mds_quota *h = msg->front.iov_base;
	struct ceph_vino vino;
	struct inode *inode;
	struct ceph_inode_info *ci;

	if (msg->front.iov_len < sizeof(*h)) {
		pr_err("%s corrupt message mds%d len %d\n", __func__,
		       session->s_mds, (int)msg->front.iov_len);
		ceph_msg_dump(msg);
		return;
	}

	/* increment msg sequence number */
	mutex_lock(&session->s_mutex);
	inc_session_sequence(session);
	mutex_unlock(&session->s_mutex);

	/* lookup inode */
	vino.ino = le64_to_cpu(h->ino);
	vino.snap = CEPH_NOSNAP;
	inode = ceph_find_inode(sb, vino);
	if (!inode) {
		pr_warn("Failed to find inode %llu\n", vino.ino);
		return;
	}
	ci = ceph_inode(inode);

	spin_lock(&ci->i_ceph_lock);
	ci->i_rbytes = le64_to_cpu(h->rbytes);
	ci->i_rfiles = le64_to_cpu(h->rfiles);
	ci->i_rsubdirs = le64_to_cpu(h->rsubdirs);
	__ceph_update_quota(ci, le64_to_cpu(h->max_bytes),
		            le64_to_cpu(h->max_files));
	spin_unlock(&ci->i_ceph_lock);

	/* avoid calling iput_final() in dispatch thread */
	ceph_async_iput(inode);
}

static struct ceph_quotarealm_inode *
find_quotarealm_inode(struct ceph_mds_client *mdsc, u64 ino)
{
	struct ceph_quotarealm_inode *qri = NULL;
	struct rb_node **node, *parent = NULL;

	mutex_lock(&mdsc->quotarealms_inodes_mutex);
	node = &(mdsc->quotarealms_inodes.rb_node);
	while (*node) {
		parent = *node;
		qri = container_of(*node, struct ceph_quotarealm_inode, node);

		if (ino < qri->ino)
			node = &((*node)->rb_left);
		else if (ino > qri->ino)
			node = &((*node)->rb_right);
		else
			break;
	}
	if (!qri || (qri->ino != ino)) {
		/* Not found, create a new one and insert it */
		qri = kmalloc(sizeof(*qri), GFP_KERNEL);
		if (qri) {
			qri->ino = ino;
			qri->inode = NULL;
			qri->timeout = 0;
			mutex_init(&qri->mutex);
			rb_link_node(&qri->node, parent, node);
			rb_insert_color(&qri->node, &mdsc->quotarealms_inodes);
		} else
			pr_warn("Failed to alloc quotarealms_inode\n");
	}
	mutex_unlock(&mdsc->quotarealms_inodes_mutex);

	return qri;
}

/*
 * This function will try to lookup a realm inode which isn't visible in the
 * filesystem mountpoint.  A list of these kind of inodes (not visible) is
 * maintained in the mdsc and freed only when the filesystem is umounted.
 *
 * Note that these inodes are kept in this list even if the lookup fails, which
 * allows to prevent useless lookup requests.
 */
static struct inode *lookup_quotarealm_inode(struct ceph_mds_client *mdsc,
					     struct super_block *sb,
					     struct ceph_snap_realm *realm)
{
	struct ceph_quotarealm_inode *qri;
	struct inode *in;

	qri = find_quotarealm_inode(mdsc, realm->ino);
	if (!qri)
		return NULL;

	mutex_lock(&qri->mutex);
	if (qri->inode && ceph_is_any_caps(qri->inode)) {
		/* A request has already returned the inode */
		mutex_unlock(&qri->mutex);
		return qri->inode;
	}
	/* Check if this inode lookup has failed recently */
	if (qri->timeout &&
	    time_before_eq(jiffies, qri->timeout)) {
		mutex_unlock(&qri->mutex);
		return NULL;
	}
	if (qri->inode) {
		/* get caps */
		int ret = __ceph_do_getattr(qri->inode, NULL,
					    CEPH_STAT_CAP_INODE, true);
		if (ret >= 0)
			in = qri->inode;
		else
			in = ERR_PTR(ret);
	}  else {
		in = ceph_lookup_inode(sb, realm->ino);
	}

	if (IS_ERR(in)) {
		dout("Can't lookup inode %llx (err: %ld)\n",
		     realm->ino, PTR_ERR(in));
		qri->timeout = jiffies + msecs_to_jiffies(60 * 1000); /* XXX */
	} else {
		qri->timeout = 0;
		qri->inode = in;
	}
	mutex_unlock(&qri->mutex);

	return in;
}

void ceph_cleanup_quotarealms_inodes(struct ceph_mds_client *mdsc)
{
	struct ceph_quotarealm_inode *qri;
	struct rb_node *node;

	/*
	 * It should now be safe to clean quotarealms_inode tree without holding
	 * mdsc->quotarealms_inodes_mutex...
	 */
	mutex_lock(&mdsc->quotarealms_inodes_mutex);
	while (!RB_EMPTY_ROOT(&mdsc->quotarealms_inodes)) {
		node = rb_first(&mdsc->quotarealms_inodes);
		qri = rb_entry(node, struct ceph_quotarealm_inode, node);
		rb_erase(node, &mdsc->quotarealms_inodes);
		iput(qri->inode);
		kfree(qri);
	}
	mutex_unlock(&mdsc->quotarealms_inodes_mutex);
}

/*
 * This function walks through the snaprealm for an inode and returns the
 * ceph_snap_realm for the first snaprealm that has quotas set (either max_files
 * or max_bytes).  If the root is reached, return the root ceph_snap_realm
 * instead.
 *
 * Note that the caller is responsible for calling ceph_put_snap_realm() on the
 * returned realm.
 *
 * Callers of this function need to hold mdsc->snap_rwsem.  However, if there's
 * a need to do an inode lookup, this rwsem will be temporarily dropped.  Hence
 * the 'retry' argument: if rwsem needs to be dropped and 'retry' is 'false'
 * this function will return -EAGAIN; otherwise, the snaprealms walk-through
 * will be restarted.
 */
static struct ceph_snap_realm *get_quota_realm(struct ceph_mds_client *mdsc,
					       struct inode *inode, bool retry)
{
	struct ceph_inode_info *ci = NULL;
	struct ceph_snap_realm *realm, *next;
	struct inode *in;
	bool has_quota;

	if (ceph_snap(inode) != CEPH_NOSNAP)
		return NULL;

restart:
	realm = ceph_inode(inode)->i_snap_realm;
	if (realm)
		ceph_get_snap_realm(mdsc, realm);
	else
		pr_err_ratelimited("get_quota_realm: ino (%llx.%llx) "
				   "null i_snap_realm\n", ceph_vinop(inode));
	while (realm) {
		bool has_inode;

		spin_lock(&realm->inodes_with_caps_lock);
		has_inode = realm->inode;
		in = has_inode ? igrab(realm->inode) : NULL;
		spin_unlock(&realm->inodes_with_caps_lock);
		if (has_inode && !in)
			break;
		if (!in) {
			up_read(&mdsc->snap_rwsem);
			in = lookup_quotarealm_inode(mdsc, inode->i_sb, realm);
			down_read(&mdsc->snap_rwsem);
			if (IS_ERR_OR_NULL(in))
				break;
			ceph_put_snap_realm(mdsc, realm);
			if (!retry)
				return ERR_PTR(-EAGAIN);
			goto restart;
		}

		ci = ceph_inode(in);
		has_quota = __ceph_has_any_quota(ci);
		/* avoid calling iput_final() while holding mdsc->snap_rwsem */
		ceph_async_iput(in);

		next = realm->parent;
		if (has_quota || !next)
		       return realm;

		ceph_get_snap_realm(mdsc, next);
		ceph_put_snap_realm(mdsc, realm);
		realm = next;
	}
	if (realm)
		ceph_put_snap_realm(mdsc, realm);

	return NULL;
}

static bool ceph_quota_is_same_realm(struct inode *old, struct inode *new)
{
	struct ceph_mds_client *mdsc = ceph_sb_to_mdsc(old->i_sb);
	struct ceph_snap_realm *old_realm, *new_realm;
	bool is_same;

restart:
	/*
	 * We need to lookup 2 quota realms atomically, i.e. with snap_rwsem.
	 * However, get_quota_realm may drop it temporarily.  By setting the
	 * 'retry' parameter to 'false', we'll get -EAGAIN if the rwsem was
	 * dropped and we can then restart the whole operation.
	 */
	down_read(&mdsc->snap_rwsem);
	old_realm = get_quota_realm(mdsc, old, true);
	new_realm = get_quota_realm(mdsc, new, false);
	if (PTR_ERR(new_realm) == -EAGAIN) {
		up_read(&mdsc->snap_rwsem);
		if (old_realm)
			ceph_put_snap_realm(mdsc, old_realm);
		goto restart;
	}
	is_same = (old_realm == new_realm);
	up_read(&mdsc->snap_rwsem);

	if (old_realm)
		ceph_put_snap_realm(mdsc, old_realm);
	if (new_realm)
		ceph_put_snap_realm(mdsc, new_realm);

	return is_same;
}

enum quota_check_op {
	QUOTA_CHECK_MAX_FILES_OP,	/* check quota max_files limit */
	QUOTA_CHECK_MAX_BYTES_OP,	/* check quota max_files limit */
	QUOTA_CHECK_MAX_BYTES_APPROACHING_OP	/* check if quota max_files
						   limit is approaching */
};

/*
 * check_quota_exceeded() will walk up the snaprealm hierarchy and, for each
 * realm, it will execute quota check operation defined by the 'op' parameter.
 * The snaprealm walk is interrupted if the quota check detects that the quota
 * is exceeded or if the root inode is reached.
 */
static bool check_quota_exceeded(struct inode *inode, enum quota_check_op op,
				 loff_t delta)
{
	struct ceph_mds_client *mdsc = ceph_sb_to_mdsc(inode->i_sb);
	struct ceph_inode_info *ci;
	struct ceph_snap_realm *realm, *next;
	struct inode *in;
	u64 max, rvalue;
	bool exceeded = false;

	if (ceph_snap(inode) != CEPH_NOSNAP)
		return false;

	down_read(&mdsc->snap_rwsem);
restart:
	realm = ceph_inode(inode)->i_snap_realm;
	if (realm)
		ceph_get_snap_realm(mdsc, realm);
	else
		pr_err_ratelimited("check_quota_exceeded: ino (%llx.%llx) "
				   "null i_snap_realm\n", ceph_vinop(inode));
	while (realm) {
		bool has_inode;

		spin_lock(&realm->inodes_with_caps_lock);
		has_inode = realm->inode;
		in = has_inode ? igrab(realm->inode) : NULL;
		spin_unlock(&realm->inodes_with_caps_lock);
		if (has_inode && !in)
			break;
		if (!in) {
			up_read(&mdsc->snap_rwsem);
			in = lookup_quotarealm_inode(mdsc, inode->i_sb, realm);
			down_read(&mdsc->snap_rwsem);
			if (IS_ERR_OR_NULL(in))
				break;
			ceph_put_snap_realm(mdsc, realm);
			goto restart;
		}
		ci = ceph_inode(in);
		spin_lock(&ci->i_ceph_lock);
		if (op == QUOTA_CHECK_MAX_FILES_OP) {
			max = ci->i_max_files;
			rvalue = ci->i_rfiles + ci->i_rsubdirs;
		} else {
			max = ci->i_max_bytes;
			rvalue = ci->i_rbytes;
		}
		spin_unlock(&ci->i_ceph_lock);
		switch (op) {
		case QUOTA_CHECK_MAX_FILES_OP:
		case QUOTA_CHECK_MAX_BYTES_OP:
			exceeded = (max && (rvalue + delta > max));
			break;
		case QUOTA_CHECK_MAX_BYTES_APPROACHING_OP:
			if (max) {
				if (rvalue >= max)
					exceeded = true;
				else {
					/*
					 * when we're writing more that 1/16th
					 * of the available space
					 */
					exceeded =
						(((max - rvalue) >> 4) < delta);
				}
			}
			break;
		default:
			/* Shouldn't happen */
			pr_warn("Invalid quota check op (%d)\n", op);
			exceeded = true; /* Just break the loop */
		}
		/* avoid calling iput_final() while holding mdsc->snap_rwsem */
		ceph_async_iput(in);

		next = realm->parent;
		if (exceeded || !next)
			break;
		ceph_get_snap_realm(mdsc, next);
		ceph_put_snap_realm(mdsc, realm);
		realm = next;
	}
	if (realm)
		ceph_put_snap_realm(mdsc, realm);
	up_read(&mdsc->snap_rwsem);

	return exceeded;
}

/*
 * ceph_quota_is_max_files_exceeded - check if we can create a new file
 * @inode:	directory where a new file is being created
 *
 * This functions returns true is max_files quota allows a new file to be
 * created.  It is necessary to walk through the snaprealm hierarchy (until the
 * FS root) to check all realms with quotas set.
 */
bool ceph_quota_is_max_files_exceeded(struct inode *inode)
{
	if (!ceph_has_realms_with_quotas(inode))
		return false;

	WARN_ON(!S_ISDIR(inode->i_mode));

	return check_quota_exceeded(inode, QUOTA_CHECK_MAX_FILES_OP, 1);
}

/*
 * ceph_quota_is_max_bytes_exceeded - check if we can write to a file
 * @inode:	inode being written
 * @newsize:	new size if write succeeds
 *
 * This functions returns true is max_bytes quota allows a file size to reach
 * @newsize; it returns false otherwise.
 */
bool ceph_quota_is_max_bytes_exceeded(struct inode *inode, loff_t newsize)
{
	loff_t size = i_size_read(inode);

	if (!ceph_has_realms_with_quotas(inode))
		return false;

	/* return immediately if we're decreasing file size */
	if (newsize <= size)
		return false;

	return check_quota_exceeded(inode, QUOTA_CHECK_MAX_BYTES_OP, (newsize - size));
}

/*
 * ceph_quota_is_max_bytes_approaching - check if we're reaching max_bytes
 * @inode:	inode being written
 * @newsize:	new size if write succeeds
 *
 * This function returns true if the new file size @newsize will be consuming
 * more than 1/16th of the available quota space; it returns false otherwise.
 */
bool ceph_quota_is_max_bytes_approaching(struct inode *inode, loff_t newsize)
{
	loff_t size = ceph_inode(inode)->i_reported_size;

	if (!ceph_has_realms_with_quotas(inode))
		return false;

	/* return immediately if we're decreasing file size */
	if (newsize <= size)
		return false;

	return check_quota_exceeded(inode, QUOTA_CHECK_MAX_BYTES_APPROACHING_OP,
				    (newsize - size));
}

/*
 * ceph_quota_update_statfs - if root has quota update statfs with quota status
 * @fsc:	filesystem client instance
 * @buf:	statfs to update
 *
 * If the mounted filesystem root has max_bytes quota set, update the filesystem
 * statistics with the quota status.
 *
 * This function returns true if the stats have been updated, false otherwise.
 */
bool ceph_quota_update_statfs(struct ceph_fs_client *fsc, struct kstatfs *buf)
{
	struct ceph_mds_client *mdsc = fsc->mdsc;
	struct ceph_inode_info *ci;
	struct ceph_snap_realm *realm;
	struct inode *in;
	u64 total = 0, used, free;
	bool is_updated = false;

	down_read(&mdsc->snap_rwsem);
	realm = get_quota_realm(mdsc, d_inode(fsc->sb->s_root), true);
	up_read(&mdsc->snap_rwsem);
	if (!realm)
		return false;

	spin_lock(&realm->inodes_with_caps_lock);
	in = realm->inode ? igrab(realm->inode) : NULL;
	spin_unlock(&realm->inodes_with_caps_lock);
	if (in) {
		ci = ceph_inode(in);
		spin_lock(&ci->i_ceph_lock);
		if (ci->i_max_bytes) {
			total = ci->i_max_bytes >> CEPH_BLOCK_SHIFT;
			used = ci->i_rbytes >> CEPH_BLOCK_SHIFT;
			/* It is possible for a quota to be exceeded.
			 * Report 'zero' in that case
			 */
			free = total > used ? total - used : 0;
		}
		spin_unlock(&ci->i_ceph_lock);
		if (total) {
			buf->f_blocks = total;
			buf->f_bfree = free;
			buf->f_bavail = free;
			is_updated = true;
		}
		iput(in);
	}
	ceph_put_snap_realm(mdsc, realm);

	return is_updated;
}

/*
 * ceph_quota_check_rename - check if a rename can be executed
 * @mdsc:	MDS client instance
 * @old:	inode to be copied
 * @new:	destination inode (directory)
 *
 * This function verifies if a rename (e.g. moving a file or directory) can be
 * executed.  It forces an rstat update in the @new target directory (and in the
 * source @old as well, if it's a directory).  The actual check is done both for
 * max_files and max_bytes.
 *
 * This function returns 0 if it's OK to do the rename, or, if quotas are
 * exceeded, -EXDEV (if @old is a directory) or -EDQUOT.
 */
int ceph_quota_check_rename(struct ceph_mds_client *mdsc,
			    struct inode *old, struct inode *new)
{
	struct ceph_inode_info *ci_old = ceph_inode(old);
	int ret = 0;

	if (ceph_quota_is_same_realm(old, new))
		return 0;

	/*
	 * Get the latest rstat for target directory (and for source, if a
	 * directory)
	 */
	ret = ceph_do_getattr(new, CEPH_STAT_RSTAT, false);
	if (ret)
		return ret;

	if (S_ISDIR(old->i_mode)) {
		ret = ceph_do_getattr(old, CEPH_STAT_RSTAT, false);
		if (ret)
			return ret;
		ret = check_quota_exceeded(new, QUOTA_CHECK_MAX_BYTES_OP,
					   ci_old->i_rbytes);
		if (!ret)
			ret = check_quota_exceeded(new,
						   QUOTA_CHECK_MAX_FILES_OP,
						   ci_old->i_rfiles +
						   ci_old->i_rsubdirs);
		if (ret)
			ret = -EXDEV;
	} else {
		ret = check_quota_exceeded(new, QUOTA_CHECK_MAX_BYTES_OP,
					   i_size_read(old));
		if (!ret)
			ret = check_quota_exceeded(new,
						   QUOTA_CHECK_MAX_FILES_OP, 1);
		if (ret)
			ret = -EDQUOT;
	}

	return ret;
}