1/*
2 * Copyright (C) 2007 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
19#ifndef __BTRFS_ORDERED_DATA__
20#define __BTRFS_ORDERED_DATA__
21
22/* one of these per inode */
23struct btrfs_ordered_inode_tree {
24	spinlock_t lock;
25	struct rb_root tree;
26	struct rb_node *last;
27};
28
29struct btrfs_ordered_sum {
30	/* bytenr is the start of this extent on disk */
31	u64 bytenr;
32
33	/*
34	 * this is the length in bytes covered by the sums array below.
35	 */
36	int len;
37	struct list_head list;
38	/* last field is a variable length array of csums */
39	u32 sums[];
40};
41
42/*
43 * bits for the flags field:
44 *
45 * BTRFS_ORDERED_IO_DONE is set when all of the blocks are written.
46 * It is used to make sure metadata is inserted into the tree only once
47 * per extent.
48 *
49 * BTRFS_ORDERED_COMPLETE is set when the extent is removed from the
50 * rbtree, just before waking any waiters.  It is used to indicate the
51 * IO is done and any metadata is inserted into the tree.
52 */
53#define BTRFS_ORDERED_IO_DONE 0 /* set when all the pages are written */
54
55#define BTRFS_ORDERED_COMPLETE 1 /* set when removed from the tree */
56
57#define BTRFS_ORDERED_NOCOW 2 /* set when we want to write in place */
58
59#define BTRFS_ORDERED_COMPRESSED 3 /* writing a zlib compressed extent */
60
61#define BTRFS_ORDERED_PREALLOC 4 /* set when writing to prealloced extent */
62
63#define BTRFS_ORDERED_DIRECT 5 /* set when we're doing DIO with this extent */
64
65#define BTRFS_ORDERED_IOERR 6 /* We had an io error when writing this out */
66
67#define BTRFS_ORDERED_UPDATED_ISIZE 7 /* indicates whether this ordered extent
68				       * has done its due diligence in updating
69				       * the isize. */
70#define BTRFS_ORDERED_LOGGED_CSUM 8 /* We've logged the csums on this ordered
71				       ordered extent */
72#define BTRFS_ORDERED_TRUNCATED 9 /* Set when we have to truncate an extent */
73
74#define BTRFS_ORDERED_LOGGED 10 /* Set when we've waited on this ordered extent
75				 * in the logging code. */
76struct btrfs_ordered_extent {
77	/* logical offset in the file */
78	u64 file_offset;
79
80	/* disk byte number */
81	u64 start;
82
83	/* ram length of the extent in bytes */
84	u64 len;
85
86	/* extent length on disk */
87	u64 disk_len;
88
89	/* number of bytes that still need writing */
90	u64 bytes_left;
91
92	/* number of bytes that still need csumming */
93	u64 csum_bytes_left;
94
95	/*
96	 * the end of the ordered extent which is behind it but
97	 * didn't update disk_i_size. Please see the comment of
98	 * btrfs_ordered_update_i_size();
99	 */
100	u64 outstanding_isize;
101
102	/*
103	 * If we get truncated we need to adjust the file extent we enter for
104	 * this ordered extent so that we do not expose stale data.
105	 */
106	u64 truncated_len;
107
108	/* flags (described above) */
109	unsigned long flags;
110
111	/* compression algorithm */
112	int compress_type;
113
114	/* reference count */
115	atomic_t refs;
116
117	/* the inode we belong to */
118	struct inode *inode;
119
120	/* list of checksums for insertion when the extent io is done */
121	struct list_head list;
122
123	/* If we need to wait on this to be done */
124	struct list_head log_list;
125
126	/* If the transaction needs to wait on this ordered extent */
127	struct list_head trans_list;
128
129	/* used to wait for the BTRFS_ORDERED_COMPLETE bit */
130	wait_queue_head_t wait;
131
132	/* our friendly rbtree entry */
133	struct rb_node rb_node;
134
135	/* a per root list of all the pending ordered extents */
136	struct list_head root_extent_list;
137
138	struct btrfs_work work;
139
140	struct completion completion;
141	struct btrfs_work flush_work;
142	struct list_head work_list;
143};
144
145/*
146 * calculates the total size you need to allocate for an ordered sum
147 * structure spanning 'bytes' in the file
148 */
149static inline int btrfs_ordered_sum_size(struct btrfs_root *root,
150					 unsigned long bytes)
151{
152	int num_sectors = (int)DIV_ROUND_UP(bytes, root->sectorsize);
153	return sizeof(struct btrfs_ordered_sum) + num_sectors * sizeof(u32);
154}
155
156static inline void
157btrfs_ordered_inode_tree_init(struct btrfs_ordered_inode_tree *t)
158{
159	spin_lock_init(&t->lock);
160	t->tree = RB_ROOT;
161	t->last = NULL;
162}
163
164void btrfs_put_ordered_extent(struct btrfs_ordered_extent *entry);
165void btrfs_remove_ordered_extent(struct inode *inode,
166				struct btrfs_ordered_extent *entry);
167int btrfs_dec_test_ordered_pending(struct inode *inode,
168				   struct btrfs_ordered_extent **cached,
169				   u64 file_offset, u64 io_size, int uptodate);
170int btrfs_dec_test_first_ordered_pending(struct inode *inode,
171				   struct btrfs_ordered_extent **cached,
172				   u64 *file_offset, u64 io_size,
173				   int uptodate);
174int btrfs_add_ordered_extent(struct inode *inode, u64 file_offset,
175			     u64 start, u64 len, u64 disk_len, int type);
176int btrfs_add_ordered_extent_dio(struct inode *inode, u64 file_offset,
177				 u64 start, u64 len, u64 disk_len, int type);
178int btrfs_add_ordered_extent_compress(struct inode *inode, u64 file_offset,
179				      u64 start, u64 len, u64 disk_len,
180				      int type, int compress_type);
181void btrfs_add_ordered_sum(struct inode *inode,
182			   struct btrfs_ordered_extent *entry,
183			   struct btrfs_ordered_sum *sum);
184struct btrfs_ordered_extent *btrfs_lookup_ordered_extent(struct inode *inode,
185							 u64 file_offset);
186void btrfs_start_ordered_extent(struct inode *inode,
187				struct btrfs_ordered_extent *entry, int wait);
188int btrfs_wait_ordered_range(struct inode *inode, u64 start, u64 len);
189struct btrfs_ordered_extent *
190btrfs_lookup_first_ordered_extent(struct inode * inode, u64 file_offset);
191struct btrfs_ordered_extent *btrfs_lookup_ordered_range(struct inode *inode,
192							u64 file_offset,
193							u64 len);
194int btrfs_ordered_update_i_size(struct inode *inode, u64 offset,
195				struct btrfs_ordered_extent *ordered);
196int btrfs_find_ordered_sum(struct inode *inode, u64 offset, u64 disk_bytenr,
197			   u32 *sum, int len);
198int btrfs_wait_ordered_extents(struct btrfs_root *root, int nr);
199void btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, int nr);
200void btrfs_get_logged_extents(struct inode *inode,
201			      struct list_head *logged_list,
202			      const loff_t start,
203			      const loff_t end);
204void btrfs_put_logged_extents(struct list_head *logged_list);
205void btrfs_submit_logged_extents(struct list_head *logged_list,
206				 struct btrfs_root *log);
207void btrfs_wait_logged_extents(struct btrfs_trans_handle *trans,
208			       struct btrfs_root *log, u64 transid);
209void btrfs_free_logged_extents(struct btrfs_root *log, u64 transid);
210int __init ordered_data_init(void);
211void ordered_data_exit(void);
212#endif
213