[mirror_ubuntu-bionic-kernel.git] / fs / btrfs / delayed-ref.h

/*
 * Copyright (C) 2008 Oracle.  All rights reserved.
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of the GNU General Public
 * License v2 as published by the Free Software Foundation.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
 * General Public License for more details.
 *
 * You should have received a copy of the GNU General Public
 * License along with this program; if not, write to the
 * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
 * Boston, MA 021110-1307, USA.
 */
#ifndef __DELAYED_REF__
#define __DELAYED_REF__

#include <linux/refcount.h>

/* these are the possible values of struct btrfs_delayed_ref_node->action */
#define BTRFS_ADD_DELAYED_REF    1 /* add one backref to the tree */
#define BTRFS_DROP_DELAYED_REF   2 /* delete one backref from the tree */
#define BTRFS_ADD_DELAYED_EXTENT 3 /* record a full extent allocation */
#define BTRFS_UPDATE_DELAYED_HEAD 4 /* not changing ref count on head ref */

struct btrfs_delayed_ref_node {
	struct rb_node ref_node;
	/*
	 * If action is BTRFS_ADD_DELAYED_REF, also link this node to
	 * ref_head->ref_add_list, then we do not need to iterate the
	 * whole ref_head->ref_list to find BTRFS_ADD_DELAYED_REF nodes.
	 */
	struct list_head add_list;

	/* the starting bytenr of the extent */
	u64 bytenr;

	/* the size of the extent */
	u64 num_bytes;

	/* seq number to keep track of insertion order */
	u64 seq;

	/* ref count on this data structure */
	refcount_t refs;

	/*
	 * how many refs is this entry adding or deleting.  For
	 * head refs, this may be a negative number because it is keeping
	 * track of the total mods done to the reference count.
	 * For individual refs, this will always be a positive number
	 *
	 * It may be more than one, since it is possible for a single
	 * parent to have more than one ref on an extent
	 */
	int ref_mod;

	unsigned int action:8;
	unsigned int type:8;
	/* is this node still in the rbtree? */
	unsigned int is_head:1;
	unsigned int in_tree:1;
};

struct btrfs_delayed_extent_op {
	struct btrfs_disk_key key;
	u8 level;
	bool update_key;
	bool update_flags;
	bool is_data;
	u64 flags_to_set;
};

/*
 * the head refs are used to hold a lock on a given extent, which allows us
 * to make sure that only one process is running the delayed refs
 * at a time for a single extent.  They also store the sum of all the
 * reference count modifications we've queued up.
 */
struct btrfs_delayed_ref_head {
	u64 bytenr;
	u64 num_bytes;
	refcount_t refs;
	/*
	 * the mutex is held while running the refs, and it is also
	 * held when checking the sum of reference modifications.
	 */
	struct mutex mutex;

	spinlock_t lock;
	struct rb_root ref_tree;
	/* accumulate add BTRFS_ADD_DELAYED_REF nodes to this ref_add_list. */
	struct list_head ref_add_list;

	struct rb_node href_node;

	struct btrfs_delayed_extent_op *extent_op;

	/*
	 * This is used to track the final ref_mod from all the refs associated
	 * with this head ref, this is not adjusted as delayed refs are run,
	 * this is meant to track if we need to do the csum accounting or not.
	 */
	int total_ref_mod;

	/*
	 * This is the current outstanding mod references for this bytenr.  This
	 * is used with lookup_extent_info to get an accurate reference count
	 * for a bytenr, so it is adjusted as delayed refs are run so that any
	 * on disk reference count + ref_mod is accurate.
	 */
	int ref_mod;

	/*
	 * For qgroup reserved space freeing.
	 *
	 * ref_root and reserved will be recorded after
	 * BTRFS_ADD_DELAYED_EXTENT is called.
	 * And will be used to free reserved qgroup space at
	 * run_delayed_refs() time.
	 */
	u64 qgroup_ref_root;
	u64 qgroup_reserved;

	/*
	 * when a new extent is allocated, it is just reserved in memory
	 * The actual extent isn't inserted into the extent allocation tree
	 * until the delayed ref is processed.  must_insert_reserved is
	 * used to flag a delayed ref so the accounting can be updated
	 * when a full insert is done.
	 *
	 * It is possible the extent will be freed before it is ever
	 * inserted into the extent allocation tree.  In this case
	 * we need to update the in ram accounting to properly reflect
	 * the free has happened.
	 */
	unsigned int must_insert_reserved:1;
	unsigned int is_data:1;
	unsigned int is_system:1;
	unsigned int processing:1;
};

struct btrfs_delayed_tree_ref {
	struct btrfs_delayed_ref_node node;
	u64 root;
	u64 parent;
	int level;
};

struct btrfs_delayed_data_ref {
	struct btrfs_delayed_ref_node node;
	u64 root;
	u64 parent;
	u64 objectid;
	u64 offset;
};

struct btrfs_delayed_ref_root {
	/* head ref rbtree */
	struct rb_root href_root;

	/* dirty extent records */
	struct rb_root dirty_extent_root;

	/* this spin lock protects the rbtree and the entries inside */
	spinlock_t lock;

	/* how many delayed ref updates we've queued, used by the
	 * throttling code
	 */
	atomic_t num_entries;

	/* total number of head nodes in tree */
	unsigned long num_heads;

	/* total number of head nodes ready for processing */
	unsigned long num_heads_ready;

	u64 pending_csums;

	/*
	 * set when the tree is flushing before a transaction commit,
	 * used by the throttling code to decide if new updates need
	 * to be run right away
	 */
	int flushing;

	u64 run_delayed_start;

	/*
	 * To make qgroup to skip given root.
	 * This is for snapshot, as btrfs_qgroup_inherit() will manually
	 * modify counters for snapshot and its source, so we should skip
	 * the snapshot in new_root/old_roots or it will get calculated twice
	 */
	u64 qgroup_to_skip;
};

extern struct kmem_cache *btrfs_delayed_ref_head_cachep;
extern struct kmem_cache *btrfs_delayed_tree_ref_cachep;
extern struct kmem_cache *btrfs_delayed_data_ref_cachep;
extern struct kmem_cache *btrfs_delayed_extent_op_cachep;

int btrfs_delayed_ref_init(void);
void btrfs_delayed_ref_exit(void);

static inline struct btrfs_delayed_extent_op *
btrfs_alloc_delayed_extent_op(void)
{
	return kmem_cache_alloc(btrfs_delayed_extent_op_cachep, GFP_NOFS);
}

static inline void
btrfs_free_delayed_extent_op(struct btrfs_delayed_extent_op *op)
{
	if (op)
		kmem_cache_free(btrfs_delayed_extent_op_cachep, op);
}

static inline void btrfs_put_delayed_ref(struct btrfs_delayed_ref_node *ref)
{
	WARN_ON(refcount_read(&ref->refs) == 0);
	if (refcount_dec_and_test(&ref->refs)) {
		WARN_ON(ref->in_tree);
		switch (ref->type) {
		case BTRFS_TREE_BLOCK_REF_KEY:
		case BTRFS_SHARED_BLOCK_REF_KEY:
			kmem_cache_free(btrfs_delayed_tree_ref_cachep, ref);
			break;
		case BTRFS_EXTENT_DATA_REF_KEY:
		case BTRFS_SHARED_DATA_REF_KEY:
			kmem_cache_free(btrfs_delayed_data_ref_cachep, ref);
			break;
		default:
			BUG();
		}
	}
}

static inline void btrfs_put_delayed_ref_head(struct btrfs_delayed_ref_head *head)
{
	if (refcount_dec_and_test(&head->refs))
		kmem_cache_free(btrfs_delayed_ref_head_cachep, head);
}

int btrfs_add_delayed_tree_ref(struct btrfs_fs_info *fs_info,
			       struct btrfs_trans_handle *trans,
			       u64 bytenr, u64 num_bytes, u64 parent,
			       u64 ref_root, int level, int action,
			       struct btrfs_delayed_extent_op *extent_op,
			       int *old_ref_mod, int *new_ref_mod);
int btrfs_add_delayed_data_ref(struct btrfs_fs_info *fs_info,
			       struct btrfs_trans_handle *trans,
			       u64 bytenr, u64 num_bytes,
			       u64 parent, u64 ref_root,
			       u64 owner, u64 offset, u64 reserved, int action,
			       int *old_ref_mod, int *new_ref_mod);
int btrfs_add_delayed_extent_op(struct btrfs_fs_info *fs_info,
				struct btrfs_trans_handle *trans,
				u64 bytenr, u64 num_bytes,
				struct btrfs_delayed_extent_op *extent_op);
void btrfs_merge_delayed_refs(struct btrfs_trans_handle *trans,
			      struct btrfs_fs_info *fs_info,
			      struct btrfs_delayed_ref_root *delayed_refs,
			      struct btrfs_delayed_ref_head *head);

struct btrfs_delayed_ref_head *
btrfs_find_delayed_ref_head(struct btrfs_delayed_ref_root *delayed_refs,
			    u64 bytenr);
int btrfs_delayed_ref_lock(struct btrfs_trans_handle *trans,
			   struct btrfs_delayed_ref_head *head);
static inline void btrfs_delayed_ref_unlock(struct btrfs_delayed_ref_head *head)
{
	mutex_unlock(&head->mutex);
}


struct btrfs_delayed_ref_head *
btrfs_select_ref_head(struct btrfs_trans_handle *trans);

int btrfs_check_delayed_seq(struct btrfs_fs_info *fs_info,
			    struct btrfs_delayed_ref_root *delayed_refs,
			    u64 seq);

/*
 * helper functions to cast a node into its container
 */
static inline struct btrfs_delayed_tree_ref *
btrfs_delayed_node_to_tree_ref(struct btrfs_delayed_ref_node *node)
{
	return container_of(node, struct btrfs_delayed_tree_ref, node);
}

static inline struct btrfs_delayed_data_ref *
btrfs_delayed_node_to_data_ref(struct btrfs_delayed_ref_node *node)
{
	return container_of(node, struct btrfs_delayed_data_ref, node);
}
#endif
Commit	Line	Data
56bec294 CM	1	/*
	2	* Copyright (C) 2008 Oracle. 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	#ifndef __DELAYED_REF__
	19	#define __DELAYED_REF__
	20
6df8cdf5 ER	21	#include <linux/refcount.h>
6df8cdf5 ER	22
44a075bd	23	/* these are the possible values of struct btrfs_delayed_ref_node->action */
56bec294 CM	24	#define BTRFS_ADD_DELAYED_REF 1 /* add one backref to the tree */
	25	#define BTRFS_DROP_DELAYED_REF 2 /* delete one backref from the tree */
	26	#define BTRFS_ADD_DELAYED_EXTENT 3 /* record a full extent allocation */
1a81af4d	27	#define BTRFS_UPDATE_DELAYED_HEAD 4 /* not changing ref count on head ref */
56bec294 CM	28
56bec294 CM	29	struct btrfs_delayed_ref_node {
0e0adbcf	30	struct rb_node ref_node;
1d57ee94 WX	31	/*
	32	* If action is BTRFS_ADD_DELAYED_REF, also link this node to
	33	* ref_head->ref_add_list, then we do not need to iterate the
	34	* whole ref_head->ref_list to find BTRFS_ADD_DELAYED_REF nodes.
	35	*/
	36	struct list_head add_list;
c6fc2454	37
56bec294 CM	38	/* the starting bytenr of the extent */
	39	u64 bytenr;
	40
56bec294 CM	41	/* the size of the extent */
	42	u64 num_bytes;
	43
00f04b88 AJ	44	/* seq number to keep track of insertion order */
	45	u64 seq;
	46
56bec294	47	/* ref count on this data structure */
6df8cdf5	48	refcount_t refs;
56bec294 CM	49
	50	/*
	51	* how many refs is this entry adding or deleting. For
	52	* head refs, this may be a negative number because it is keeping
	53	* track of the total mods done to the reference count.
	54	* For individual refs, this will always be a positive number
	55	*
	56	* It may be more than one, since it is possible for a single
	57	* parent to have more than one ref on an extent
	58	*/
	59	int ref_mod;
	60
5d4f98a2 YZ	61	unsigned int action:8;
5d4f98a2 YZ	62	unsigned int type:8;
56bec294	63	/* is this node still in the rbtree? */
5d4f98a2	64	unsigned int is_head:1;
56bec294 CM	65	unsigned int in_tree:1;
	66	};
	67
5d4f98a2 YZ	68	struct btrfs_delayed_extent_op {
5d4f98a2 YZ	69	struct btrfs_disk_key key;
35b3ad50 DS	70	u8 level;
	71	bool update_key;
	72	bool update_flags;
	73	bool is_data;
5d4f98a2	74	u64 flags_to_set;
5d4f98a2 YZ	75	};
5d4f98a2 YZ	76
56bec294 CM	77	/*
	78	* the head refs are used to hold a lock on a given extent, which allows us
	79	* to make sure that only one process is running the delayed refs
	80	* at a time for a single extent. They also store the sum of all the
	81	* reference count modifications we've queued up.
	82	*/
	83	struct btrfs_delayed_ref_head {
d278850e JB	84	u64 bytenr;
	85	u64 num_bytes;
	86	refcount_t refs;
56bec294 CM	87	/*
	88	* the mutex is held while running the refs, and it is also
	89	* held when checking the sum of reference modifications.
	90	*/
	91	struct mutex mutex;
	92
d7df2c79	93	spinlock_t lock;
0e0adbcf	94	struct rb_root ref_tree;
1d57ee94 WX	95	/* accumulate add BTRFS_ADD_DELAYED_REF nodes to this ref_add_list. */
1d57ee94 WX	96	struct list_head ref_add_list;
c3e69d58	97
c46effa6 LB	98	struct rb_node href_node;
c46effa6 LB	99
5d4f98a2	100	struct btrfs_delayed_extent_op *extent_op;
1262133b JB	101
	102	/*
	103	* This is used to track the final ref_mod from all the refs associated
	104	* with this head ref, this is not adjusted as delayed refs are run,
	105	* this is meant to track if we need to do the csum accounting or not.
	106	*/
	107	int total_ref_mod;
	108
d278850e JB	109	/*
	110	* This is the current outstanding mod references for this bytenr. This
	111	* is used with lookup_extent_info to get an accurate reference count
	112	* for a bytenr, so it is adjusted as delayed refs are run so that any
	113	* on disk reference count + ref_mod is accurate.
	114	*/
	115	int ref_mod;
	116
f64d5ca8 QW	117	/*
	118	* For qgroup reserved space freeing.
	119	*
	120	* ref_root and reserved will be recorded after
	121	* BTRFS_ADD_DELAYED_EXTENT is called.
	122	* And will be used to free reserved qgroup space at
	123	* run_delayed_refs() time.
	124	*/
	125	u64 qgroup_ref_root;
	126	u64 qgroup_reserved;
	127
56bec294 CM	128	/*
	129	* when a new extent is allocated, it is just reserved in memory
	130	* The actual extent isn't inserted into the extent allocation tree
	131	* until the delayed ref is processed. must_insert_reserved is
	132	* used to flag a delayed ref so the accounting can be updated
	133	* when a full insert is done.
	134	*
	135	* It is possible the extent will be freed before it is ever
	136	* inserted into the extent allocation tree. In this case
	137	* we need to update the in ram accounting to properly reflect
	138	* the free has happened.
	139	*/
	140	unsigned int must_insert_reserved:1;
5d4f98a2	141	unsigned int is_data:1;
d59e2961	142	unsigned int is_system:1;
d7df2c79	143	unsigned int processing:1;
56bec294 CM	144	};
56bec294 CM	145
5d4f98a2	146	struct btrfs_delayed_tree_ref {
56bec294	147	struct btrfs_delayed_ref_node node;
eebe063b AJ	148	u64 root;
eebe063b AJ	149	u64 parent;
5d4f98a2 YZ	150	int level;
5d4f98a2 YZ	151	};
56bec294	152
5d4f98a2 YZ	153	struct btrfs_delayed_data_ref {
5d4f98a2 YZ	154	struct btrfs_delayed_ref_node node;
eebe063b AJ	155	u64 root;
eebe063b AJ	156	u64 parent;
5d4f98a2 YZ	157	u64 objectid;
5d4f98a2 YZ	158	u64 offset;
56bec294 CM	159	};
	160
	161	struct btrfs_delayed_ref_root {
c46effa6 LB	162	/* head ref rbtree */
	163	struct rb_root href_root;
	164
3368d001 QW	165	/* dirty extent records */
	166	struct rb_root dirty_extent_root;
	167
56bec294 CM	168	/* this spin lock protects the rbtree and the entries inside */
	169	spinlock_t lock;
	170
	171	/* how many delayed ref updates we've queued, used by the
	172	* throttling code
	173	*/
d7df2c79	174	atomic_t num_entries;
56bec294	175
c3e69d58 CM	176	/* total number of head nodes in tree */
	177	unsigned long num_heads;
	178
	179	/* total number of head nodes ready for processing */
	180	unsigned long num_heads_ready;
	181
1262133b JB	182	u64 pending_csums;
1262133b JB	183
56bec294 CM	184	/*
	185	* set when the tree is flushing before a transaction commit,
	186	* used by the throttling code to decide if new updates need
	187	* to be run right away
	188	*/
	189	int flushing;
c3e69d58 CM	190
c3e69d58 CM	191	u64 run_delayed_start;
9086db86 QW	192
	193	/*
	194	* To make qgroup to skip given root.
01327610	195	* This is for snapshot, as btrfs_qgroup_inherit() will manually
9086db86 QW	196	* modify counters for snapshot and its source, so we should skip
	197	* the snapshot in new_root/old_roots or it will get calculated twice
	198	*/
	199	u64 qgroup_to_skip;
56bec294 CM	200	};
56bec294 CM	201
78a6184a MX	202	extern struct kmem_cache *btrfs_delayed_ref_head_cachep;
	203	extern struct kmem_cache *btrfs_delayed_tree_ref_cachep;
	204	extern struct kmem_cache *btrfs_delayed_data_ref_cachep;
	205	extern struct kmem_cache *btrfs_delayed_extent_op_cachep;
	206
	207	int btrfs_delayed_ref_init(void);
	208	void btrfs_delayed_ref_exit(void);
	209
	210	static inline struct btrfs_delayed_extent_op *
	211	btrfs_alloc_delayed_extent_op(void)
	212	{
	213	return kmem_cache_alloc(btrfs_delayed_extent_op_cachep, GFP_NOFS);
	214	}
	215
	216	static inline void
	217	btrfs_free_delayed_extent_op(struct btrfs_delayed_extent_op *op)
	218	{
	219	if (op)
	220	kmem_cache_free(btrfs_delayed_extent_op_cachep, op);
	221	}
	222
56bec294 CM	223	static inline void btrfs_put_delayed_ref(struct btrfs_delayed_ref_node *ref)
56bec294 CM	224	{
6df8cdf5 ER	225	WARN_ON(refcount_read(&ref->refs) == 0);
6df8cdf5 ER	226	if (refcount_dec_and_test(&ref->refs)) {
56bec294	227	WARN_ON(ref->in_tree);
78a6184a MX	228	switch (ref->type) {
	229	case BTRFS_TREE_BLOCK_REF_KEY:
	230	case BTRFS_SHARED_BLOCK_REF_KEY:
	231	kmem_cache_free(btrfs_delayed_tree_ref_cachep, ref);
	232	break;
	233	case BTRFS_EXTENT_DATA_REF_KEY:
	234	case BTRFS_SHARED_DATA_REF_KEY:
	235	kmem_cache_free(btrfs_delayed_data_ref_cachep, ref);
	236	break;
78a6184a MX	237	default:
	238	BUG();
	239	}
56bec294 CM	240	}
	241	}
	242
d278850e JB	243	static inline void btrfs_put_delayed_ref_head(struct btrfs_delayed_ref_head *head)
	244	{
	245	if (refcount_dec_and_test(&head->refs))
	246	kmem_cache_free(btrfs_delayed_ref_head_cachep, head);
	247	}
	248
66d7e7f0 AJ	249	int btrfs_add_delayed_tree_ref(struct btrfs_fs_info *fs_info,
66d7e7f0 AJ	250	struct btrfs_trans_handle *trans,
5d4f98a2 YZ	251	u64 bytenr, u64 num_bytes, u64 parent,
5d4f98a2 YZ	252	u64 ref_root, int level, int action,
7be07912 OS	253	struct btrfs_delayed_extent_op *extent_op,
7be07912 OS	254	int old_ref_mod, int new_ref_mod);
66d7e7f0 AJ	255	int btrfs_add_delayed_data_ref(struct btrfs_fs_info *fs_info,
66d7e7f0 AJ	256	struct btrfs_trans_handle *trans,
5d4f98a2 YZ	257	u64 bytenr, u64 num_bytes,
5d4f98a2 YZ	258	u64 parent, u64 ref_root,
7be07912 OS	259	u64 owner, u64 offset, u64 reserved, int action,
7be07912 OS	260	int old_ref_mod, int new_ref_mod);
66d7e7f0 AJ	261	int btrfs_add_delayed_extent_op(struct btrfs_fs_info *fs_info,
66d7e7f0 AJ	262	struct btrfs_trans_handle *trans,
5d4f98a2 YZ	263	u64 bytenr, u64 num_bytes,
5d4f98a2 YZ	264	struct btrfs_delayed_extent_op *extent_op);
ae1e206b JB	265	void btrfs_merge_delayed_refs(struct btrfs_trans_handle *trans,
	266	struct btrfs_fs_info *fs_info,
	267	struct btrfs_delayed_ref_root *delayed_refs,
	268	struct btrfs_delayed_ref_head *head);
56bec294	269
1887be66	270	struct btrfs_delayed_ref_head *
f72ad18e LB	271	btrfs_find_delayed_ref_head(struct btrfs_delayed_ref_root *delayed_refs,
f72ad18e LB	272	u64 bytenr);
c3e69d58 CM	273	int btrfs_delayed_ref_lock(struct btrfs_trans_handle *trans,
c3e69d58 CM	274	struct btrfs_delayed_ref_head *head);
093486c4 MX	275	static inline void btrfs_delayed_ref_unlock(struct btrfs_delayed_ref_head *head)
	276	{
	277	mutex_unlock(&head->mutex);
	278	}
	279
d7df2c79 JB	280
	281	struct btrfs_delayed_ref_head *
	282	btrfs_select_ref_head(struct btrfs_trans_handle *trans);
00f04b88	283
097b8a7c JS	284	int btrfs_check_delayed_seq(struct btrfs_fs_info *fs_info,
097b8a7c JS	285	struct btrfs_delayed_ref_root *delayed_refs,
00f04b88 AJ	286	u64 seq);
00f04b88 AJ	287
56bec294 CM	288	/*
	289	* helper functions to cast a node into its container
	290	*/
5d4f98a2 YZ	291	static inline struct btrfs_delayed_tree_ref *
5d4f98a2 YZ	292	btrfs_delayed_node_to_tree_ref(struct btrfs_delayed_ref_node *node)
56bec294	293	{
5d4f98a2 YZ	294	return container_of(node, struct btrfs_delayed_tree_ref, node);
5d4f98a2 YZ	295	}
56bec294	296
5d4f98a2 YZ	297	static inline struct btrfs_delayed_data_ref *
	298	btrfs_delayed_node_to_data_ref(struct btrfs_delayed_ref_node *node)
	299	{
5d4f98a2	300	return container_of(node, struct btrfs_delayed_data_ref, node);
56bec294	301	}
56bec294	302	#endif