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
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
|
//SPDX-License-Identifier: GPL-2.0
#include <linux/bpf-cgroup.h>
#include <linux/bpf.h>
#include <linux/btf.h>
#include <linux/bug.h>
#include <linux/filter.h>
#include <linux/mm.h>
#include <linux/rbtree.h>
#include <linux/slab.h>
#include <uapi/linux/btf.h>
DEFINE_PER_CPU(struct bpf_cgroup_storage*, bpf_cgroup_storage[MAX_BPF_CGROUP_STORAGE_TYPE]);
#ifdef CONFIG_CGROUP_BPF
#define LOCAL_STORAGE_CREATE_FLAG_MASK \
(BPF_F_NUMA_NODE | BPF_F_RDONLY | BPF_F_WRONLY)
struct bpf_cgroup_storage_map {
struct bpf_map map;
spinlock_t lock;
struct bpf_prog *prog;
struct rb_root root;
struct list_head list;
};
static struct bpf_cgroup_storage_map *map_to_storage(struct bpf_map *map)
{
return container_of(map, struct bpf_cgroup_storage_map, map);
}
static int bpf_cgroup_storage_key_cmp(
const struct bpf_cgroup_storage_key *key1,
const struct bpf_cgroup_storage_key *key2)
{
if (key1->cgroup_inode_id < key2->cgroup_inode_id)
return -1;
else if (key1->cgroup_inode_id > key2->cgroup_inode_id)
return 1;
else if (key1->attach_type < key2->attach_type)
return -1;
else if (key1->attach_type > key2->attach_type)
return 1;
return 0;
}
static struct bpf_cgroup_storage *cgroup_storage_lookup(
struct bpf_cgroup_storage_map *map, struct bpf_cgroup_storage_key *key,
bool locked)
{
struct rb_root *root = &map->root;
struct rb_node *node;
if (!locked)
spin_lock_bh(&map->lock);
node = root->rb_node;
while (node) {
struct bpf_cgroup_storage *storage;
storage = container_of(node, struct bpf_cgroup_storage, node);
switch (bpf_cgroup_storage_key_cmp(key, &storage->key)) {
case -1:
node = node->rb_left;
break;
case 1:
node = node->rb_right;
break;
default:
if (!locked)
spin_unlock_bh(&map->lock);
return storage;
}
}
if (!locked)
spin_unlock_bh(&map->lock);
return NULL;
}
static int cgroup_storage_insert(struct bpf_cgroup_storage_map *map,
struct bpf_cgroup_storage *storage)
{
struct rb_root *root = &map->root;
struct rb_node **new = &(root->rb_node), *parent = NULL;
while (*new) {
struct bpf_cgroup_storage *this;
this = container_of(*new, struct bpf_cgroup_storage, node);
parent = *new;
switch (bpf_cgroup_storage_key_cmp(&storage->key, &this->key)) {
case -1:
new = &((*new)->rb_left);
break;
case 1:
new = &((*new)->rb_right);
break;
default:
return -EEXIST;
}
}
rb_link_node(&storage->node, parent, new);
rb_insert_color(&storage->node, root);
return 0;
}
static void *cgroup_storage_lookup_elem(struct bpf_map *_map, void *_key)
{
struct bpf_cgroup_storage_map *map = map_to_storage(_map);
struct bpf_cgroup_storage_key *key = _key;
struct bpf_cgroup_storage *storage;
storage = cgroup_storage_lookup(map, key, false);
if (!storage)
return NULL;
return &READ_ONCE(storage->buf)->data[0];
}
static int cgroup_storage_update_elem(struct bpf_map *map, void *_key,
void *value, u64 flags)
{
struct bpf_cgroup_storage_key *key = _key;
struct bpf_cgroup_storage *storage;
struct bpf_storage_buffer *new;
if (flags != BPF_ANY && flags != BPF_EXIST)
return -EINVAL;
storage = cgroup_storage_lookup((struct bpf_cgroup_storage_map *)map,
key, false);
if (!storage)
return -ENOENT;
new = kmalloc_node(sizeof(struct bpf_storage_buffer) +
map->value_size,
__GFP_ZERO | GFP_ATOMIC | __GFP_NOWARN,
map->numa_node);
if (!new)
return -ENOMEM;
memcpy(&new->data[0], value, map->value_size);
new = xchg(&storage->buf, new);
kfree_rcu(new, rcu);
return 0;
}
int bpf_percpu_cgroup_storage_copy(struct bpf_map *_map, void *_key,
void *value)
{
struct bpf_cgroup_storage_map *map = map_to_storage(_map);
struct bpf_cgroup_storage_key *key = _key;
struct bpf_cgroup_storage *storage;
int cpu, off = 0;
u32 size;
rcu_read_lock();
storage = cgroup_storage_lookup(map, key, false);
if (!storage) {
rcu_read_unlock();
return -ENOENT;
}
/* per_cpu areas are zero-filled and bpf programs can only
* access 'value_size' of them, so copying rounded areas
* will not leak any kernel data
*/
size = round_up(_map->value_size, 8);
for_each_possible_cpu(cpu) {
bpf_long_memcpy(value + off,
per_cpu_ptr(storage->percpu_buf, cpu), size);
off += size;
}
rcu_read_unlock();
return 0;
}
int bpf_percpu_cgroup_storage_update(struct bpf_map *_map, void *_key,
void *value, u64 map_flags)
{
struct bpf_cgroup_storage_map *map = map_to_storage(_map);
struct bpf_cgroup_storage_key *key = _key;
struct bpf_cgroup_storage *storage;
int cpu, off = 0;
u32 size;
if (map_flags != BPF_ANY && map_flags != BPF_EXIST)
return -EINVAL;
rcu_read_lock();
storage = cgroup_storage_lookup(map, key, false);
if (!storage) {
rcu_read_unlock();
return -ENOENT;
}
/* the user space will provide round_up(value_size, 8) bytes that
* will be copied into per-cpu area. bpf programs can only access
* value_size of it. During lookup the same extra bytes will be
* returned or zeros which were zero-filled by percpu_alloc,
* so no kernel data leaks possible
*/
size = round_up(_map->value_size, 8);
for_each_possible_cpu(cpu) {
bpf_long_memcpy(per_cpu_ptr(storage->percpu_buf, cpu),
value + off, size);
off += size;
}
rcu_read_unlock();
return 0;
}
static int cgroup_storage_get_next_key(struct bpf_map *_map, void *_key,
void *_next_key)
{
struct bpf_cgroup_storage_map *map = map_to_storage(_map);
struct bpf_cgroup_storage_key *key = _key;
struct bpf_cgroup_storage_key *next = _next_key;
struct bpf_cgroup_storage *storage;
spin_lock_bh(&map->lock);
if (list_empty(&map->list))
goto enoent;
if (key) {
storage = cgroup_storage_lookup(map, key, true);
if (!storage)
goto enoent;
storage = list_next_entry(storage, list);
if (!storage)
goto enoent;
} else {
storage = list_first_entry(&map->list,
struct bpf_cgroup_storage, list);
}
spin_unlock_bh(&map->lock);
next->attach_type = storage->key.attach_type;
next->cgroup_inode_id = storage->key.cgroup_inode_id;
return 0;
enoent:
spin_unlock_bh(&map->lock);
return -ENOENT;
}
static struct bpf_map *cgroup_storage_map_alloc(union bpf_attr *attr)
{
int numa_node = bpf_map_attr_numa_node(attr);
struct bpf_cgroup_storage_map *map;
if (attr->key_size != sizeof(struct bpf_cgroup_storage_key))
return ERR_PTR(-EINVAL);
if (attr->value_size == 0)
return ERR_PTR(-EINVAL);
if (attr->value_size > PAGE_SIZE)
return ERR_PTR(-E2BIG);
if (attr->map_flags & ~LOCAL_STORAGE_CREATE_FLAG_MASK)
/* reserved bits should not be used */
return ERR_PTR(-EINVAL);
if (attr->max_entries)
/* max_entries is not used and enforced to be 0 */
return ERR_PTR(-EINVAL);
map = kmalloc_node(sizeof(struct bpf_cgroup_storage_map),
__GFP_ZERO | GFP_USER, numa_node);
if (!map)
return ERR_PTR(-ENOMEM);
map->map.pages = round_up(sizeof(struct bpf_cgroup_storage_map),
PAGE_SIZE) >> PAGE_SHIFT;
/* copy mandatory map attributes */
bpf_map_init_from_attr(&map->map, attr);
spin_lock_init(&map->lock);
map->root = RB_ROOT;
INIT_LIST_HEAD(&map->list);
return &map->map;
}
static void cgroup_storage_map_free(struct bpf_map *_map)
{
struct bpf_cgroup_storage_map *map = map_to_storage(_map);
WARN_ON(!RB_EMPTY_ROOT(&map->root));
WARN_ON(!list_empty(&map->list));
kfree(map);
}
static int cgroup_storage_delete_elem(struct bpf_map *map, void *key)
{
return -EINVAL;
}
static int cgroup_storage_check_btf(const struct bpf_map *map,
const struct btf *btf,
const struct btf_type *key_type,
const struct btf_type *value_type)
{
const struct btf_type *t;
struct btf_member *m;
u32 id, size;
/* Key is expected to be of struct bpf_cgroup_storage_key type,
* which is:
* struct bpf_cgroup_storage_key {
* __u64 cgroup_inode_id;
* __u32 attach_type;
* };
*/
/*
* Key_type must be a structure with two fields.
*/
if (BTF_INFO_KIND(key_type->info) != BTF_KIND_STRUCT ||
BTF_INFO_VLEN(key_type->info) != 2)
return -EINVAL;
/*
* The first field must be a 64 bit integer at 0 offset.
*/
m = (struct btf_member *)(key_type + 1);
if (m->offset)
return -EINVAL;
id = m->type;
t = btf_type_id_size(btf, &id, NULL);
size = FIELD_SIZEOF(struct bpf_cgroup_storage_key, cgroup_inode_id);
if (!t || !btf_type_is_reg_int(t, size))
return -EINVAL;
/*
* The second field must be a 32 bit integer at 64 bit offset.
*/
m++;
if (m->offset != offsetof(struct bpf_cgroup_storage_key, attach_type) *
BITS_PER_BYTE)
return -EINVAL;
id = m->type;
t = btf_type_id_size(btf, &id, NULL);
size = FIELD_SIZEOF(struct bpf_cgroup_storage_key, attach_type);
if (!t || !btf_type_is_reg_int(t, size))
return -EINVAL;
return 0;
}
static void cgroup_storage_seq_show_elem(struct bpf_map *map, void *_key,
struct seq_file *m)
{
enum bpf_cgroup_storage_type stype = cgroup_storage_type(map);
struct bpf_cgroup_storage_key *key = _key;
struct bpf_cgroup_storage *storage;
int cpu;
rcu_read_lock();
storage = cgroup_storage_lookup(map_to_storage(map), key, false);
if (!storage) {
rcu_read_unlock();
return;
}
btf_type_seq_show(map->btf, map->btf_key_type_id, key, m);
stype = cgroup_storage_type(map);
if (stype == BPF_CGROUP_STORAGE_SHARED) {
seq_puts(m, ": ");
btf_type_seq_show(map->btf, map->btf_value_type_id,
&READ_ONCE(storage->buf)->data[0], m);
seq_puts(m, "\n");
} else {
seq_puts(m, ": {\n");
for_each_possible_cpu(cpu) {
seq_printf(m, "\tcpu%d: ", cpu);
btf_type_seq_show(map->btf, map->btf_value_type_id,
per_cpu_ptr(storage->percpu_buf, cpu),
m);
seq_puts(m, "\n");
}
seq_puts(m, "}\n");
}
rcu_read_unlock();
}
const struct bpf_map_ops cgroup_storage_map_ops = {
.map_alloc = cgroup_storage_map_alloc,
.map_free = cgroup_storage_map_free,
.map_get_next_key = cgroup_storage_get_next_key,
.map_lookup_elem = cgroup_storage_lookup_elem,
.map_update_elem = cgroup_storage_update_elem,
.map_delete_elem = cgroup_storage_delete_elem,
.map_check_btf = cgroup_storage_check_btf,
.map_seq_show_elem = cgroup_storage_seq_show_elem,
};
int bpf_cgroup_storage_assign(struct bpf_prog *prog, struct bpf_map *_map)
{
enum bpf_cgroup_storage_type stype = cgroup_storage_type(_map);
struct bpf_cgroup_storage_map *map = map_to_storage(_map);
int ret = -EBUSY;
spin_lock_bh(&map->lock);
if (map->prog && map->prog != prog)
goto unlock;
if (prog->aux->cgroup_storage[stype] &&
prog->aux->cgroup_storage[stype] != _map)
goto unlock;
map->prog = prog;
prog->aux->cgroup_storage[stype] = _map;
ret = 0;
unlock:
spin_unlock_bh(&map->lock);
return ret;
}
void bpf_cgroup_storage_release(struct bpf_prog *prog, struct bpf_map *_map)
{
enum bpf_cgroup_storage_type stype = cgroup_storage_type(_map);
struct bpf_cgroup_storage_map *map = map_to_storage(_map);
spin_lock_bh(&map->lock);
if (map->prog == prog) {
WARN_ON(prog->aux->cgroup_storage[stype] != _map);
map->prog = NULL;
prog->aux->cgroup_storage[stype] = NULL;
}
spin_unlock_bh(&map->lock);
}
static size_t bpf_cgroup_storage_calculate_size(struct bpf_map *map, u32 *pages)
{
size_t size;
if (cgroup_storage_type(map) == BPF_CGROUP_STORAGE_SHARED) {
size = sizeof(struct bpf_storage_buffer) + map->value_size;
*pages = round_up(sizeof(struct bpf_cgroup_storage) + size,
PAGE_SIZE) >> PAGE_SHIFT;
} else {
size = map->value_size;
*pages = round_up(round_up(size, 8) * num_possible_cpus(),
PAGE_SIZE) >> PAGE_SHIFT;
}
return size;
}
struct bpf_cgroup_storage *bpf_cgroup_storage_alloc(struct bpf_prog *prog,
enum bpf_cgroup_storage_type stype)
{
struct bpf_cgroup_storage *storage;
struct bpf_map *map;
gfp_t flags;
size_t size;
u32 pages;
map = prog->aux->cgroup_storage[stype];
if (!map)
return NULL;
size = bpf_cgroup_storage_calculate_size(map, &pages);
if (bpf_map_charge_memlock(map, pages))
return ERR_PTR(-EPERM);
storage = kmalloc_node(sizeof(struct bpf_cgroup_storage),
__GFP_ZERO | GFP_USER, map->numa_node);
if (!storage)
goto enomem;
flags = __GFP_ZERO | GFP_USER;
if (stype == BPF_CGROUP_STORAGE_SHARED) {
storage->buf = kmalloc_node(size, flags, map->numa_node);
if (!storage->buf)
goto enomem;
} else {
storage->percpu_buf = __alloc_percpu_gfp(size, 8, flags);
if (!storage->percpu_buf)
goto enomem;
}
storage->map = (struct bpf_cgroup_storage_map *)map;
return storage;
enomem:
bpf_map_uncharge_memlock(map, pages);
kfree(storage);
return ERR_PTR(-ENOMEM);
}
static void free_shared_cgroup_storage_rcu(struct rcu_head *rcu)
{
struct bpf_cgroup_storage *storage =
container_of(rcu, struct bpf_cgroup_storage, rcu);
kfree(storage->buf);
kfree(storage);
}
static void free_percpu_cgroup_storage_rcu(struct rcu_head *rcu)
{
struct bpf_cgroup_storage *storage =
container_of(rcu, struct bpf_cgroup_storage, rcu);
free_percpu(storage->percpu_buf);
kfree(storage);
}
void bpf_cgroup_storage_free(struct bpf_cgroup_storage *storage)
{
enum bpf_cgroup_storage_type stype;
struct bpf_map *map;
u32 pages;
if (!storage)
return;
map = &storage->map->map;
bpf_cgroup_storage_calculate_size(map, &pages);
bpf_map_uncharge_memlock(map, pages);
stype = cgroup_storage_type(map);
if (stype == BPF_CGROUP_STORAGE_SHARED)
call_rcu(&storage->rcu, free_shared_cgroup_storage_rcu);
else
call_rcu(&storage->rcu, free_percpu_cgroup_storage_rcu);
}
void bpf_cgroup_storage_link(struct bpf_cgroup_storage *storage,
struct cgroup *cgroup,
enum bpf_attach_type type)
{
struct bpf_cgroup_storage_map *map;
if (!storage)
return;
storage->key.attach_type = type;
storage->key.cgroup_inode_id = cgroup->kn->id.id;
map = storage->map;
spin_lock_bh(&map->lock);
WARN_ON(cgroup_storage_insert(map, storage));
list_add(&storage->list, &map->list);
spin_unlock_bh(&map->lock);
}
void bpf_cgroup_storage_unlink(struct bpf_cgroup_storage *storage)
{
struct bpf_cgroup_storage_map *map;
struct rb_root *root;
if (!storage)
return;
map = storage->map;
spin_lock_bh(&map->lock);
root = &map->root;
rb_erase(&storage->node, root);
list_del(&storage->list);
spin_unlock_bh(&map->lock);
}
#endif
|