summaryrefslogtreecommitdiffstats
path: root/fs/bcachefs/btree_update.h
blob: 60393e98084d7952771e0d02ba9246183977e233 (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
/* SPDX-License-Identifier: GPL-2.0 */
#ifndef _BCACHEFS_BTREE_UPDATE_H
#define _BCACHEFS_BTREE_UPDATE_H

#include "btree_iter.h"
#include "journal.h"

struct bch_fs;
struct btree;

void bch2_btree_node_prep_for_write(struct btree_trans *,
				    struct btree_path *, struct btree *);
bool bch2_btree_bset_insert_key(struct btree_trans *, struct btree_path *,
				struct btree *, struct btree_node_iter *,
				struct bkey_i *);

int bch2_btree_node_flush0(struct journal *, struct journal_entry_pin *, u64);
int bch2_btree_node_flush1(struct journal *, struct journal_entry_pin *, u64);
void bch2_btree_add_journal_pin(struct bch_fs *, struct btree *, u64);

void bch2_btree_insert_key_leaf(struct btree_trans *, struct btree_path *,
				struct bkey_i *, u64);

#define BCH_TRANS_COMMIT_FLAGS()							\
	x(no_enospc,	"don't check for enospc")					\
	x(no_check_rw,	"don't attempt to take a ref on c->writes")			\
	x(lazy_rw,	"go read-write if we haven't yet - only for use in recovery")	\
	x(no_journal_res, "don't take a journal reservation, instead "			\
			"pin journal entry referred to by trans->journal_res.seq")	\
	x(journal_reclaim, "operation required for journal reclaim; may return error"	\
			"instead of deadlocking if BCH_WATERMARK_reclaim not specified")\
	x(skip_accounting_apply, "we're in journal replay - accounting updates have already been applied")

enum __bch_trans_commit_flags {
	/* First bits for bch_watermark: */
	__BCH_TRANS_COMMIT_FLAGS_START = BCH_WATERMARK_BITS,
#define x(n, ...)	__BCH_TRANS_COMMIT_##n,
	BCH_TRANS_COMMIT_FLAGS()
#undef x
};

enum bch_trans_commit_flags {
#define x(n, ...)	BCH_TRANS_COMMIT_##n = BIT(__BCH_TRANS_COMMIT_##n),
	BCH_TRANS_COMMIT_FLAGS()
#undef x
};

void bch2_trans_commit_flags_to_text(struct printbuf *, enum bch_trans_commit_flags);

int bch2_btree_delete_extent_at(struct btree_trans *, struct btree_iter *,
				unsigned, unsigned);
int bch2_btree_delete_at(struct btree_trans *, struct btree_iter *, unsigned);
int bch2_btree_delete(struct btree_trans *, enum btree_id, struct bpos, unsigned);

int bch2_btree_insert_nonextent(struct btree_trans *, enum btree_id,
				struct bkey_i *, enum btree_iter_update_trigger_flags);

int bch2_btree_insert_trans(struct btree_trans *, enum btree_id, struct bkey_i *,
			enum btree_iter_update_trigger_flags);
int bch2_btree_insert(struct bch_fs *, enum btree_id, struct bkey_i *, struct
		disk_reservation *, int flags, enum
		btree_iter_update_trigger_flags iter_flags);

int bch2_btree_delete_range_trans(struct btree_trans *, enum btree_id,
				  struct bpos, struct bpos, unsigned, u64 *);
int bch2_btree_delete_range(struct bch_fs *, enum btree_id,
			    struct bpos, struct bpos, unsigned, u64 *);

int bch2_btree_bit_mod(struct btree_trans *, enum btree_id, struct bpos, bool);
int bch2_btree_bit_mod_buffered(struct btree_trans *, enum btree_id, struct bpos, bool);

static inline int bch2_btree_delete_at_buffered(struct btree_trans *trans,
						enum btree_id btree, struct bpos pos)
{
	return bch2_btree_bit_mod_buffered(trans, btree, pos, false);
}

int __bch2_insert_snapshot_whiteouts(struct btree_trans *, enum btree_id,
				     struct bpos, struct bpos);

/*
 * For use when splitting extents in existing snapshots:
 *
 * If @old_pos is an interior snapshot node, iterate over descendent snapshot
 * nodes: for every descendent snapshot in whiche @old_pos is overwritten and
 * not visible, emit a whiteout at @new_pos.
 */
static inline int bch2_insert_snapshot_whiteouts(struct btree_trans *trans,
						 enum btree_id btree,
						 struct bpos old_pos,
						 struct bpos new_pos)
{
	if (!btree_type_has_snapshots(btree) ||
	    bkey_eq(old_pos, new_pos))
		return 0;

	return __bch2_insert_snapshot_whiteouts(trans, btree, old_pos, new_pos);
}

int bch2_trans_update_extent_overwrite(struct btree_trans *, struct btree_iter *,
				       enum btree_iter_update_trigger_flags,
				       struct bkey_s_c, struct bkey_s_c);

int bch2_bkey_get_empty_slot(struct btree_trans *, struct btree_iter *,
			     enum btree_id, struct bpos);

int __must_check bch2_trans_update(struct btree_trans *, struct btree_iter *,
				   struct bkey_i *, enum btree_iter_update_trigger_flags);

struct jset_entry *__bch2_trans_jset_entry_alloc(struct btree_trans *, unsigned);

static inline struct jset_entry *btree_trans_journal_entries_top(struct btree_trans *trans)
{
	return (void *) ((u64 *) trans->journal_entries + trans->journal_entries_u64s);
}

static inline struct jset_entry *
bch2_trans_jset_entry_alloc(struct btree_trans *trans, unsigned u64s)
{
	if (!trans->journal_entries ||
	    trans->journal_entries_u64s + u64s > trans->journal_entries_size)
		return __bch2_trans_jset_entry_alloc(trans, u64s);

	struct jset_entry *e = btree_trans_journal_entries_top(trans);
	trans->journal_entries_u64s += u64s;
	return e;
}

int bch2_btree_insert_clone_trans(struct btree_trans *, enum btree_id, struct bkey_i *);

static inline int __must_check bch2_trans_update_buffered(struct btree_trans *trans,
					    enum btree_id btree,
					    struct bkey_i *k)
{
	/*
	 * Most updates skip the btree write buffer until journal replay is
	 * finished because synchronization with journal replay relies on having
	 * a btree node locked - if we're overwriting a key in the journal that
	 * journal replay hasn't yet replayed, we have to mark it as
	 * overwritten.
	 *
	 * But accounting updates don't overwrite, they're deltas, and they have
	 * to be flushed to the btree strictly in order for journal replay to be
	 * able to tell which updates need to be applied:
	 */
	if (k->k.type != KEY_TYPE_accounting &&
	    unlikely(trans->journal_replay_not_finished))
		return bch2_btree_insert_clone_trans(trans, btree, k);

	struct jset_entry *e = bch2_trans_jset_entry_alloc(trans, jset_u64s(k->k.u64s));
	int ret = PTR_ERR_OR_ZERO(e);
	if (ret)
		return ret;

	journal_entry_init(e, BCH_JSET_ENTRY_write_buffer_keys, btree, 0, k->k.u64s);
	bkey_copy(e->start, k);
	return 0;
}

void bch2_trans_commit_hook(struct btree_trans *,
			    struct btree_trans_commit_hook *);
int __bch2_trans_commit(struct btree_trans *, unsigned);

__printf(2, 3) int bch2_fs_log_msg(struct bch_fs *, const char *, ...);
__printf(2, 3) int bch2_journal_log_msg(struct bch_fs *, const char *, ...);

/**
 * bch2_trans_commit - insert keys at given iterator positions
 *
 * This is main entry point for btree updates.
 *
 * Return values:
 * -EROFS: filesystem read only
 * -EIO: journal or btree node IO error
 */
static inline int bch2_trans_commit(struct btree_trans *trans,
				    struct disk_reservation *disk_res,
				    u64 *journal_seq,
				    unsigned flags)
{
	trans->disk_res		= disk_res;
	trans->journal_seq	= journal_seq;

	return __bch2_trans_commit(trans, flags);
}

#define commit_do(_trans, _disk_res, _journal_seq, _flags, _do)	\
	lockrestart_do(_trans, _do ?: bch2_trans_commit(_trans, (_disk_res),\
					(_journal_seq), (_flags)))

#define nested_commit_do(_trans, _disk_res, _journal_seq, _flags, _do)	\
	nested_lockrestart_do(_trans, _do ?: bch2_trans_commit(_trans, (_disk_res),\
					(_journal_seq), (_flags)))

#define bch2_trans_do(_c, _disk_res, _journal_seq, _flags, _do)		\
	bch2_trans_run(_c, commit_do(trans, _disk_res, _journal_seq, _flags, _do))

#define trans_for_each_update(_trans, _i)				\
	for (struct btree_insert_entry *_i = (_trans)->updates;		\
	     (_i) < (_trans)->updates + (_trans)->nr_updates;		\
	     (_i)++)

static inline void bch2_trans_reset_updates(struct btree_trans *trans)
{
	trans_for_each_update(trans, i)
		bch2_path_put(trans, i->path, true);

	trans->nr_updates		= 0;
	trans->journal_entries_u64s	= 0;
	trans->hooks			= NULL;
	trans->extra_disk_res		= 0;
}

static inline struct bkey_i *__bch2_bkey_make_mut_noupdate(struct btree_trans *trans, struct bkey_s_c k,
						  unsigned type, unsigned min_bytes)
{
	unsigned bytes = max_t(unsigned, min_bytes, bkey_bytes(k.k));
	struct bkey_i *mut;

	if (type && k.k->type != type)
		return ERR_PTR(-ENOENT);

	mut = bch2_trans_kmalloc_nomemzero(trans, bytes);
	if (!IS_ERR(mut)) {
		bkey_reassemble(mut, k);

		if (unlikely(bytes > bkey_bytes(k.k))) {
			memset((void *) mut + bkey_bytes(k.k), 0,
			       bytes - bkey_bytes(k.k));
			mut->k.u64s = DIV_ROUND_UP(bytes, sizeof(u64));
		}
	}
	return mut;
}

static inline struct bkey_i *bch2_bkey_make_mut_noupdate(struct btree_trans *trans, struct bkey_s_c k)
{
	return __bch2_bkey_make_mut_noupdate(trans, k, 0, 0);
}

#define bch2_bkey_make_mut_noupdate_typed(_trans, _k, _type)		\
	bkey_i_to_##_type(__bch2_bkey_make_mut_noupdate(_trans, _k,	\
				KEY_TYPE_##_type, sizeof(struct bkey_i_##_type)))

static inline struct bkey_i *__bch2_bkey_make_mut(struct btree_trans *trans, struct btree_iter *iter,
					struct bkey_s_c *k, unsigned flags,
					unsigned type, unsigned min_bytes)
{
	struct bkey_i *mut = __bch2_bkey_make_mut_noupdate(trans, *k, type, min_bytes);
	int ret;

	if (IS_ERR(mut))
		return mut;

	ret = bch2_trans_update(trans, iter, mut, flags);
	if (ret)
		return ERR_PTR(ret);

	*k = bkey_i_to_s_c(mut);
	return mut;
}

static inline struct bkey_i *bch2_bkey_make_mut(struct btree_trans *trans, struct btree_iter *iter,
						struct bkey_s_c *k, unsigned flags)
{
	return __bch2_bkey_make_mut(trans, iter, k, flags, 0, 0);
}

#define bch2_bkey_make_mut_typed(_trans, _iter, _k, _flags, _type)	\
	bkey_i_to_##_type(__bch2_bkey_make_mut(_trans, _iter, _k, _flags,\
				KEY_TYPE_##_type, sizeof(struct bkey_i_##_type)))

static inline struct bkey_i *__bch2_bkey_get_mut_noupdate(struct btree_trans *trans,
					 struct btree_iter *iter,
					 unsigned btree_id, struct bpos pos,
					 unsigned flags, unsigned type, unsigned min_bytes)
{
	struct bkey_s_c k = __bch2_bkey_get_iter(trans, iter,
				btree_id, pos, flags|BTREE_ITER_intent, type);
	struct bkey_i *ret = IS_ERR(k.k)
		? ERR_CAST(k.k)
		: __bch2_bkey_make_mut_noupdate(trans, k, 0, min_bytes);
	if (IS_ERR(ret))
		bch2_trans_iter_exit(trans, iter);
	return ret;
}

static inline struct bkey_i *bch2_bkey_get_mut_noupdate(struct btree_trans *trans,
					       struct btree_iter *iter,
					       unsigned btree_id, struct bpos pos,
					       unsigned flags)
{
	return __bch2_bkey_get_mut_noupdate(trans, iter, btree_id, pos, flags, 0, 0);
}

static inline struct bkey_i *__bch2_bkey_get_mut(struct btree_trans *trans,
					 struct btree_iter *iter,
					 unsigned btree_id, struct bpos pos,
					 unsigned flags, unsigned type, unsigned min_bytes)
{
	struct bkey_i *mut = __bch2_bkey_get_mut_noupdate(trans, iter,
				btree_id, pos, flags|BTREE_ITER_intent, type, min_bytes);
	int ret;

	if (IS_ERR(mut))
		return mut;

	ret = bch2_trans_update(trans, iter, mut, flags);
	if (ret) {
		bch2_trans_iter_exit(trans, iter);
		return ERR_PTR(ret);
	}

	return mut;
}

static inline struct bkey_i *bch2_bkey_get_mut_minsize(struct btree_trans *trans,
						       struct btree_iter *iter,
						       unsigned btree_id, struct bpos pos,
						       unsigned flags, unsigned min_bytes)
{
	return __bch2_bkey_get_mut(trans, iter, btree_id, pos, flags, 0, min_bytes);
}

static inline struct bkey_i *bch2_bkey_get_mut(struct btree_trans *trans,
					       struct btree_iter *iter,
					       unsigned btree_id, struct bpos pos,
					       unsigned flags)
{
	return __bch2_bkey_get_mut(trans, iter, btree_id, pos, flags, 0, 0);
}

#define bch2_bkey_get_mut_typed(_trans, _iter, _btree_id, _pos, _flags, _type)\
	bkey_i_to_##_type(__bch2_bkey_get_mut(_trans, _iter,		\
			_btree_id, _pos, _flags,			\
			KEY_TYPE_##_type, sizeof(struct bkey_i_##_type)))

static inline struct bkey_i *__bch2_bkey_alloc(struct btree_trans *trans, struct btree_iter *iter,
					       unsigned flags, unsigned type, unsigned val_size)
{
	struct bkey_i *k = bch2_trans_kmalloc(trans, sizeof(*k) + val_size);
	int ret;

	if (IS_ERR(k))
		return k;

	bkey_init(&k->k);
	k->k.p = iter->pos;
	k->k.type = type;
	set_bkey_val_bytes(&k->k, val_size);

	ret = bch2_trans_update(trans, iter, k, flags);
	if (unlikely(ret))
		return ERR_PTR(ret);
	return k;
}

#define bch2_bkey_alloc(_trans, _iter, _flags, _type)			\
	bkey_i_to_##_type(__bch2_bkey_alloc(_trans, _iter, _flags,	\
				KEY_TYPE_##_type, sizeof(struct bch_##_type)))

#endif /* _BCACHEFS_BTREE_UPDATE_H */