xref: /openbmc/linux/fs/btrfs/qgroup.c (revision 97da55fc)
1 /*
2  * Copyright (C) 2011 STRATO.  All rights reserved.
3  *
4  * This program is free software; you can redistribute it and/or
5  * modify it under the terms of the GNU General Public
6  * License v2 as published by the Free Software Foundation.
7  *
8  * This program is distributed in the hope that it will be useful,
9  * but WITHOUT ANY WARRANTY; without even the implied warranty of
10  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
11  * General Public License for more details.
12  *
13  * You should have received a copy of the GNU General Public
14  * License along with this program; if not, write to the
15  * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
16  * Boston, MA 021110-1307, USA.
17  */
18 
19 #include <linux/sched.h>
20 #include <linux/pagemap.h>
21 #include <linux/writeback.h>
22 #include <linux/blkdev.h>
23 #include <linux/rbtree.h>
24 #include <linux/slab.h>
25 #include <linux/workqueue.h>
26 #include <linux/btrfs.h>
27 
28 #include "ctree.h"
29 #include "transaction.h"
30 #include "disk-io.h"
31 #include "locking.h"
32 #include "ulist.h"
33 #include "backref.h"
34 
35 /* TODO XXX FIXME
36  *  - subvol delete -> delete when ref goes to 0? delete limits also?
37  *  - reorganize keys
38  *  - compressed
39  *  - sync
40  *  - rescan
41  *  - copy also limits on subvol creation
42  *  - limit
43  *  - caches fuer ulists
44  *  - performance benchmarks
45  *  - check all ioctl parameters
46  */
47 
48 /*
49  * one struct for each qgroup, organized in fs_info->qgroup_tree.
50  */
51 struct btrfs_qgroup {
52 	u64 qgroupid;
53 
54 	/*
55 	 * state
56 	 */
57 	u64 rfer;	/* referenced */
58 	u64 rfer_cmpr;	/* referenced compressed */
59 	u64 excl;	/* exclusive */
60 	u64 excl_cmpr;	/* exclusive compressed */
61 
62 	/*
63 	 * limits
64 	 */
65 	u64 lim_flags;	/* which limits are set */
66 	u64 max_rfer;
67 	u64 max_excl;
68 	u64 rsv_rfer;
69 	u64 rsv_excl;
70 
71 	/*
72 	 * reservation tracking
73 	 */
74 	u64 reserved;
75 
76 	/*
77 	 * lists
78 	 */
79 	struct list_head groups;  /* groups this group is member of */
80 	struct list_head members; /* groups that are members of this group */
81 	struct list_head dirty;   /* dirty groups */
82 	struct rb_node node;	  /* tree of qgroups */
83 
84 	/*
85 	 * temp variables for accounting operations
86 	 */
87 	u64 tag;
88 	u64 refcnt;
89 };
90 
91 /*
92  * glue structure to represent the relations between qgroups.
93  */
94 struct btrfs_qgroup_list {
95 	struct list_head next_group;
96 	struct list_head next_member;
97 	struct btrfs_qgroup *group;
98 	struct btrfs_qgroup *member;
99 };
100 
101 /* must be called with qgroup_lock held */
102 static struct btrfs_qgroup *find_qgroup_rb(struct btrfs_fs_info *fs_info,
103 					   u64 qgroupid)
104 {
105 	struct rb_node *n = fs_info->qgroup_tree.rb_node;
106 	struct btrfs_qgroup *qgroup;
107 
108 	while (n) {
109 		qgroup = rb_entry(n, struct btrfs_qgroup, node);
110 		if (qgroup->qgroupid < qgroupid)
111 			n = n->rb_left;
112 		else if (qgroup->qgroupid > qgroupid)
113 			n = n->rb_right;
114 		else
115 			return qgroup;
116 	}
117 	return NULL;
118 }
119 
120 /* must be called with qgroup_lock held */
121 static struct btrfs_qgroup *add_qgroup_rb(struct btrfs_fs_info *fs_info,
122 					  u64 qgroupid)
123 {
124 	struct rb_node **p = &fs_info->qgroup_tree.rb_node;
125 	struct rb_node *parent = NULL;
126 	struct btrfs_qgroup *qgroup;
127 
128 	while (*p) {
129 		parent = *p;
130 		qgroup = rb_entry(parent, struct btrfs_qgroup, node);
131 
132 		if (qgroup->qgroupid < qgroupid)
133 			p = &(*p)->rb_left;
134 		else if (qgroup->qgroupid > qgroupid)
135 			p = &(*p)->rb_right;
136 		else
137 			return qgroup;
138 	}
139 
140 	qgroup = kzalloc(sizeof(*qgroup), GFP_ATOMIC);
141 	if (!qgroup)
142 		return ERR_PTR(-ENOMEM);
143 
144 	qgroup->qgroupid = qgroupid;
145 	INIT_LIST_HEAD(&qgroup->groups);
146 	INIT_LIST_HEAD(&qgroup->members);
147 	INIT_LIST_HEAD(&qgroup->dirty);
148 
149 	rb_link_node(&qgroup->node, parent, p);
150 	rb_insert_color(&qgroup->node, &fs_info->qgroup_tree);
151 
152 	return qgroup;
153 }
154 
155 /* must be called with qgroup_lock held */
156 static int del_qgroup_rb(struct btrfs_fs_info *fs_info, u64 qgroupid)
157 {
158 	struct btrfs_qgroup *qgroup = find_qgroup_rb(fs_info, qgroupid);
159 	struct btrfs_qgroup_list *list;
160 
161 	if (!qgroup)
162 		return -ENOENT;
163 
164 	rb_erase(&qgroup->node, &fs_info->qgroup_tree);
165 	list_del(&qgroup->dirty);
166 
167 	while (!list_empty(&qgroup->groups)) {
168 		list = list_first_entry(&qgroup->groups,
169 					struct btrfs_qgroup_list, next_group);
170 		list_del(&list->next_group);
171 		list_del(&list->next_member);
172 		kfree(list);
173 	}
174 
175 	while (!list_empty(&qgroup->members)) {
176 		list = list_first_entry(&qgroup->members,
177 					struct btrfs_qgroup_list, next_member);
178 		list_del(&list->next_group);
179 		list_del(&list->next_member);
180 		kfree(list);
181 	}
182 	kfree(qgroup);
183 
184 	return 0;
185 }
186 
187 /* must be called with qgroup_lock held */
188 static int add_relation_rb(struct btrfs_fs_info *fs_info,
189 			   u64 memberid, u64 parentid)
190 {
191 	struct btrfs_qgroup *member;
192 	struct btrfs_qgroup *parent;
193 	struct btrfs_qgroup_list *list;
194 
195 	member = find_qgroup_rb(fs_info, memberid);
196 	parent = find_qgroup_rb(fs_info, parentid);
197 	if (!member || !parent)
198 		return -ENOENT;
199 
200 	list = kzalloc(sizeof(*list), GFP_ATOMIC);
201 	if (!list)
202 		return -ENOMEM;
203 
204 	list->group = parent;
205 	list->member = member;
206 	list_add_tail(&list->next_group, &member->groups);
207 	list_add_tail(&list->next_member, &parent->members);
208 
209 	return 0;
210 }
211 
212 /* must be called with qgroup_lock held */
213 static int del_relation_rb(struct btrfs_fs_info *fs_info,
214 			   u64 memberid, u64 parentid)
215 {
216 	struct btrfs_qgroup *member;
217 	struct btrfs_qgroup *parent;
218 	struct btrfs_qgroup_list *list;
219 
220 	member = find_qgroup_rb(fs_info, memberid);
221 	parent = find_qgroup_rb(fs_info, parentid);
222 	if (!member || !parent)
223 		return -ENOENT;
224 
225 	list_for_each_entry(list, &member->groups, next_group) {
226 		if (list->group == parent) {
227 			list_del(&list->next_group);
228 			list_del(&list->next_member);
229 			kfree(list);
230 			return 0;
231 		}
232 	}
233 	return -ENOENT;
234 }
235 
236 /*
237  * The full config is read in one go, only called from open_ctree()
238  * It doesn't use any locking, as at this point we're still single-threaded
239  */
240 int btrfs_read_qgroup_config(struct btrfs_fs_info *fs_info)
241 {
242 	struct btrfs_key key;
243 	struct btrfs_key found_key;
244 	struct btrfs_root *quota_root = fs_info->quota_root;
245 	struct btrfs_path *path = NULL;
246 	struct extent_buffer *l;
247 	int slot;
248 	int ret = 0;
249 	u64 flags = 0;
250 
251 	if (!fs_info->quota_enabled)
252 		return 0;
253 
254 	path = btrfs_alloc_path();
255 	if (!path) {
256 		ret = -ENOMEM;
257 		goto out;
258 	}
259 
260 	/* default this to quota off, in case no status key is found */
261 	fs_info->qgroup_flags = 0;
262 
263 	/*
264 	 * pass 1: read status, all qgroup infos and limits
265 	 */
266 	key.objectid = 0;
267 	key.type = 0;
268 	key.offset = 0;
269 	ret = btrfs_search_slot_for_read(quota_root, &key, path, 1, 1);
270 	if (ret)
271 		goto out;
272 
273 	while (1) {
274 		struct btrfs_qgroup *qgroup;
275 
276 		slot = path->slots[0];
277 		l = path->nodes[0];
278 		btrfs_item_key_to_cpu(l, &found_key, slot);
279 
280 		if (found_key.type == BTRFS_QGROUP_STATUS_KEY) {
281 			struct btrfs_qgroup_status_item *ptr;
282 
283 			ptr = btrfs_item_ptr(l, slot,
284 					     struct btrfs_qgroup_status_item);
285 
286 			if (btrfs_qgroup_status_version(l, ptr) !=
287 			    BTRFS_QGROUP_STATUS_VERSION) {
288 				printk(KERN_ERR
289 				 "btrfs: old qgroup version, quota disabled\n");
290 				goto out;
291 			}
292 			if (btrfs_qgroup_status_generation(l, ptr) !=
293 			    fs_info->generation) {
294 				flags |= BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT;
295 				printk(KERN_ERR
296 					"btrfs: qgroup generation mismatch, "
297 					"marked as inconsistent\n");
298 			}
299 			fs_info->qgroup_flags = btrfs_qgroup_status_flags(l,
300 									  ptr);
301 			/* FIXME read scan element */
302 			goto next1;
303 		}
304 
305 		if (found_key.type != BTRFS_QGROUP_INFO_KEY &&
306 		    found_key.type != BTRFS_QGROUP_LIMIT_KEY)
307 			goto next1;
308 
309 		qgroup = find_qgroup_rb(fs_info, found_key.offset);
310 		if ((qgroup && found_key.type == BTRFS_QGROUP_INFO_KEY) ||
311 		    (!qgroup && found_key.type == BTRFS_QGROUP_LIMIT_KEY)) {
312 			printk(KERN_ERR "btrfs: inconsitent qgroup config\n");
313 			flags |= BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT;
314 		}
315 		if (!qgroup) {
316 			qgroup = add_qgroup_rb(fs_info, found_key.offset);
317 			if (IS_ERR(qgroup)) {
318 				ret = PTR_ERR(qgroup);
319 				goto out;
320 			}
321 		}
322 		switch (found_key.type) {
323 		case BTRFS_QGROUP_INFO_KEY: {
324 			struct btrfs_qgroup_info_item *ptr;
325 
326 			ptr = btrfs_item_ptr(l, slot,
327 					     struct btrfs_qgroup_info_item);
328 			qgroup->rfer = btrfs_qgroup_info_rfer(l, ptr);
329 			qgroup->rfer_cmpr = btrfs_qgroup_info_rfer_cmpr(l, ptr);
330 			qgroup->excl = btrfs_qgroup_info_excl(l, ptr);
331 			qgroup->excl_cmpr = btrfs_qgroup_info_excl_cmpr(l, ptr);
332 			/* generation currently unused */
333 			break;
334 		}
335 		case BTRFS_QGROUP_LIMIT_KEY: {
336 			struct btrfs_qgroup_limit_item *ptr;
337 
338 			ptr = btrfs_item_ptr(l, slot,
339 					     struct btrfs_qgroup_limit_item);
340 			qgroup->lim_flags = btrfs_qgroup_limit_flags(l, ptr);
341 			qgroup->max_rfer = btrfs_qgroup_limit_max_rfer(l, ptr);
342 			qgroup->max_excl = btrfs_qgroup_limit_max_excl(l, ptr);
343 			qgroup->rsv_rfer = btrfs_qgroup_limit_rsv_rfer(l, ptr);
344 			qgroup->rsv_excl = btrfs_qgroup_limit_rsv_excl(l, ptr);
345 			break;
346 		}
347 		}
348 next1:
349 		ret = btrfs_next_item(quota_root, path);
350 		if (ret < 0)
351 			goto out;
352 		if (ret)
353 			break;
354 	}
355 	btrfs_release_path(path);
356 
357 	/*
358 	 * pass 2: read all qgroup relations
359 	 */
360 	key.objectid = 0;
361 	key.type = BTRFS_QGROUP_RELATION_KEY;
362 	key.offset = 0;
363 	ret = btrfs_search_slot_for_read(quota_root, &key, path, 1, 0);
364 	if (ret)
365 		goto out;
366 	while (1) {
367 		slot = path->slots[0];
368 		l = path->nodes[0];
369 		btrfs_item_key_to_cpu(l, &found_key, slot);
370 
371 		if (found_key.type != BTRFS_QGROUP_RELATION_KEY)
372 			goto next2;
373 
374 		if (found_key.objectid > found_key.offset) {
375 			/* parent <- member, not needed to build config */
376 			/* FIXME should we omit the key completely? */
377 			goto next2;
378 		}
379 
380 		ret = add_relation_rb(fs_info, found_key.objectid,
381 				      found_key.offset);
382 		if (ret == -ENOENT) {
383 			printk(KERN_WARNING
384 				"btrfs: orphan qgroup relation 0x%llx->0x%llx\n",
385 				(unsigned long long)found_key.objectid,
386 				(unsigned long long)found_key.offset);
387 			ret = 0;	/* ignore the error */
388 		}
389 		if (ret)
390 			goto out;
391 next2:
392 		ret = btrfs_next_item(quota_root, path);
393 		if (ret < 0)
394 			goto out;
395 		if (ret)
396 			break;
397 	}
398 out:
399 	fs_info->qgroup_flags |= flags;
400 	if (!(fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_ON)) {
401 		fs_info->quota_enabled = 0;
402 		fs_info->pending_quota_state = 0;
403 	}
404 	btrfs_free_path(path);
405 
406 	return ret < 0 ? ret : 0;
407 }
408 
409 /*
410  * This is only called from close_ctree() or open_ctree(), both in single-
411  * treaded paths. Clean up the in-memory structures. No locking needed.
412  */
413 void btrfs_free_qgroup_config(struct btrfs_fs_info *fs_info)
414 {
415 	struct rb_node *n;
416 	struct btrfs_qgroup *qgroup;
417 	struct btrfs_qgroup_list *list;
418 
419 	while ((n = rb_first(&fs_info->qgroup_tree))) {
420 		qgroup = rb_entry(n, struct btrfs_qgroup, node);
421 		rb_erase(n, &fs_info->qgroup_tree);
422 
423 		WARN_ON(!list_empty(&qgroup->dirty));
424 
425 		while (!list_empty(&qgroup->groups)) {
426 			list = list_first_entry(&qgroup->groups,
427 						struct btrfs_qgroup_list,
428 						next_group);
429 			list_del(&list->next_group);
430 			list_del(&list->next_member);
431 			kfree(list);
432 		}
433 
434 		while (!list_empty(&qgroup->members)) {
435 			list = list_first_entry(&qgroup->members,
436 						struct btrfs_qgroup_list,
437 						next_member);
438 			list_del(&list->next_group);
439 			list_del(&list->next_member);
440 			kfree(list);
441 		}
442 		kfree(qgroup);
443 	}
444 }
445 
446 static int add_qgroup_relation_item(struct btrfs_trans_handle *trans,
447 				    struct btrfs_root *quota_root,
448 				    u64 src, u64 dst)
449 {
450 	int ret;
451 	struct btrfs_path *path;
452 	struct btrfs_key key;
453 
454 	path = btrfs_alloc_path();
455 	if (!path)
456 		return -ENOMEM;
457 
458 	key.objectid = src;
459 	key.type = BTRFS_QGROUP_RELATION_KEY;
460 	key.offset = dst;
461 
462 	ret = btrfs_insert_empty_item(trans, quota_root, path, &key, 0);
463 
464 	btrfs_mark_buffer_dirty(path->nodes[0]);
465 
466 	btrfs_free_path(path);
467 	return ret;
468 }
469 
470 static int del_qgroup_relation_item(struct btrfs_trans_handle *trans,
471 				    struct btrfs_root *quota_root,
472 				    u64 src, u64 dst)
473 {
474 	int ret;
475 	struct btrfs_path *path;
476 	struct btrfs_key key;
477 
478 	path = btrfs_alloc_path();
479 	if (!path)
480 		return -ENOMEM;
481 
482 	key.objectid = src;
483 	key.type = BTRFS_QGROUP_RELATION_KEY;
484 	key.offset = dst;
485 
486 	ret = btrfs_search_slot(trans, quota_root, &key, path, -1, 1);
487 	if (ret < 0)
488 		goto out;
489 
490 	if (ret > 0) {
491 		ret = -ENOENT;
492 		goto out;
493 	}
494 
495 	ret = btrfs_del_item(trans, quota_root, path);
496 out:
497 	btrfs_free_path(path);
498 	return ret;
499 }
500 
501 static int add_qgroup_item(struct btrfs_trans_handle *trans,
502 			   struct btrfs_root *quota_root, u64 qgroupid)
503 {
504 	int ret;
505 	struct btrfs_path *path;
506 	struct btrfs_qgroup_info_item *qgroup_info;
507 	struct btrfs_qgroup_limit_item *qgroup_limit;
508 	struct extent_buffer *leaf;
509 	struct btrfs_key key;
510 
511 	path = btrfs_alloc_path();
512 	if (!path)
513 		return -ENOMEM;
514 
515 	key.objectid = 0;
516 	key.type = BTRFS_QGROUP_INFO_KEY;
517 	key.offset = qgroupid;
518 
519 	ret = btrfs_insert_empty_item(trans, quota_root, path, &key,
520 				      sizeof(*qgroup_info));
521 	if (ret)
522 		goto out;
523 
524 	leaf = path->nodes[0];
525 	qgroup_info = btrfs_item_ptr(leaf, path->slots[0],
526 				 struct btrfs_qgroup_info_item);
527 	btrfs_set_qgroup_info_generation(leaf, qgroup_info, trans->transid);
528 	btrfs_set_qgroup_info_rfer(leaf, qgroup_info, 0);
529 	btrfs_set_qgroup_info_rfer_cmpr(leaf, qgroup_info, 0);
530 	btrfs_set_qgroup_info_excl(leaf, qgroup_info, 0);
531 	btrfs_set_qgroup_info_excl_cmpr(leaf, qgroup_info, 0);
532 
533 	btrfs_mark_buffer_dirty(leaf);
534 
535 	btrfs_release_path(path);
536 
537 	key.type = BTRFS_QGROUP_LIMIT_KEY;
538 	ret = btrfs_insert_empty_item(trans, quota_root, path, &key,
539 				      sizeof(*qgroup_limit));
540 	if (ret)
541 		goto out;
542 
543 	leaf = path->nodes[0];
544 	qgroup_limit = btrfs_item_ptr(leaf, path->slots[0],
545 				  struct btrfs_qgroup_limit_item);
546 	btrfs_set_qgroup_limit_flags(leaf, qgroup_limit, 0);
547 	btrfs_set_qgroup_limit_max_rfer(leaf, qgroup_limit, 0);
548 	btrfs_set_qgroup_limit_max_excl(leaf, qgroup_limit, 0);
549 	btrfs_set_qgroup_limit_rsv_rfer(leaf, qgroup_limit, 0);
550 	btrfs_set_qgroup_limit_rsv_excl(leaf, qgroup_limit, 0);
551 
552 	btrfs_mark_buffer_dirty(leaf);
553 
554 	ret = 0;
555 out:
556 	btrfs_free_path(path);
557 	return ret;
558 }
559 
560 static int del_qgroup_item(struct btrfs_trans_handle *trans,
561 			   struct btrfs_root *quota_root, u64 qgroupid)
562 {
563 	int ret;
564 	struct btrfs_path *path;
565 	struct btrfs_key key;
566 
567 	path = btrfs_alloc_path();
568 	if (!path)
569 		return -ENOMEM;
570 
571 	key.objectid = 0;
572 	key.type = BTRFS_QGROUP_INFO_KEY;
573 	key.offset = qgroupid;
574 	ret = btrfs_search_slot(trans, quota_root, &key, path, -1, 1);
575 	if (ret < 0)
576 		goto out;
577 
578 	if (ret > 0) {
579 		ret = -ENOENT;
580 		goto out;
581 	}
582 
583 	ret = btrfs_del_item(trans, quota_root, path);
584 	if (ret)
585 		goto out;
586 
587 	btrfs_release_path(path);
588 
589 	key.type = BTRFS_QGROUP_LIMIT_KEY;
590 	ret = btrfs_search_slot(trans, quota_root, &key, path, -1, 1);
591 	if (ret < 0)
592 		goto out;
593 
594 	if (ret > 0) {
595 		ret = -ENOENT;
596 		goto out;
597 	}
598 
599 	ret = btrfs_del_item(trans, quota_root, path);
600 
601 out:
602 	btrfs_free_path(path);
603 	return ret;
604 }
605 
606 static int update_qgroup_limit_item(struct btrfs_trans_handle *trans,
607 				    struct btrfs_root *root, u64 qgroupid,
608 				    u64 flags, u64 max_rfer, u64 max_excl,
609 				    u64 rsv_rfer, u64 rsv_excl)
610 {
611 	struct btrfs_path *path;
612 	struct btrfs_key key;
613 	struct extent_buffer *l;
614 	struct btrfs_qgroup_limit_item *qgroup_limit;
615 	int ret;
616 	int slot;
617 
618 	key.objectid = 0;
619 	key.type = BTRFS_QGROUP_LIMIT_KEY;
620 	key.offset = qgroupid;
621 
622 	path = btrfs_alloc_path();
623 	if (!path)
624 		return -ENOMEM;
625 
626 	ret = btrfs_search_slot(trans, root, &key, path, 0, 1);
627 	if (ret > 0)
628 		ret = -ENOENT;
629 
630 	if (ret)
631 		goto out;
632 
633 	l = path->nodes[0];
634 	slot = path->slots[0];
635 	qgroup_limit = btrfs_item_ptr(l, path->slots[0],
636 				      struct btrfs_qgroup_limit_item);
637 	btrfs_set_qgroup_limit_flags(l, qgroup_limit, flags);
638 	btrfs_set_qgroup_limit_max_rfer(l, qgroup_limit, max_rfer);
639 	btrfs_set_qgroup_limit_max_excl(l, qgroup_limit, max_excl);
640 	btrfs_set_qgroup_limit_rsv_rfer(l, qgroup_limit, rsv_rfer);
641 	btrfs_set_qgroup_limit_rsv_excl(l, qgroup_limit, rsv_excl);
642 
643 	btrfs_mark_buffer_dirty(l);
644 
645 out:
646 	btrfs_free_path(path);
647 	return ret;
648 }
649 
650 static int update_qgroup_info_item(struct btrfs_trans_handle *trans,
651 				   struct btrfs_root *root,
652 				   struct btrfs_qgroup *qgroup)
653 {
654 	struct btrfs_path *path;
655 	struct btrfs_key key;
656 	struct extent_buffer *l;
657 	struct btrfs_qgroup_info_item *qgroup_info;
658 	int ret;
659 	int slot;
660 
661 	key.objectid = 0;
662 	key.type = BTRFS_QGROUP_INFO_KEY;
663 	key.offset = qgroup->qgroupid;
664 
665 	path = btrfs_alloc_path();
666 	if (!path)
667 		return -ENOMEM;
668 
669 	ret = btrfs_search_slot(trans, root, &key, path, 0, 1);
670 	if (ret > 0)
671 		ret = -ENOENT;
672 
673 	if (ret)
674 		goto out;
675 
676 	l = path->nodes[0];
677 	slot = path->slots[0];
678 	qgroup_info = btrfs_item_ptr(l, path->slots[0],
679 				 struct btrfs_qgroup_info_item);
680 	btrfs_set_qgroup_info_generation(l, qgroup_info, trans->transid);
681 	btrfs_set_qgroup_info_rfer(l, qgroup_info, qgroup->rfer);
682 	btrfs_set_qgroup_info_rfer_cmpr(l, qgroup_info, qgroup->rfer_cmpr);
683 	btrfs_set_qgroup_info_excl(l, qgroup_info, qgroup->excl);
684 	btrfs_set_qgroup_info_excl_cmpr(l, qgroup_info, qgroup->excl_cmpr);
685 
686 	btrfs_mark_buffer_dirty(l);
687 
688 out:
689 	btrfs_free_path(path);
690 	return ret;
691 }
692 
693 static int update_qgroup_status_item(struct btrfs_trans_handle *trans,
694 				     struct btrfs_fs_info *fs_info,
695 				    struct btrfs_root *root)
696 {
697 	struct btrfs_path *path;
698 	struct btrfs_key key;
699 	struct extent_buffer *l;
700 	struct btrfs_qgroup_status_item *ptr;
701 	int ret;
702 	int slot;
703 
704 	key.objectid = 0;
705 	key.type = BTRFS_QGROUP_STATUS_KEY;
706 	key.offset = 0;
707 
708 	path = btrfs_alloc_path();
709 	if (!path)
710 		return -ENOMEM;
711 
712 	ret = btrfs_search_slot(trans, root, &key, path, 0, 1);
713 	if (ret > 0)
714 		ret = -ENOENT;
715 
716 	if (ret)
717 		goto out;
718 
719 	l = path->nodes[0];
720 	slot = path->slots[0];
721 	ptr = btrfs_item_ptr(l, slot, struct btrfs_qgroup_status_item);
722 	btrfs_set_qgroup_status_flags(l, ptr, fs_info->qgroup_flags);
723 	btrfs_set_qgroup_status_generation(l, ptr, trans->transid);
724 	/* XXX scan */
725 
726 	btrfs_mark_buffer_dirty(l);
727 
728 out:
729 	btrfs_free_path(path);
730 	return ret;
731 }
732 
733 /*
734  * called with qgroup_lock held
735  */
736 static int btrfs_clean_quota_tree(struct btrfs_trans_handle *trans,
737 				  struct btrfs_root *root)
738 {
739 	struct btrfs_path *path;
740 	struct btrfs_key key;
741 	struct extent_buffer *leaf = NULL;
742 	int ret;
743 	int nr = 0;
744 
745 	path = btrfs_alloc_path();
746 	if (!path)
747 		return -ENOMEM;
748 
749 	path->leave_spinning = 1;
750 
751 	key.objectid = 0;
752 	key.offset = 0;
753 	key.type = 0;
754 
755 	while (1) {
756 		ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
757 		if (ret < 0)
758 			goto out;
759 		leaf = path->nodes[0];
760 		nr = btrfs_header_nritems(leaf);
761 		if (!nr)
762 			break;
763 		/*
764 		 * delete the leaf one by one
765 		 * since the whole tree is going
766 		 * to be deleted.
767 		 */
768 		path->slots[0] = 0;
769 		ret = btrfs_del_items(trans, root, path, 0, nr);
770 		if (ret)
771 			goto out;
772 
773 		btrfs_release_path(path);
774 	}
775 	ret = 0;
776 out:
777 	root->fs_info->pending_quota_state = 0;
778 	btrfs_free_path(path);
779 	return ret;
780 }
781 
782 int btrfs_quota_enable(struct btrfs_trans_handle *trans,
783 		       struct btrfs_fs_info *fs_info)
784 {
785 	struct btrfs_root *quota_root;
786 	struct btrfs_path *path = NULL;
787 	struct btrfs_qgroup_status_item *ptr;
788 	struct extent_buffer *leaf;
789 	struct btrfs_key key;
790 	int ret = 0;
791 
792 	spin_lock(&fs_info->qgroup_lock);
793 	if (fs_info->quota_root) {
794 		fs_info->pending_quota_state = 1;
795 		spin_unlock(&fs_info->qgroup_lock);
796 		goto out;
797 	}
798 	spin_unlock(&fs_info->qgroup_lock);
799 
800 	/*
801 	 * initially create the quota tree
802 	 */
803 	quota_root = btrfs_create_tree(trans, fs_info,
804 				       BTRFS_QUOTA_TREE_OBJECTID);
805 	if (IS_ERR(quota_root)) {
806 		ret =  PTR_ERR(quota_root);
807 		goto out;
808 	}
809 
810 	path = btrfs_alloc_path();
811 	if (!path) {
812 		ret = -ENOMEM;
813 		goto out_free_root;
814 	}
815 
816 	key.objectid = 0;
817 	key.type = BTRFS_QGROUP_STATUS_KEY;
818 	key.offset = 0;
819 
820 	ret = btrfs_insert_empty_item(trans, quota_root, path, &key,
821 				      sizeof(*ptr));
822 	if (ret)
823 		goto out_free_path;
824 
825 	leaf = path->nodes[0];
826 	ptr = btrfs_item_ptr(leaf, path->slots[0],
827 				 struct btrfs_qgroup_status_item);
828 	btrfs_set_qgroup_status_generation(leaf, ptr, trans->transid);
829 	btrfs_set_qgroup_status_version(leaf, ptr, BTRFS_QGROUP_STATUS_VERSION);
830 	fs_info->qgroup_flags = BTRFS_QGROUP_STATUS_FLAG_ON |
831 				BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT;
832 	btrfs_set_qgroup_status_flags(leaf, ptr, fs_info->qgroup_flags);
833 	btrfs_set_qgroup_status_scan(leaf, ptr, 0);
834 
835 	btrfs_mark_buffer_dirty(leaf);
836 
837 	spin_lock(&fs_info->qgroup_lock);
838 	fs_info->quota_root = quota_root;
839 	fs_info->pending_quota_state = 1;
840 	spin_unlock(&fs_info->qgroup_lock);
841 out_free_path:
842 	btrfs_free_path(path);
843 out_free_root:
844 	if (ret) {
845 		free_extent_buffer(quota_root->node);
846 		free_extent_buffer(quota_root->commit_root);
847 		kfree(quota_root);
848 	}
849 out:
850 	return ret;
851 }
852 
853 int btrfs_quota_disable(struct btrfs_trans_handle *trans,
854 			struct btrfs_fs_info *fs_info)
855 {
856 	struct btrfs_root *tree_root = fs_info->tree_root;
857 	struct btrfs_root *quota_root;
858 	int ret = 0;
859 
860 	spin_lock(&fs_info->qgroup_lock);
861 	if (!fs_info->quota_root) {
862 		spin_unlock(&fs_info->qgroup_lock);
863 		return 0;
864 	}
865 	fs_info->quota_enabled = 0;
866 	fs_info->pending_quota_state = 0;
867 	quota_root = fs_info->quota_root;
868 	fs_info->quota_root = NULL;
869 	btrfs_free_qgroup_config(fs_info);
870 	spin_unlock(&fs_info->qgroup_lock);
871 
872 	if (!quota_root)
873 		return -EINVAL;
874 
875 	ret = btrfs_clean_quota_tree(trans, quota_root);
876 	if (ret)
877 		goto out;
878 
879 	ret = btrfs_del_root(trans, tree_root, &quota_root->root_key);
880 	if (ret)
881 		goto out;
882 
883 	list_del(&quota_root->dirty_list);
884 
885 	btrfs_tree_lock(quota_root->node);
886 	clean_tree_block(trans, tree_root, quota_root->node);
887 	btrfs_tree_unlock(quota_root->node);
888 	btrfs_free_tree_block(trans, quota_root, quota_root->node, 0, 1);
889 
890 	free_extent_buffer(quota_root->node);
891 	free_extent_buffer(quota_root->commit_root);
892 	kfree(quota_root);
893 out:
894 	return ret;
895 }
896 
897 int btrfs_quota_rescan(struct btrfs_fs_info *fs_info)
898 {
899 	/* FIXME */
900 	return 0;
901 }
902 
903 int btrfs_add_qgroup_relation(struct btrfs_trans_handle *trans,
904 			      struct btrfs_fs_info *fs_info, u64 src, u64 dst)
905 {
906 	struct btrfs_root *quota_root;
907 	int ret = 0;
908 
909 	quota_root = fs_info->quota_root;
910 	if (!quota_root)
911 		return -EINVAL;
912 
913 	ret = add_qgroup_relation_item(trans, quota_root, src, dst);
914 	if (ret)
915 		return ret;
916 
917 	ret = add_qgroup_relation_item(trans, quota_root, dst, src);
918 	if (ret) {
919 		del_qgroup_relation_item(trans, quota_root, src, dst);
920 		return ret;
921 	}
922 
923 	spin_lock(&fs_info->qgroup_lock);
924 	ret = add_relation_rb(quota_root->fs_info, src, dst);
925 	spin_unlock(&fs_info->qgroup_lock);
926 
927 	return ret;
928 }
929 
930 int btrfs_del_qgroup_relation(struct btrfs_trans_handle *trans,
931 			      struct btrfs_fs_info *fs_info, u64 src, u64 dst)
932 {
933 	struct btrfs_root *quota_root;
934 	int ret = 0;
935 	int err;
936 
937 	quota_root = fs_info->quota_root;
938 	if (!quota_root)
939 		return -EINVAL;
940 
941 	ret = del_qgroup_relation_item(trans, quota_root, src, dst);
942 	err = del_qgroup_relation_item(trans, quota_root, dst, src);
943 	if (err && !ret)
944 		ret = err;
945 
946 	spin_lock(&fs_info->qgroup_lock);
947 	del_relation_rb(fs_info, src, dst);
948 
949 	spin_unlock(&fs_info->qgroup_lock);
950 
951 	return ret;
952 }
953 
954 int btrfs_create_qgroup(struct btrfs_trans_handle *trans,
955 			struct btrfs_fs_info *fs_info, u64 qgroupid, char *name)
956 {
957 	struct btrfs_root *quota_root;
958 	struct btrfs_qgroup *qgroup;
959 	int ret = 0;
960 
961 	quota_root = fs_info->quota_root;
962 	if (!quota_root)
963 		return -EINVAL;
964 
965 	ret = add_qgroup_item(trans, quota_root, qgroupid);
966 
967 	spin_lock(&fs_info->qgroup_lock);
968 	qgroup = add_qgroup_rb(fs_info, qgroupid);
969 	spin_unlock(&fs_info->qgroup_lock);
970 
971 	if (IS_ERR(qgroup))
972 		ret = PTR_ERR(qgroup);
973 
974 	return ret;
975 }
976 
977 int btrfs_remove_qgroup(struct btrfs_trans_handle *trans,
978 			struct btrfs_fs_info *fs_info, u64 qgroupid)
979 {
980 	struct btrfs_root *quota_root;
981 	struct btrfs_qgroup *qgroup;
982 	int ret = 0;
983 
984 	quota_root = fs_info->quota_root;
985 	if (!quota_root)
986 		return -EINVAL;
987 
988 	/* check if there are no relations to this qgroup */
989 	spin_lock(&fs_info->qgroup_lock);
990 	qgroup = find_qgroup_rb(fs_info, qgroupid);
991 	if (qgroup) {
992 		if (!list_empty(&qgroup->groups) || !list_empty(&qgroup->members)) {
993 			spin_unlock(&fs_info->qgroup_lock);
994 			return -EBUSY;
995 		}
996 	}
997 	spin_unlock(&fs_info->qgroup_lock);
998 
999 	ret = del_qgroup_item(trans, quota_root, qgroupid);
1000 
1001 	spin_lock(&fs_info->qgroup_lock);
1002 	del_qgroup_rb(quota_root->fs_info, qgroupid);
1003 	spin_unlock(&fs_info->qgroup_lock);
1004 
1005 	return ret;
1006 }
1007 
1008 int btrfs_limit_qgroup(struct btrfs_trans_handle *trans,
1009 		       struct btrfs_fs_info *fs_info, u64 qgroupid,
1010 		       struct btrfs_qgroup_limit *limit)
1011 {
1012 	struct btrfs_root *quota_root = fs_info->quota_root;
1013 	struct btrfs_qgroup *qgroup;
1014 	int ret = 0;
1015 
1016 	if (!quota_root)
1017 		return -EINVAL;
1018 
1019 	ret = update_qgroup_limit_item(trans, quota_root, qgroupid,
1020 				       limit->flags, limit->max_rfer,
1021 				       limit->max_excl, limit->rsv_rfer,
1022 				       limit->rsv_excl);
1023 	if (ret) {
1024 		fs_info->qgroup_flags |= BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT;
1025 		printk(KERN_INFO "unable to update quota limit for %llu\n",
1026 		       (unsigned long long)qgroupid);
1027 	}
1028 
1029 	spin_lock(&fs_info->qgroup_lock);
1030 
1031 	qgroup = find_qgroup_rb(fs_info, qgroupid);
1032 	if (!qgroup) {
1033 		ret = -ENOENT;
1034 		goto unlock;
1035 	}
1036 	qgroup->lim_flags = limit->flags;
1037 	qgroup->max_rfer = limit->max_rfer;
1038 	qgroup->max_excl = limit->max_excl;
1039 	qgroup->rsv_rfer = limit->rsv_rfer;
1040 	qgroup->rsv_excl = limit->rsv_excl;
1041 
1042 unlock:
1043 	spin_unlock(&fs_info->qgroup_lock);
1044 
1045 	return ret;
1046 }
1047 
1048 static void qgroup_dirty(struct btrfs_fs_info *fs_info,
1049 			 struct btrfs_qgroup *qgroup)
1050 {
1051 	if (list_empty(&qgroup->dirty))
1052 		list_add(&qgroup->dirty, &fs_info->dirty_qgroups);
1053 }
1054 
1055 /*
1056  * btrfs_qgroup_record_ref is called when the ref is added or deleted. it puts
1057  * the modification into a list that's later used by btrfs_end_transaction to
1058  * pass the recorded modifications on to btrfs_qgroup_account_ref.
1059  */
1060 int btrfs_qgroup_record_ref(struct btrfs_trans_handle *trans,
1061 			    struct btrfs_delayed_ref_node *node,
1062 			    struct btrfs_delayed_extent_op *extent_op)
1063 {
1064 	struct qgroup_update *u;
1065 
1066 	BUG_ON(!trans->delayed_ref_elem.seq);
1067 	u = kmalloc(sizeof(*u), GFP_NOFS);
1068 	if (!u)
1069 		return -ENOMEM;
1070 
1071 	u->node = node;
1072 	u->extent_op = extent_op;
1073 	list_add_tail(&u->list, &trans->qgroup_ref_list);
1074 
1075 	return 0;
1076 }
1077 
1078 /*
1079  * btrfs_qgroup_account_ref is called for every ref that is added to or deleted
1080  * from the fs. First, all roots referencing the extent are searched, and
1081  * then the space is accounted accordingly to the different roots. The
1082  * accounting algorithm works in 3 steps documented inline.
1083  */
1084 int btrfs_qgroup_account_ref(struct btrfs_trans_handle *trans,
1085 			     struct btrfs_fs_info *fs_info,
1086 			     struct btrfs_delayed_ref_node *node,
1087 			     struct btrfs_delayed_extent_op *extent_op)
1088 {
1089 	struct btrfs_key ins;
1090 	struct btrfs_root *quota_root;
1091 	u64 ref_root;
1092 	struct btrfs_qgroup *qgroup;
1093 	struct ulist_node *unode;
1094 	struct ulist *roots = NULL;
1095 	struct ulist *tmp = NULL;
1096 	struct ulist_iterator uiter;
1097 	u64 seq;
1098 	int ret = 0;
1099 	int sgn;
1100 
1101 	if (!fs_info->quota_enabled)
1102 		return 0;
1103 
1104 	BUG_ON(!fs_info->quota_root);
1105 
1106 	ins.objectid = node->bytenr;
1107 	ins.offset = node->num_bytes;
1108 	ins.type = BTRFS_EXTENT_ITEM_KEY;
1109 
1110 	if (node->type == BTRFS_TREE_BLOCK_REF_KEY ||
1111 	    node->type == BTRFS_SHARED_BLOCK_REF_KEY) {
1112 		struct btrfs_delayed_tree_ref *ref;
1113 		ref = btrfs_delayed_node_to_tree_ref(node);
1114 		ref_root = ref->root;
1115 	} else if (node->type == BTRFS_EXTENT_DATA_REF_KEY ||
1116 		   node->type == BTRFS_SHARED_DATA_REF_KEY) {
1117 		struct btrfs_delayed_data_ref *ref;
1118 		ref = btrfs_delayed_node_to_data_ref(node);
1119 		ref_root = ref->root;
1120 	} else {
1121 		BUG();
1122 	}
1123 
1124 	if (!is_fstree(ref_root)) {
1125 		/*
1126 		 * non-fs-trees are not being accounted
1127 		 */
1128 		return 0;
1129 	}
1130 
1131 	switch (node->action) {
1132 	case BTRFS_ADD_DELAYED_REF:
1133 	case BTRFS_ADD_DELAYED_EXTENT:
1134 		sgn = 1;
1135 		break;
1136 	case BTRFS_DROP_DELAYED_REF:
1137 		sgn = -1;
1138 		break;
1139 	case BTRFS_UPDATE_DELAYED_HEAD:
1140 		return 0;
1141 	default:
1142 		BUG();
1143 	}
1144 
1145 	/*
1146 	 * the delayed ref sequence number we pass depends on the direction of
1147 	 * the operation. for add operations, we pass (node->seq - 1) to skip
1148 	 * the delayed ref's current sequence number, because we need the state
1149 	 * of the tree before the add operation. for delete operations, we pass
1150 	 * (node->seq) to include the delayed ref's current sequence number,
1151 	 * because we need the state of the tree after the delete operation.
1152 	 */
1153 	ret = btrfs_find_all_roots(trans, fs_info, node->bytenr,
1154 				   sgn > 0 ? node->seq - 1 : node->seq, &roots);
1155 	if (ret < 0)
1156 		goto out;
1157 
1158 	spin_lock(&fs_info->qgroup_lock);
1159 	quota_root = fs_info->quota_root;
1160 	if (!quota_root)
1161 		goto unlock;
1162 
1163 	qgroup = find_qgroup_rb(fs_info, ref_root);
1164 	if (!qgroup)
1165 		goto unlock;
1166 
1167 	/*
1168 	 * step 1: for each old ref, visit all nodes once and inc refcnt
1169 	 */
1170 	tmp = ulist_alloc(GFP_ATOMIC);
1171 	if (!tmp) {
1172 		ret = -ENOMEM;
1173 		goto unlock;
1174 	}
1175 	seq = fs_info->qgroup_seq;
1176 	fs_info->qgroup_seq += roots->nnodes + 1; /* max refcnt */
1177 
1178 	ULIST_ITER_INIT(&uiter);
1179 	while ((unode = ulist_next(roots, &uiter))) {
1180 		struct ulist_node *tmp_unode;
1181 		struct ulist_iterator tmp_uiter;
1182 		struct btrfs_qgroup *qg;
1183 
1184 		qg = find_qgroup_rb(fs_info, unode->val);
1185 		if (!qg)
1186 			continue;
1187 
1188 		ulist_reinit(tmp);
1189 						/* XXX id not needed */
1190 		ulist_add(tmp, qg->qgroupid, (u64)(uintptr_t)qg, GFP_ATOMIC);
1191 		ULIST_ITER_INIT(&tmp_uiter);
1192 		while ((tmp_unode = ulist_next(tmp, &tmp_uiter))) {
1193 			struct btrfs_qgroup_list *glist;
1194 
1195 			qg = (struct btrfs_qgroup *)(uintptr_t)tmp_unode->aux;
1196 			if (qg->refcnt < seq)
1197 				qg->refcnt = seq + 1;
1198 			else
1199 				++qg->refcnt;
1200 
1201 			list_for_each_entry(glist, &qg->groups, next_group) {
1202 				ulist_add(tmp, glist->group->qgroupid,
1203 					  (u64)(uintptr_t)glist->group,
1204 					  GFP_ATOMIC);
1205 			}
1206 		}
1207 	}
1208 
1209 	/*
1210 	 * step 2: walk from the new root
1211 	 */
1212 	ulist_reinit(tmp);
1213 	ulist_add(tmp, qgroup->qgroupid, (uintptr_t)qgroup, GFP_ATOMIC);
1214 	ULIST_ITER_INIT(&uiter);
1215 	while ((unode = ulist_next(tmp, &uiter))) {
1216 		struct btrfs_qgroup *qg;
1217 		struct btrfs_qgroup_list *glist;
1218 
1219 		qg = (struct btrfs_qgroup *)(uintptr_t)unode->aux;
1220 		if (qg->refcnt < seq) {
1221 			/* not visited by step 1 */
1222 			qg->rfer += sgn * node->num_bytes;
1223 			qg->rfer_cmpr += sgn * node->num_bytes;
1224 			if (roots->nnodes == 0) {
1225 				qg->excl += sgn * node->num_bytes;
1226 				qg->excl_cmpr += sgn * node->num_bytes;
1227 			}
1228 			qgroup_dirty(fs_info, qg);
1229 		}
1230 		WARN_ON(qg->tag >= seq);
1231 		qg->tag = seq;
1232 
1233 		list_for_each_entry(glist, &qg->groups, next_group) {
1234 			ulist_add(tmp, glist->group->qgroupid,
1235 				  (uintptr_t)glist->group, GFP_ATOMIC);
1236 		}
1237 	}
1238 
1239 	/*
1240 	 * step 3: walk again from old refs
1241 	 */
1242 	ULIST_ITER_INIT(&uiter);
1243 	while ((unode = ulist_next(roots, &uiter))) {
1244 		struct btrfs_qgroup *qg;
1245 		struct ulist_node *tmp_unode;
1246 		struct ulist_iterator tmp_uiter;
1247 
1248 		qg = find_qgroup_rb(fs_info, unode->val);
1249 		if (!qg)
1250 			continue;
1251 
1252 		ulist_reinit(tmp);
1253 		ulist_add(tmp, qg->qgroupid, (uintptr_t)qg, GFP_ATOMIC);
1254 		ULIST_ITER_INIT(&tmp_uiter);
1255 		while ((tmp_unode = ulist_next(tmp, &tmp_uiter))) {
1256 			struct btrfs_qgroup_list *glist;
1257 
1258 			qg = (struct btrfs_qgroup *)(uintptr_t)tmp_unode->aux;
1259 			if (qg->tag == seq)
1260 				continue;
1261 
1262 			if (qg->refcnt - seq == roots->nnodes) {
1263 				qg->excl -= sgn * node->num_bytes;
1264 				qg->excl_cmpr -= sgn * node->num_bytes;
1265 				qgroup_dirty(fs_info, qg);
1266 			}
1267 
1268 			list_for_each_entry(glist, &qg->groups, next_group) {
1269 				ulist_add(tmp, glist->group->qgroupid,
1270 					  (uintptr_t)glist->group,
1271 					  GFP_ATOMIC);
1272 			}
1273 		}
1274 	}
1275 	ret = 0;
1276 unlock:
1277 	spin_unlock(&fs_info->qgroup_lock);
1278 out:
1279 	ulist_free(roots);
1280 	ulist_free(tmp);
1281 
1282 	return ret;
1283 }
1284 
1285 /*
1286  * called from commit_transaction. Writes all changed qgroups to disk.
1287  */
1288 int btrfs_run_qgroups(struct btrfs_trans_handle *trans,
1289 		      struct btrfs_fs_info *fs_info)
1290 {
1291 	struct btrfs_root *quota_root = fs_info->quota_root;
1292 	int ret = 0;
1293 
1294 	if (!quota_root)
1295 		goto out;
1296 
1297 	fs_info->quota_enabled = fs_info->pending_quota_state;
1298 
1299 	spin_lock(&fs_info->qgroup_lock);
1300 	while (!list_empty(&fs_info->dirty_qgroups)) {
1301 		struct btrfs_qgroup *qgroup;
1302 		qgroup = list_first_entry(&fs_info->dirty_qgroups,
1303 					  struct btrfs_qgroup, dirty);
1304 		list_del_init(&qgroup->dirty);
1305 		spin_unlock(&fs_info->qgroup_lock);
1306 		ret = update_qgroup_info_item(trans, quota_root, qgroup);
1307 		if (ret)
1308 			fs_info->qgroup_flags |=
1309 					BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT;
1310 		spin_lock(&fs_info->qgroup_lock);
1311 	}
1312 	if (fs_info->quota_enabled)
1313 		fs_info->qgroup_flags |= BTRFS_QGROUP_STATUS_FLAG_ON;
1314 	else
1315 		fs_info->qgroup_flags &= ~BTRFS_QGROUP_STATUS_FLAG_ON;
1316 	spin_unlock(&fs_info->qgroup_lock);
1317 
1318 	ret = update_qgroup_status_item(trans, fs_info, quota_root);
1319 	if (ret)
1320 		fs_info->qgroup_flags |= BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT;
1321 
1322 out:
1323 
1324 	return ret;
1325 }
1326 
1327 /*
1328  * copy the acounting information between qgroups. This is necessary when a
1329  * snapshot or a subvolume is created
1330  */
1331 int btrfs_qgroup_inherit(struct btrfs_trans_handle *trans,
1332 			 struct btrfs_fs_info *fs_info, u64 srcid, u64 objectid,
1333 			 struct btrfs_qgroup_inherit *inherit)
1334 {
1335 	int ret = 0;
1336 	int i;
1337 	u64 *i_qgroups;
1338 	struct btrfs_root *quota_root = fs_info->quota_root;
1339 	struct btrfs_qgroup *srcgroup;
1340 	struct btrfs_qgroup *dstgroup;
1341 	u32 level_size = 0;
1342 
1343 	if (!fs_info->quota_enabled)
1344 		return 0;
1345 
1346 	if (!quota_root)
1347 		return -EINVAL;
1348 
1349 	/*
1350 	 * create a tracking group for the subvol itself
1351 	 */
1352 	ret = add_qgroup_item(trans, quota_root, objectid);
1353 	if (ret)
1354 		goto out;
1355 
1356 	if (inherit && inherit->flags & BTRFS_QGROUP_INHERIT_SET_LIMITS) {
1357 		ret = update_qgroup_limit_item(trans, quota_root, objectid,
1358 					       inherit->lim.flags,
1359 					       inherit->lim.max_rfer,
1360 					       inherit->lim.max_excl,
1361 					       inherit->lim.rsv_rfer,
1362 					       inherit->lim.rsv_excl);
1363 		if (ret)
1364 			goto out;
1365 	}
1366 
1367 	if (srcid) {
1368 		struct btrfs_root *srcroot;
1369 		struct btrfs_key srckey;
1370 		int srcroot_level;
1371 
1372 		srckey.objectid = srcid;
1373 		srckey.type = BTRFS_ROOT_ITEM_KEY;
1374 		srckey.offset = (u64)-1;
1375 		srcroot = btrfs_read_fs_root_no_name(fs_info, &srckey);
1376 		if (IS_ERR(srcroot)) {
1377 			ret = PTR_ERR(srcroot);
1378 			goto out;
1379 		}
1380 
1381 		rcu_read_lock();
1382 		srcroot_level = btrfs_header_level(srcroot->node);
1383 		level_size = btrfs_level_size(srcroot, srcroot_level);
1384 		rcu_read_unlock();
1385 	}
1386 
1387 	/*
1388 	 * add qgroup to all inherited groups
1389 	 */
1390 	if (inherit) {
1391 		i_qgroups = (u64 *)(inherit + 1);
1392 		for (i = 0; i < inherit->num_qgroups; ++i) {
1393 			ret = add_qgroup_relation_item(trans, quota_root,
1394 						       objectid, *i_qgroups);
1395 			if (ret)
1396 				goto out;
1397 			ret = add_qgroup_relation_item(trans, quota_root,
1398 						       *i_qgroups, objectid);
1399 			if (ret)
1400 				goto out;
1401 			++i_qgroups;
1402 		}
1403 	}
1404 
1405 
1406 	spin_lock(&fs_info->qgroup_lock);
1407 
1408 	dstgroup = add_qgroup_rb(fs_info, objectid);
1409 	if (IS_ERR(dstgroup)) {
1410 		ret = PTR_ERR(dstgroup);
1411 		goto unlock;
1412 	}
1413 
1414 	if (srcid) {
1415 		srcgroup = find_qgroup_rb(fs_info, srcid);
1416 		if (!srcgroup)
1417 			goto unlock;
1418 		dstgroup->rfer = srcgroup->rfer - level_size;
1419 		dstgroup->rfer_cmpr = srcgroup->rfer_cmpr - level_size;
1420 		srcgroup->excl = level_size;
1421 		srcgroup->excl_cmpr = level_size;
1422 		qgroup_dirty(fs_info, dstgroup);
1423 		qgroup_dirty(fs_info, srcgroup);
1424 	}
1425 
1426 	if (!inherit)
1427 		goto unlock;
1428 
1429 	i_qgroups = (u64 *)(inherit + 1);
1430 	for (i = 0; i < inherit->num_qgroups; ++i) {
1431 		ret = add_relation_rb(quota_root->fs_info, objectid,
1432 				      *i_qgroups);
1433 		if (ret)
1434 			goto unlock;
1435 		++i_qgroups;
1436 	}
1437 
1438 	for (i = 0; i <  inherit->num_ref_copies; ++i) {
1439 		struct btrfs_qgroup *src;
1440 		struct btrfs_qgroup *dst;
1441 
1442 		src = find_qgroup_rb(fs_info, i_qgroups[0]);
1443 		dst = find_qgroup_rb(fs_info, i_qgroups[1]);
1444 
1445 		if (!src || !dst) {
1446 			ret = -EINVAL;
1447 			goto unlock;
1448 		}
1449 
1450 		dst->rfer = src->rfer - level_size;
1451 		dst->rfer_cmpr = src->rfer_cmpr - level_size;
1452 		i_qgroups += 2;
1453 	}
1454 	for (i = 0; i <  inherit->num_excl_copies; ++i) {
1455 		struct btrfs_qgroup *src;
1456 		struct btrfs_qgroup *dst;
1457 
1458 		src = find_qgroup_rb(fs_info, i_qgroups[0]);
1459 		dst = find_qgroup_rb(fs_info, i_qgroups[1]);
1460 
1461 		if (!src || !dst) {
1462 			ret = -EINVAL;
1463 			goto unlock;
1464 		}
1465 
1466 		dst->excl = src->excl + level_size;
1467 		dst->excl_cmpr = src->excl_cmpr + level_size;
1468 		i_qgroups += 2;
1469 	}
1470 
1471 unlock:
1472 	spin_unlock(&fs_info->qgroup_lock);
1473 out:
1474 	return ret;
1475 }
1476 
1477 /*
1478  * reserve some space for a qgroup and all its parents. The reservation takes
1479  * place with start_transaction or dealloc_reserve, similar to ENOSPC
1480  * accounting. If not enough space is available, EDQUOT is returned.
1481  * We assume that the requested space is new for all qgroups.
1482  */
1483 int btrfs_qgroup_reserve(struct btrfs_root *root, u64 num_bytes)
1484 {
1485 	struct btrfs_root *quota_root;
1486 	struct btrfs_qgroup *qgroup;
1487 	struct btrfs_fs_info *fs_info = root->fs_info;
1488 	u64 ref_root = root->root_key.objectid;
1489 	int ret = 0;
1490 	struct ulist *ulist = NULL;
1491 	struct ulist_node *unode;
1492 	struct ulist_iterator uiter;
1493 
1494 	if (!is_fstree(ref_root))
1495 		return 0;
1496 
1497 	if (num_bytes == 0)
1498 		return 0;
1499 
1500 	spin_lock(&fs_info->qgroup_lock);
1501 	quota_root = fs_info->quota_root;
1502 	if (!quota_root)
1503 		goto out;
1504 
1505 	qgroup = find_qgroup_rb(fs_info, ref_root);
1506 	if (!qgroup)
1507 		goto out;
1508 
1509 	/*
1510 	 * in a first step, we check all affected qgroups if any limits would
1511 	 * be exceeded
1512 	 */
1513 	ulist = ulist_alloc(GFP_ATOMIC);
1514 	if (!ulist) {
1515 		ret = -ENOMEM;
1516 		goto out;
1517 	}
1518 	ulist_add(ulist, qgroup->qgroupid, (uintptr_t)qgroup, GFP_ATOMIC);
1519 	ULIST_ITER_INIT(&uiter);
1520 	while ((unode = ulist_next(ulist, &uiter))) {
1521 		struct btrfs_qgroup *qg;
1522 		struct btrfs_qgroup_list *glist;
1523 
1524 		qg = (struct btrfs_qgroup *)(uintptr_t)unode->aux;
1525 
1526 		if ((qg->lim_flags & BTRFS_QGROUP_LIMIT_MAX_RFER) &&
1527 		    qg->reserved + qg->rfer + num_bytes >
1528 		    qg->max_rfer)
1529 			ret = -EDQUOT;
1530 
1531 		if ((qg->lim_flags & BTRFS_QGROUP_LIMIT_MAX_EXCL) &&
1532 		    qg->reserved + qg->excl + num_bytes >
1533 		    qg->max_excl)
1534 			ret = -EDQUOT;
1535 
1536 		list_for_each_entry(glist, &qg->groups, next_group) {
1537 			ulist_add(ulist, glist->group->qgroupid,
1538 				  (uintptr_t)glist->group, GFP_ATOMIC);
1539 		}
1540 	}
1541 	if (ret)
1542 		goto out;
1543 
1544 	/*
1545 	 * no limits exceeded, now record the reservation into all qgroups
1546 	 */
1547 	ULIST_ITER_INIT(&uiter);
1548 	while ((unode = ulist_next(ulist, &uiter))) {
1549 		struct btrfs_qgroup *qg;
1550 
1551 		qg = (struct btrfs_qgroup *)(uintptr_t)unode->aux;
1552 
1553 		qg->reserved += num_bytes;
1554 	}
1555 
1556 out:
1557 	spin_unlock(&fs_info->qgroup_lock);
1558 	ulist_free(ulist);
1559 
1560 	return ret;
1561 }
1562 
1563 void btrfs_qgroup_free(struct btrfs_root *root, u64 num_bytes)
1564 {
1565 	struct btrfs_root *quota_root;
1566 	struct btrfs_qgroup *qgroup;
1567 	struct btrfs_fs_info *fs_info = root->fs_info;
1568 	struct ulist *ulist = NULL;
1569 	struct ulist_node *unode;
1570 	struct ulist_iterator uiter;
1571 	u64 ref_root = root->root_key.objectid;
1572 
1573 	if (!is_fstree(ref_root))
1574 		return;
1575 
1576 	if (num_bytes == 0)
1577 		return;
1578 
1579 	spin_lock(&fs_info->qgroup_lock);
1580 
1581 	quota_root = fs_info->quota_root;
1582 	if (!quota_root)
1583 		goto out;
1584 
1585 	qgroup = find_qgroup_rb(fs_info, ref_root);
1586 	if (!qgroup)
1587 		goto out;
1588 
1589 	ulist = ulist_alloc(GFP_ATOMIC);
1590 	if (!ulist) {
1591 		btrfs_std_error(fs_info, -ENOMEM);
1592 		goto out;
1593 	}
1594 	ulist_add(ulist, qgroup->qgroupid, (uintptr_t)qgroup, GFP_ATOMIC);
1595 	ULIST_ITER_INIT(&uiter);
1596 	while ((unode = ulist_next(ulist, &uiter))) {
1597 		struct btrfs_qgroup *qg;
1598 		struct btrfs_qgroup_list *glist;
1599 
1600 		qg = (struct btrfs_qgroup *)(uintptr_t)unode->aux;
1601 
1602 		qg->reserved -= num_bytes;
1603 
1604 		list_for_each_entry(glist, &qg->groups, next_group) {
1605 			ulist_add(ulist, glist->group->qgroupid,
1606 				  (uintptr_t)glist->group, GFP_ATOMIC);
1607 		}
1608 	}
1609 
1610 out:
1611 	spin_unlock(&fs_info->qgroup_lock);
1612 	ulist_free(ulist);
1613 }
1614 
1615 void assert_qgroups_uptodate(struct btrfs_trans_handle *trans)
1616 {
1617 	if (list_empty(&trans->qgroup_ref_list) && !trans->delayed_ref_elem.seq)
1618 		return;
1619 	printk(KERN_ERR "btrfs: qgroups not uptodate in trans handle %p: list is%s empty, seq is %llu\n",
1620 		trans, list_empty(&trans->qgroup_ref_list) ? "" : " not",
1621 		trans->delayed_ref_elem.seq);
1622 	BUG();
1623 }
1624