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
|
/* SPDX-License-Identifier: GPL-2.0 */
/*
* Code for manipulating bucket marks for garbage collection.
*
* Copyright 2014 Datera, Inc.
*/
#ifndef _BUCKETS_H
#define _BUCKETS_H
#include "buckets_types.h"
#include "super.h"
#define for_each_bucket(_b, _buckets) \
for (_b = (_buckets)->b + (_buckets)->first_bucket; \
_b < (_buckets)->b + (_buckets)->nbuckets; _b++)
#define bucket_cmpxchg(g, new, expr) \
({ \
struct bucket *_g = g; \
u64 _v = atomic64_read(&(g)->_mark.v); \
struct bucket_mark _old; \
\
do { \
(new).v.counter = _old.v.counter = _v; \
expr; \
} while ((_v = atomic64_cmpxchg(&(_g)->_mark.v, \
_old.v.counter, \
(new).v.counter)) != _old.v.counter);\
_old; \
})
static inline struct bucket_array *__bucket_array(struct bch_dev *ca,
bool gc)
{
return rcu_dereference_check(ca->buckets[gc],
!ca->fs ||
percpu_rwsem_is_held(&ca->fs->mark_lock) ||
lockdep_is_held(&ca->fs->gc_lock) ||
lockdep_is_held(&ca->bucket_lock));
}
static inline struct bucket_array *bucket_array(struct bch_dev *ca)
{
return __bucket_array(ca, false);
}
static inline struct bucket *__bucket(struct bch_dev *ca, size_t b, bool gc)
{
struct bucket_array *buckets = __bucket_array(ca, gc);
BUG_ON(b < buckets->first_bucket || b >= buckets->nbuckets);
return buckets->b + b;
}
static inline struct bucket *gc_bucket(struct bch_dev *ca, size_t b)
{
return __bucket(ca, b, true);
}
static inline struct bucket *bucket(struct bch_dev *ca, size_t b)
{
return __bucket(ca, b, false);
}
static inline struct bucket_gens *bucket_gens(struct bch_dev *ca)
{
return rcu_dereference_check(ca->bucket_gens,
!ca->fs ||
percpu_rwsem_is_held(&ca->fs->mark_lock) ||
lockdep_is_held(&ca->fs->gc_lock) ||
lockdep_is_held(&ca->bucket_lock));
}
static inline u8 *bucket_gen(struct bch_dev *ca, size_t b)
{
struct bucket_gens *gens = bucket_gens(ca);
BUG_ON(b < gens->first_bucket || b >= gens->nbuckets);
return gens->b + b;
}
/*
* bucket_gc_gen() returns the difference between the bucket's current gen and
* the oldest gen of any pointer into that bucket in the btree.
*/
static inline u8 bucket_gc_gen(struct bucket *g)
{
return g->mark.gen - g->oldest_gen;
}
static inline size_t PTR_BUCKET_NR(const struct bch_dev *ca,
const struct bch_extent_ptr *ptr)
{
return sector_to_bucket(ca, ptr->offset);
}
static inline struct bucket *PTR_GC_BUCKET(struct bch_dev *ca,
const struct bch_extent_ptr *ptr)
{
return gc_bucket(ca, PTR_BUCKET_NR(ca, ptr));
}
static inline enum bch_data_type ptr_data_type(const struct bkey *k,
const struct bch_extent_ptr *ptr)
{
if (k->type == KEY_TYPE_btree_ptr ||
k->type == KEY_TYPE_btree_ptr_v2)
return BCH_DATA_btree;
return ptr->cached ? BCH_DATA_cached : BCH_DATA_user;
}
static inline int gen_cmp(u8 a, u8 b)
{
return (s8) (a - b);
}
static inline int gen_after(u8 a, u8 b)
{
int r = gen_cmp(a, b);
return r > 0 ? r : 0;
}
/**
* ptr_stale() - check if a pointer points into a bucket that has been
* invalidated.
*/
static inline u8 ptr_stale(struct bch_dev *ca,
const struct bch_extent_ptr *ptr)
{
u8 ret;
rcu_read_lock();
ret = gen_after(*bucket_gen(ca, PTR_BUCKET_NR(ca, ptr)), ptr->gen);
rcu_read_unlock();
return ret;
}
/* bucket gc marks */
static inline bool is_available_bucket(struct bucket_mark mark)
{
return !mark.dirty_sectors && !mark.stripe;
}
/* Device usage: */
struct bch_dev_usage bch2_dev_usage_read(struct bch_dev *);
static inline u64 __dev_buckets_available(struct bch_dev *ca,
struct bch_dev_usage stats)
{
u64 total = ca->mi.nbuckets - ca->mi.first_bucket;
if (WARN_ONCE(stats.buckets_unavailable > total,
"buckets_unavailable overflow (%llu > %llu)\n",
stats.buckets_unavailable, total))
return 0;
return total - stats.buckets_unavailable;
}
static inline u64 dev_buckets_available(struct bch_dev *ca)
{
return __dev_buckets_available(ca, bch2_dev_usage_read(ca));
}
static inline u64 __dev_buckets_reclaimable(struct bch_dev *ca,
struct bch_dev_usage stats)
{
struct bch_fs *c = ca->fs;
s64 available = __dev_buckets_available(ca, stats);
unsigned i;
spin_lock(&c->freelist_lock);
for (i = 0; i < RESERVE_NR; i++)
available -= fifo_used(&ca->free[i]);
available -= fifo_used(&ca->free_inc);
available -= ca->nr_open_buckets;
spin_unlock(&c->freelist_lock);
return max(available, 0LL);
}
static inline u64 dev_buckets_reclaimable(struct bch_dev *ca)
{
return __dev_buckets_reclaimable(ca, bch2_dev_usage_read(ca));
}
/* Filesystem usage: */
static inline unsigned fs_usage_u64s(struct bch_fs *c)
{
return sizeof(struct bch_fs_usage) / sizeof(u64) +
READ_ONCE(c->replicas.nr);
}
static inline unsigned dev_usage_u64s(void)
{
return sizeof(struct bch_dev_usage) / sizeof(u64);
}
u64 bch2_fs_usage_read_one(struct bch_fs *, u64 *);
struct bch_fs_usage_online *bch2_fs_usage_read(struct bch_fs *);
void bch2_fs_usage_acc_to_base(struct bch_fs *, unsigned);
void bch2_fs_usage_to_text(struct printbuf *,
struct bch_fs *, struct bch_fs_usage_online *);
u64 bch2_fs_sectors_used(struct bch_fs *, struct bch_fs_usage_online *);
struct bch_fs_usage_short
bch2_fs_usage_read_short(struct bch_fs *);
/* key/bucket marking: */
void bch2_fs_usage_initialize(struct bch_fs *);
void bch2_mark_alloc_bucket(struct bch_fs *, struct bch_dev *, size_t, bool);
void bch2_mark_metadata_bucket(struct bch_fs *, struct bch_dev *,
size_t, enum bch_data_type, unsigned,
struct gc_pos, unsigned);
int bch2_mark_alloc(struct btree_trans *, struct bkey_s_c, struct bkey_s_c, unsigned);
int bch2_mark_extent(struct btree_trans *, struct bkey_s_c, struct bkey_s_c, unsigned);
int bch2_mark_stripe(struct btree_trans *, struct bkey_s_c, struct bkey_s_c, unsigned);
int bch2_mark_inode(struct btree_trans *, struct bkey_s_c, struct bkey_s_c, unsigned);
int bch2_mark_reservation(struct btree_trans *, struct bkey_s_c, struct bkey_s_c, unsigned);
int bch2_mark_reflink_p(struct btree_trans *, struct bkey_s_c, struct bkey_s_c, unsigned);
int bch2_trans_mark_extent(struct btree_trans *, struct bkey_s_c, struct bkey_i *, unsigned);
int bch2_trans_mark_stripe(struct btree_trans *, struct bkey_s_c, struct bkey_i *, unsigned);
int bch2_trans_mark_inode(struct btree_trans *, struct bkey_s_c, struct bkey_i *, unsigned);
int bch2_trans_mark_reservation(struct btree_trans *, struct bkey_s_c, struct bkey_i *, unsigned);
int bch2_trans_mark_reflink_p(struct btree_trans *, struct bkey_s_c, struct bkey_i *, unsigned);
int bch2_mark_key(struct btree_trans *, struct bkey_s_c, struct bkey_s_c, unsigned);
int bch2_trans_mark_key(struct btree_trans *, struct bkey_s_c,
struct bkey_i *, unsigned);
static inline int bch2_trans_mark_old(struct btree_trans *trans,
struct bkey_s_c old, unsigned flags)
{
struct bkey_i deleted;
bkey_init(&deleted.k);
deleted.k.p = old.k->p;
return bch2_trans_mark_key(trans, old, &deleted,
BTREE_TRIGGER_OVERWRITE|flags);
}
static inline int bch2_trans_mark_new(struct btree_trans *trans,
struct bkey_i *new, unsigned flags)
{
struct bkey_i deleted;
bkey_init(&deleted.k);
deleted.k.p = new->k.p;
return bch2_trans_mark_key(trans, bkey_i_to_s_c(&deleted), new,
BTREE_TRIGGER_INSERT|flags);
}
int bch2_trans_fs_usage_apply(struct btree_trans *, struct replicas_delta_list *);
int bch2_trans_mark_metadata_bucket(struct btree_trans *, struct bch_dev *,
size_t, enum bch_data_type, unsigned);
int bch2_trans_mark_dev_sb(struct bch_fs *, struct bch_dev *);
/* disk reservations: */
static inline void bch2_disk_reservation_put(struct bch_fs *c,
struct disk_reservation *res)
{
if (res->sectors) {
this_cpu_sub(*c->online_reserved, res->sectors);
res->sectors = 0;
}
}
#define BCH_DISK_RESERVATION_NOFAIL (1 << 0)
int __bch2_disk_reservation_add(struct bch_fs *,
struct disk_reservation *,
u64, int);
static inline int bch2_disk_reservation_add(struct bch_fs *c, struct disk_reservation *res,
u64 sectors, int flags)
{
u64 old, new;
do {
old = this_cpu_read(c->pcpu->sectors_available);
if (sectors > old)
return __bch2_disk_reservation_add(c, res, sectors, flags);
new = old - sectors;
} while (this_cpu_cmpxchg(c->pcpu->sectors_available, old, new) != old);
this_cpu_add(*c->online_reserved, sectors);
res->sectors += sectors;
return 0;
}
static inline struct disk_reservation
bch2_disk_reservation_init(struct bch_fs *c, unsigned nr_replicas)
{
return (struct disk_reservation) {
.sectors = 0,
#if 0
/* not used yet: */
.gen = c->capacity_gen,
#endif
.nr_replicas = nr_replicas,
};
}
static inline int bch2_disk_reservation_get(struct bch_fs *c,
struct disk_reservation *res,
u64 sectors, unsigned nr_replicas,
int flags)
{
*res = bch2_disk_reservation_init(c, nr_replicas);
return bch2_disk_reservation_add(c, res, sectors * nr_replicas, flags);
}
#define RESERVE_FACTOR 6
static inline u64 avail_factor(u64 r)
{
return div_u64(r << RESERVE_FACTOR, (1 << RESERVE_FACTOR) + 1);
}
int bch2_dev_buckets_resize(struct bch_fs *, struct bch_dev *, u64);
void bch2_dev_buckets_free(struct bch_dev *);
int bch2_dev_buckets_alloc(struct bch_fs *, struct bch_dev *);
#endif /* _BUCKETS_H */
|