osdmap.c 31.1 KB
Newer Older
S
Sage Weil 已提交
1

2
#include <linux/ceph/ceph_debug.h>
3

4
#include <linux/module.h>
5
#include <linux/slab.h>
S
Sage Weil 已提交
6 7
#include <asm/div64.h>

8 9 10 11 12
#include <linux/ceph/libceph.h>
#include <linux/ceph/osdmap.h>
#include <linux/ceph/decode.h>
#include <linux/crush/hash.h>
#include <linux/crush/mapper.h>
S
Sage Weil 已提交
13 14 15 16

char *ceph_osdmap_state_str(char *str, int len, int state)
{
	if (!len)
17
		return str;
S
Sage Weil 已提交
18

19 20 21 22 23 24 25
	if ((state & CEPH_OSD_EXISTS) && (state & CEPH_OSD_UP))
		snprintf(str, len, "exists, up");
	else if (state & CEPH_OSD_EXISTS)
		snprintf(str, len, "exists");
	else if (state & CEPH_OSD_UP)
		snprintf(str, len, "up");
	else
S
Sage Weil 已提交
26
		snprintf(str, len, "doesn't exist");
27

S
Sage Weil 已提交
28 29 30 31 32
	return str;
}

/* maps */

33
static int calc_bits_of(unsigned int t)
S
Sage Weil 已提交
34 35 36 37 38 39 40 41 42 43 44 45 46 47
{
	int b = 0;
	while (t) {
		t = t >> 1;
		b++;
	}
	return b;
}

/*
 * the foo_mask is the smallest value 2^n-1 that is >= foo.
 */
static void calc_pg_masks(struct ceph_pg_pool_info *pi)
{
48 49
	pi->pg_num_mask = (1 << calc_bits_of(pi->pg_num-1)) - 1;
	pi->pgp_num_mask = (1 << calc_bits_of(pi->pgp_num-1)) - 1;
S
Sage Weil 已提交
50 51 52 53 54 55 56 57 58 59
}

/*
 * decode crush map
 */
static int crush_decode_uniform_bucket(void **p, void *end,
				       struct crush_bucket_uniform *b)
{
	dout("crush_decode_uniform_bucket %p to %p\n", *p, end);
	ceph_decode_need(p, end, (1+b->h.size) * sizeof(u32), bad);
60
	b->item_weight = ceph_decode_32(p);
S
Sage Weil 已提交
61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78
	return 0;
bad:
	return -EINVAL;
}

static int crush_decode_list_bucket(void **p, void *end,
				    struct crush_bucket_list *b)
{
	int j;
	dout("crush_decode_list_bucket %p to %p\n", *p, end);
	b->item_weights = kcalloc(b->h.size, sizeof(u32), GFP_NOFS);
	if (b->item_weights == NULL)
		return -ENOMEM;
	b->sum_weights = kcalloc(b->h.size, sizeof(u32), GFP_NOFS);
	if (b->sum_weights == NULL)
		return -ENOMEM;
	ceph_decode_need(p, end, 2 * b->h.size * sizeof(u32), bad);
	for (j = 0; j < b->h.size; j++) {
79 80
		b->item_weights[j] = ceph_decode_32(p);
		b->sum_weights[j] = ceph_decode_32(p);
S
Sage Weil 已提交
81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97
	}
	return 0;
bad:
	return -EINVAL;
}

static int crush_decode_tree_bucket(void **p, void *end,
				    struct crush_bucket_tree *b)
{
	int j;
	dout("crush_decode_tree_bucket %p to %p\n", *p, end);
	ceph_decode_32_safe(p, end, b->num_nodes, bad);
	b->node_weights = kcalloc(b->num_nodes, sizeof(u32), GFP_NOFS);
	if (b->node_weights == NULL)
		return -ENOMEM;
	ceph_decode_need(p, end, b->num_nodes * sizeof(u32), bad);
	for (j = 0; j < b->num_nodes; j++)
98
		b->node_weights[j] = ceph_decode_32(p);
S
Sage Weil 已提交
99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116
	return 0;
bad:
	return -EINVAL;
}

static int crush_decode_straw_bucket(void **p, void *end,
				     struct crush_bucket_straw *b)
{
	int j;
	dout("crush_decode_straw_bucket %p to %p\n", *p, end);
	b->item_weights = kcalloc(b->h.size, sizeof(u32), GFP_NOFS);
	if (b->item_weights == NULL)
		return -ENOMEM;
	b->straws = kcalloc(b->h.size, sizeof(u32), GFP_NOFS);
	if (b->straws == NULL)
		return -ENOMEM;
	ceph_decode_need(p, end, 2 * b->h.size * sizeof(u32), bad);
	for (j = 0; j < b->h.size; j++) {
117 118
		b->item_weights[j] = ceph_decode_32(p);
		b->straws[j] = ceph_decode_32(p);
S
Sage Weil 已提交
119 120 121 122 123 124
	}
	return 0;
bad:
	return -EINVAL;
}

S
Sage Weil 已提交
125 126 127 128 129 130 131 132 133 134 135 136 137 138 139
static int skip_name_map(void **p, void *end)
{
        int len;
        ceph_decode_32_safe(p, end, len ,bad);
        while (len--) {
                int strlen;
                *p += sizeof(u32);
                ceph_decode_32_safe(p, end, strlen, bad);
                *p += strlen;
}
        return 0;
bad:
        return -EINVAL;
}

S
Sage Weil 已提交
140 141 142 143 144 145 146 147
static struct crush_map *crush_decode(void *pbyval, void *end)
{
	struct crush_map *c;
	int err = -EINVAL;
	int i, j;
	void **p = &pbyval;
	void *start = pbyval;
	u32 magic;
S
Sage Weil 已提交
148
	u32 num_name_maps;
S
Sage Weil 已提交
149 150 151 152 153 154 155

	dout("crush_decode %p to %p len %d\n", *p, end, (int)(end - *p));

	c = kzalloc(sizeof(*c), GFP_NOFS);
	if (c == NULL)
		return ERR_PTR(-ENOMEM);

S
Sage Weil 已提交
156 157 158 159
        /* set tunables to default values */
        c->choose_local_tries = 2;
        c->choose_local_fallback_tries = 5;
        c->choose_total_tries = 19;
160
	c->chooseleaf_descend_once = 0;
S
Sage Weil 已提交
161

S
Sage Weil 已提交
162
	ceph_decode_need(p, end, 4*sizeof(u32), bad);
163
	magic = ceph_decode_32(p);
S
Sage Weil 已提交
164 165
	if (magic != CRUSH_MAGIC) {
		pr_err("crush_decode magic %x != current %x\n",
166
		       (unsigned int)magic, (unsigned int)CRUSH_MAGIC);
S
Sage Weil 已提交
167 168
		goto bad;
	}
169 170 171
	c->max_buckets = ceph_decode_32(p);
	c->max_rules = ceph_decode_32(p);
	c->max_devices = ceph_decode_32(p);
S
Sage Weil 已提交
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

	c->buckets = kcalloc(c->max_buckets, sizeof(*c->buckets), GFP_NOFS);
	if (c->buckets == NULL)
		goto badmem;
	c->rules = kcalloc(c->max_rules, sizeof(*c->rules), GFP_NOFS);
	if (c->rules == NULL)
		goto badmem;

	/* buckets */
	for (i = 0; i < c->max_buckets; i++) {
		int size = 0;
		u32 alg;
		struct crush_bucket *b;

		ceph_decode_32_safe(p, end, alg, bad);
		if (alg == 0) {
			c->buckets[i] = NULL;
			continue;
		}
		dout("crush_decode bucket %d off %x %p to %p\n",
		     i, (int)(*p-start), *p, end);

		switch (alg) {
		case CRUSH_BUCKET_UNIFORM:
			size = sizeof(struct crush_bucket_uniform);
			break;
		case CRUSH_BUCKET_LIST:
			size = sizeof(struct crush_bucket_list);
			break;
		case CRUSH_BUCKET_TREE:
			size = sizeof(struct crush_bucket_tree);
			break;
		case CRUSH_BUCKET_STRAW:
			size = sizeof(struct crush_bucket_straw);
			break;
		default:
208
			err = -EINVAL;
S
Sage Weil 已提交
209 210 211 212 213 214 215 216
			goto bad;
		}
		BUG_ON(size == 0);
		b = c->buckets[i] = kzalloc(size, GFP_NOFS);
		if (b == NULL)
			goto badmem;

		ceph_decode_need(p, end, 4*sizeof(u32), bad);
217 218
		b->id = ceph_decode_32(p);
		b->type = ceph_decode_16(p);
219 220
		b->alg = ceph_decode_8(p);
		b->hash = ceph_decode_8(p);
221 222
		b->weight = ceph_decode_32(p);
		b->size = ceph_decode_32(p);
S
Sage Weil 已提交
223 224 225 226 227 228 229 230 231 232 233 234 235 236

		dout("crush_decode bucket size %d off %x %p to %p\n",
		     b->size, (int)(*p-start), *p, end);

		b->items = kcalloc(b->size, sizeof(__s32), GFP_NOFS);
		if (b->items == NULL)
			goto badmem;
		b->perm = kcalloc(b->size, sizeof(u32), GFP_NOFS);
		if (b->perm == NULL)
			goto badmem;
		b->perm_n = 0;

		ceph_decode_need(p, end, b->size*sizeof(u32), bad);
		for (j = 0; j < b->size; j++)
237
			b->items[j] = ceph_decode_32(p);
S
Sage Weil 已提交
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

		switch (b->alg) {
		case CRUSH_BUCKET_UNIFORM:
			err = crush_decode_uniform_bucket(p, end,
				  (struct crush_bucket_uniform *)b);
			if (err < 0)
				goto bad;
			break;
		case CRUSH_BUCKET_LIST:
			err = crush_decode_list_bucket(p, end,
			       (struct crush_bucket_list *)b);
			if (err < 0)
				goto bad;
			break;
		case CRUSH_BUCKET_TREE:
			err = crush_decode_tree_bucket(p, end,
				(struct crush_bucket_tree *)b);
			if (err < 0)
				goto bad;
			break;
		case CRUSH_BUCKET_STRAW:
			err = crush_decode_straw_bucket(p, end,
				(struct crush_bucket_straw *)b);
			if (err < 0)
				goto bad;
			break;
		}
	}

	/* rules */
	dout("rule vec is %p\n", c->rules);
	for (i = 0; i < c->max_rules; i++) {
		u32 yes;
		struct crush_rule *r;

		ceph_decode_32_safe(p, end, yes, bad);
		if (!yes) {
			dout("crush_decode NO rule %d off %x %p to %p\n",
			     i, (int)(*p-start), *p, end);
			c->rules[i] = NULL;
			continue;
		}

		dout("crush_decode rule %d off %x %p to %p\n",
		     i, (int)(*p-start), *p, end);

		/* len */
		ceph_decode_32_safe(p, end, yes, bad);
#if BITS_PER_LONG == 32
287
		err = -EINVAL;
288 289
		if (yes > (ULONG_MAX - sizeof(*r))
			  / sizeof(struct crush_rule_step))
S
Sage Weil 已提交
290 291 292 293 294 295 296 297 298 299 300 301
			goto bad;
#endif
		r = c->rules[i] = kmalloc(sizeof(*r) +
					  yes*sizeof(struct crush_rule_step),
					  GFP_NOFS);
		if (r == NULL)
			goto badmem;
		dout(" rule %d is at %p\n", i, r);
		r->len = yes;
		ceph_decode_copy_safe(p, end, &r->mask, 4, bad); /* 4 u8's */
		ceph_decode_need(p, end, r->len*3*sizeof(u32), bad);
		for (j = 0; j < r->len; j++) {
302 303 304
			r->steps[j].op = ceph_decode_32(p);
			r->steps[j].arg1 = ceph_decode_32(p);
			r->steps[j].arg2 = ceph_decode_32(p);
S
Sage Weil 已提交
305 306 307 308
		}
	}

	/* ignore trailing name maps. */
S
Sage Weil 已提交
309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325
        for (num_name_maps = 0; num_name_maps < 3; num_name_maps++) {
                err = skip_name_map(p, end);
                if (err < 0)
                        goto done;
        }

        /* tunables */
        ceph_decode_need(p, end, 3*sizeof(u32), done);
        c->choose_local_tries = ceph_decode_32(p);
        c->choose_local_fallback_tries =  ceph_decode_32(p);
        c->choose_total_tries = ceph_decode_32(p);
        dout("crush decode tunable choose_local_tries = %d",
             c->choose_local_tries);
        dout("crush decode tunable choose_local_fallback_tries = %d",
             c->choose_local_fallback_tries);
        dout("crush decode tunable choose_total_tries = %d",
             c->choose_total_tries);
S
Sage Weil 已提交
326

327 328 329 330 331
	ceph_decode_need(p, end, sizeof(u32), done);
	c->chooseleaf_descend_once = ceph_decode_32(p);
	dout("crush decode tunable chooseleaf_descend_once = %d",
	     c->chooseleaf_descend_once);

S
Sage Weil 已提交
332
done:
S
Sage Weil 已提交
333 334 335 336 337 338 339 340 341 342 343 344
	dout("crush_decode success\n");
	return c;

badmem:
	err = -ENOMEM;
bad:
	dout("crush_decode fail %d\n", err);
	crush_destroy(c);
	return ERR_PTR(err);
}

/*
345 346
 * rbtree of pg_mapping for handling pg_temp (explicit mapping of pgid
 * to a set of osds)
S
Sage Weil 已提交
347
 */
348
static int pgid_cmp(struct ceph_pg l, struct ceph_pg r)
349
{
350 351 352 353 354
	if (l.pool < r.pool)
		return -1;
	if (l.pool > r.pool)
		return 1;
	if (l.seed < r.seed)
355
		return -1;
356
	if (l.seed > r.seed)
357 358 359 360
		return 1;
	return 0;
}

361 362
static int __insert_pg_mapping(struct ceph_pg_mapping *new,
			       struct rb_root *root)
S
Sage Weil 已提交
363 364 365 366
{
	struct rb_node **p = &root->rb_node;
	struct rb_node *parent = NULL;
	struct ceph_pg_mapping *pg = NULL;
367
	int c;
S
Sage Weil 已提交
368

S
Sage Weil 已提交
369
	dout("__insert_pg_mapping %llx %p\n", *(u64 *)&new->pgid, new);
S
Sage Weil 已提交
370 371 372
	while (*p) {
		parent = *p;
		pg = rb_entry(parent, struct ceph_pg_mapping, node);
373 374
		c = pgid_cmp(new->pgid, pg->pgid);
		if (c < 0)
S
Sage Weil 已提交
375
			p = &(*p)->rb_left;
376
		else if (c > 0)
S
Sage Weil 已提交
377 378
			p = &(*p)->rb_right;
		else
379
			return -EEXIST;
S
Sage Weil 已提交
380 381 382 383
	}

	rb_link_node(&new->node, parent, p);
	rb_insert_color(&new->node, root);
384
	return 0;
S
Sage Weil 已提交
385 386
}

387
static struct ceph_pg_mapping *__lookup_pg_mapping(struct rb_root *root,
388
						   struct ceph_pg pgid)
389 390 391 392 393 394 395 396
{
	struct rb_node *n = root->rb_node;
	struct ceph_pg_mapping *pg;
	int c;

	while (n) {
		pg = rb_entry(n, struct ceph_pg_mapping, node);
		c = pgid_cmp(pgid, pg->pgid);
S
Sage Weil 已提交
397
		if (c < 0) {
398
			n = n->rb_left;
S
Sage Weil 已提交
399
		} else if (c > 0) {
400
			n = n->rb_right;
S
Sage Weil 已提交
401
		} else {
402 403
			dout("__lookup_pg_mapping %lld.%x got %p\n",
			     pgid.pool, pgid.seed, pg);
404
			return pg;
S
Sage Weil 已提交
405
		}
406 407 408 409
	}
	return NULL;
}

410
static int __remove_pg_mapping(struct rb_root *root, struct ceph_pg pgid)
S
Sage Weil 已提交
411 412 413 414
{
	struct ceph_pg_mapping *pg = __lookup_pg_mapping(root, pgid);

	if (pg) {
415 416
		dout("__remove_pg_mapping %lld.%x %p\n", pgid.pool, pgid.seed,
		     pg);
S
Sage Weil 已提交
417 418 419 420
		rb_erase(&pg->node, root);
		kfree(pg);
		return 0;
	}
421
	dout("__remove_pg_mapping %lld.%x dne\n", pgid.pool, pgid.seed);
S
Sage Weil 已提交
422 423 424
	return -ENOENT;
}

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
/*
 * rbtree of pg pool info
 */
static int __insert_pg_pool(struct rb_root *root, struct ceph_pg_pool_info *new)
{
	struct rb_node **p = &root->rb_node;
	struct rb_node *parent = NULL;
	struct ceph_pg_pool_info *pi = NULL;

	while (*p) {
		parent = *p;
		pi = rb_entry(parent, struct ceph_pg_pool_info, node);
		if (new->id < pi->id)
			p = &(*p)->rb_left;
		else if (new->id > pi->id)
			p = &(*p)->rb_right;
		else
			return -EEXIST;
	}

	rb_link_node(&new->node, parent, p);
	rb_insert_color(&new->node, root);
	return 0;
}

450
static struct ceph_pg_pool_info *__lookup_pg_pool(struct rb_root *root, u64 id)
451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466
{
	struct ceph_pg_pool_info *pi;
	struct rb_node *n = root->rb_node;

	while (n) {
		pi = rb_entry(n, struct ceph_pg_pool_info, node);
		if (id < pi->id)
			n = n->rb_left;
		else if (id > pi->id)
			n = n->rb_right;
		else
			return pi;
	}
	return NULL;
}

I
Ilya Dryomov 已提交
467 468 469 470 471
struct ceph_pg_pool_info *ceph_pg_pool_by_id(struct ceph_osdmap *map, u64 id)
{
	return __lookup_pg_pool(&map->pg_pools, id);
}

472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487
const char *ceph_pg_pool_name_by_id(struct ceph_osdmap *map, u64 id)
{
	struct ceph_pg_pool_info *pi;

	if (id == CEPH_NOPOOL)
		return NULL;

	if (WARN_ON_ONCE(id > (u64) INT_MAX))
		return NULL;

	pi = __lookup_pg_pool(&map->pg_pools, (int) id);

	return pi ? pi->name : NULL;
}
EXPORT_SYMBOL(ceph_pg_pool_name_by_id);

Y
Yehuda Sadeh 已提交
488 489 490 491 492 493 494 495 496 497 498 499
int ceph_pg_poolid_by_name(struct ceph_osdmap *map, const char *name)
{
	struct rb_node *rbp;

	for (rbp = rb_first(&map->pg_pools); rbp; rbp = rb_next(rbp)) {
		struct ceph_pg_pool_info *pi =
			rb_entry(rbp, struct ceph_pg_pool_info, node);
		if (pi->name && strcmp(pi->name, name) == 0)
			return pi->id;
	}
	return -ENOENT;
}
500
EXPORT_SYMBOL(ceph_pg_poolid_by_name);
Y
Yehuda Sadeh 已提交
501

502 503 504 505 506 507 508
static void __remove_pg_pool(struct rb_root *root, struct ceph_pg_pool_info *pi)
{
	rb_erase(&pi->node, root);
	kfree(pi->name);
	kfree(pi);
}

509
static int decode_pool(void **p, void *end, struct ceph_pg_pool_info *pi)
510
{
511 512 513 514 515 516 517 518 519 520 521
	u8 ev, cv;
	unsigned len, num;
	void *pool_end;

	ceph_decode_need(p, end, 2 + 4, bad);
	ev = ceph_decode_8(p);  /* encoding version */
	cv = ceph_decode_8(p); /* compat version */
	if (ev < 5) {
		pr_warning("got v %d < 5 cv %d of ceph_pg_pool\n", ev, cv);
		return -EINVAL;
	}
522 523
	if (cv > 9) {
		pr_warning("got v %d cv %d > 9 of ceph_pg_pool\n", ev, cv);
524 525 526 527 528
		return -EINVAL;
	}
	len = ceph_decode_32(p);
	ceph_decode_need(p, end, len, bad);
	pool_end = *p + len;
S
Sage Weil 已提交
529

530 531 532 533
	pi->type = ceph_decode_8(p);
	pi->size = ceph_decode_8(p);
	pi->crush_ruleset = ceph_decode_8(p);
	pi->object_hash = ceph_decode_8(p);
S
Sage Weil 已提交
534

535 536 537 538 539 540 541 542 543 544 545 546 547 548
	pi->pg_num = ceph_decode_32(p);
	pi->pgp_num = ceph_decode_32(p);

	*p += 4 + 4;  /* skip lpg* */
	*p += 4;      /* skip last_change */
	*p += 8 + 4;  /* skip snap_seq, snap_epoch */

	/* skip snaps */
	num = ceph_decode_32(p);
	while (num--) {
		*p += 8;  /* snapid key */
		*p += 1 + 1; /* versions */
		len = ceph_decode_32(p);
		*p += len;
S
Sage Weil 已提交
549 550
	}

551
	/* skip removed_snaps */
552 553 554 555 556
	num = ceph_decode_32(p);
	*p += num * (8 + 8);

	*p += 8;  /* skip auid */
	pi->flags = ceph_decode_64(p);
557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578
	*p += 4;  /* skip crash_replay_interval */

	if (ev >= 7)
		*p += 1;  /* skip min_size */

	if (ev >= 8)
		*p += 8 + 8;  /* skip quota_max_* */

	if (ev >= 9) {
		/* skip tiers */
		num = ceph_decode_32(p);
		*p += num * 8;

		*p += 8;  /* skip tier_of */
		*p += 1;  /* skip cache_mode */

		pi->read_tier = ceph_decode_64(p);
		pi->write_tier = ceph_decode_64(p);
	} else {
		pi->read_tier = -1;
		pi->write_tier = -1;
	}
579 580 581 582 583

	/* ignore the rest */

	*p = pool_end;
	calc_pg_masks(pi);
S
Sage Weil 已提交
584 585 586 587
	return 0;

bad:
	return -EINVAL;
588 589
}

590
static int decode_pool_names(void **p, void *end, struct ceph_osdmap *map)
591 592
{
	struct ceph_pg_pool_info *pi;
593 594
	u32 num, len;
	u64 pool;
595 596 597 598

	ceph_decode_32_safe(p, end, num, bad);
	dout(" %d pool names\n", num);
	while (num--) {
599
		ceph_decode_64_safe(p, end, pool, bad);
600
		ceph_decode_32_safe(p, end, len, bad);
601
		dout("  pool %llu len %d\n", pool, len);
602
		ceph_decode_need(p, end, len, bad);
603 604
		pi = __lookup_pg_pool(&map->pg_pools, pool);
		if (pi) {
605 606 607 608
			char *name = kstrndup(*p, len, GFP_NOFS);

			if (!name)
				return -ENOMEM;
609
			kfree(pi->name);
610 611
			pi->name = name;
			dout("  name is %s\n", pi->name);
612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648
		}
		*p += len;
	}
	return 0;

bad:
	return -EINVAL;
}

/*
 * osd map
 */
void ceph_osdmap_destroy(struct ceph_osdmap *map)
{
	dout("osdmap_destroy %p\n", map);
	if (map->crush)
		crush_destroy(map->crush);
	while (!RB_EMPTY_ROOT(&map->pg_temp)) {
		struct ceph_pg_mapping *pg =
			rb_entry(rb_first(&map->pg_temp),
				 struct ceph_pg_mapping, node);
		rb_erase(&pg->node, &map->pg_temp);
		kfree(pg);
	}
	while (!RB_EMPTY_ROOT(&map->pg_pools)) {
		struct ceph_pg_pool_info *pi =
			rb_entry(rb_first(&map->pg_pools),
				 struct ceph_pg_pool_info, node);
		__remove_pg_pool(&map->pg_pools, pi);
	}
	kfree(map->osd_state);
	kfree(map->osd_weight);
	kfree(map->osd_addr);
	kfree(map);
}

/*
649 650 651
 * Adjust max_osd value, (re)allocate arrays.
 *
 * The new elements are properly initialized.
652 653 654 655 656
 */
static int osdmap_set_max_osd(struct ceph_osdmap *map, int max)
{
	u8 *state;
	u32 *weight;
657 658
	struct ceph_entity_addr *addr;
	int i;
659

660 661 662 663
	state = krealloc(map->osd_state, max*sizeof(*state), GFP_NOFS);
	weight = krealloc(map->osd_weight, max*sizeof(*weight), GFP_NOFS);
	addr = krealloc(map->osd_addr, max*sizeof(*addr), GFP_NOFS);
	if (!state || !weight || !addr) {
664 665
		kfree(state);
		kfree(weight);
666 667
		kfree(addr);

668 669 670
		return -ENOMEM;
	}

671 672 673 674
	for (i = map->max_osd; i < max; i++) {
		state[i] = 0;
		weight[i] = CEPH_OSD_OUT;
		memset(addr + i, 0, sizeof(*addr));
675 676 677 678 679
	}

	map->osd_state = state;
	map->osd_weight = weight;
	map->osd_addr = addr;
680

681
	map->max_osd = max;
682

683 684 685
	return 0;
}

686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742
#define OSDMAP_WRAPPER_COMPAT_VER	7
#define OSDMAP_CLIENT_DATA_COMPAT_VER	1

/*
 * Return 0 or error.  On success, *v is set to 0 for old (v6) osdmaps,
 * to struct_v of the client_data section for new (v7 and above)
 * osdmaps.
 */
static int get_osdmap_client_data_v(void **p, void *end,
				    const char *prefix, u8 *v)
{
	u8 struct_v;

	ceph_decode_8_safe(p, end, struct_v, e_inval);
	if (struct_v >= 7) {
		u8 struct_compat;

		ceph_decode_8_safe(p, end, struct_compat, e_inval);
		if (struct_compat > OSDMAP_WRAPPER_COMPAT_VER) {
			pr_warning("got v %d cv %d > %d of %s ceph_osdmap\n",
				   struct_v, struct_compat,
				   OSDMAP_WRAPPER_COMPAT_VER, prefix);
			return -EINVAL;
		}
		*p += 4; /* ignore wrapper struct_len */

		ceph_decode_8_safe(p, end, struct_v, e_inval);
		ceph_decode_8_safe(p, end, struct_compat, e_inval);
		if (struct_compat > OSDMAP_CLIENT_DATA_COMPAT_VER) {
			pr_warning("got v %d cv %d > %d of %s ceph_osdmap client data\n",
				   struct_v, struct_compat,
				   OSDMAP_CLIENT_DATA_COMPAT_VER, prefix);
			return -EINVAL;
		}
		*p += 4; /* ignore client data struct_len */
	} else {
		u16 version;

		*p -= 1;
		ceph_decode_16_safe(p, end, version, e_inval);
		if (version < 6) {
			pr_warning("got v %d < 6 of %s ceph_osdmap\n", version,
				   prefix);
			return -EINVAL;
		}

		/* old osdmap enconding */
		struct_v = 0;
	}

	*v = struct_v;
	return 0;

e_inval:
	return -EINVAL;
}

743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791
static int __decode_pools(void **p, void *end, struct ceph_osdmap *map,
			  bool incremental)
{
	u32 n;

	ceph_decode_32_safe(p, end, n, e_inval);
	while (n--) {
		struct ceph_pg_pool_info *pi;
		u64 pool;
		int ret;

		ceph_decode_64_safe(p, end, pool, e_inval);

		pi = __lookup_pg_pool(&map->pg_pools, pool);
		if (!incremental || !pi) {
			pi = kzalloc(sizeof(*pi), GFP_NOFS);
			if (!pi)
				return -ENOMEM;

			pi->id = pool;

			ret = __insert_pg_pool(&map->pg_pools, pi);
			if (ret) {
				kfree(pi);
				return ret;
			}
		}

		ret = decode_pool(p, end, pi);
		if (ret)
			return ret;
	}

	return 0;

e_inval:
	return -EINVAL;
}

static int decode_pools(void **p, void *end, struct ceph_osdmap *map)
{
	return __decode_pools(p, end, map, false);
}

static int decode_new_pools(void **p, void *end, struct ceph_osdmap *map)
{
	return __decode_pools(p, end, map, true);
}

792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852
static int __decode_pg_temp(void **p, void *end, struct ceph_osdmap *map,
			    bool incremental)
{
	u32 n;

	ceph_decode_32_safe(p, end, n, e_inval);
	while (n--) {
		struct ceph_pg pgid;
		u32 len, i;
		int ret;

		ret = ceph_decode_pgid(p, end, &pgid);
		if (ret)
			return ret;

		ceph_decode_32_safe(p, end, len, e_inval);

		ret = __remove_pg_mapping(&map->pg_temp, pgid);
		BUG_ON(!incremental && ret != -ENOENT);

		if (!incremental || len > 0) {
			struct ceph_pg_mapping *pg;

			ceph_decode_need(p, end, len*sizeof(u32), e_inval);

			if (len > (UINT_MAX - sizeof(*pg)) / sizeof(u32))
				return -EINVAL;

			pg = kzalloc(sizeof(*pg) + len*sizeof(u32), GFP_NOFS);
			if (!pg)
				return -ENOMEM;

			pg->pgid = pgid;
			pg->len = len;
			for (i = 0; i < len; i++)
				pg->osds[i] = ceph_decode_32(p);

			ret = __insert_pg_mapping(pg, &map->pg_temp);
			if (ret) {
				kfree(pg);
				return ret;
			}
		}
	}

	return 0;

e_inval:
	return -EINVAL;
}

static int decode_pg_temp(void **p, void *end, struct ceph_osdmap *map)
{
	return __decode_pg_temp(p, end, map, false);
}

static int decode_new_pg_temp(void **p, void *end, struct ceph_osdmap *map)
{
	return __decode_pg_temp(p, end, map, true);
}

S
Sage Weil 已提交
853 854 855
/*
 * decode a full map.
 */
856
static int osdmap_decode(void **p, void *end, struct ceph_osdmap *map)
S
Sage Weil 已提交
857
{
858
	u8 struct_v;
859
	u32 epoch = 0;
S
Sage Weil 已提交
860
	void *start = *p;
861 862
	u32 max;
	u32 len, i;
863
	int err;
S
Sage Weil 已提交
864

865
	dout("%s %p to %p len %d\n", __func__, *p, end, (int)(end - *p));
S
Sage Weil 已提交
866

867 868 869
	err = get_osdmap_client_data_v(p, end, "full", &struct_v);
	if (err)
		goto bad;
S
Sage Weil 已提交
870

871 872 873
	/* fsid, epoch, created, modified */
	ceph_decode_need(p, end, sizeof(map->fsid) + sizeof(u32) +
			 sizeof(map->created) + sizeof(map->modified), e_inval);
S
Sage Weil 已提交
874
	ceph_decode_copy(p, &map->fsid, sizeof(map->fsid));
875
	epoch = map->epoch = ceph_decode_32(p);
S
Sage Weil 已提交
876 877 878
	ceph_decode_copy(p, &map->created, sizeof(map->created));
	ceph_decode_copy(p, &map->modified, sizeof(map->modified));

879 880 881 882
	/* pools */
	err = decode_pools(p, end, map);
	if (err)
		goto bad;
883

884 885
	/* pool_name */
	err = decode_pool_names(p, end, map);
886
	if (err)
887
		goto bad;
888

889
	ceph_decode_32_safe(p, end, map->pool_max, e_inval);
S
Sage Weil 已提交
890

891
	ceph_decode_32_safe(p, end, map->flags, e_inval);
S
Sage Weil 已提交
892

893 894
	/* max_osd */
	ceph_decode_32_safe(p, end, max, e_inval);
S
Sage Weil 已提交
895 896 897

	/* (re)alloc osd arrays */
	err = osdmap_set_max_osd(map, max);
898
	if (err)
S
Sage Weil 已提交
899 900
		goto bad;

901
	/* osd_state, osd_weight, osd_addrs->client_addr */
S
Sage Weil 已提交
902 903
	ceph_decode_need(p, end, 3*sizeof(u32) +
			 map->max_osd*(1 + sizeof(*map->osd_weight) +
904 905
				       sizeof(*map->osd_addr)), e_inval);

906 907 908
	if (ceph_decode_32(p) != map->max_osd)
		goto e_inval;

S
Sage Weil 已提交
909 910
	ceph_decode_copy(p, map->osd_state, map->max_osd);

911 912 913
	if (ceph_decode_32(p) != map->max_osd)
		goto e_inval;

S
Sage Weil 已提交
914
	for (i = 0; i < map->max_osd; i++)
915
		map->osd_weight[i] = ceph_decode_32(p);
S
Sage Weil 已提交
916

917 918 919
	if (ceph_decode_32(p) != map->max_osd)
		goto e_inval;

S
Sage Weil 已提交
920
	ceph_decode_copy(p, map->osd_addr, map->max_osd*sizeof(*map->osd_addr));
921 922
	for (i = 0; i < map->max_osd; i++)
		ceph_decode_addr(&map->osd_addr[i]);
S
Sage Weil 已提交
923 924

	/* pg_temp */
925 926 927
	err = decode_pg_temp(p, end, map);
	if (err)
		goto bad;
S
Sage Weil 已提交
928 929

	/* crush */
930
	ceph_decode_32_safe(p, end, len, e_inval);
931
	map->crush = crush_decode(*p, min(*p + len, end));
S
Sage Weil 已提交
932 933 934 935 936
	if (IS_ERR(map->crush)) {
		err = PTR_ERR(map->crush);
		map->crush = NULL;
		goto bad;
	}
937
	*p += len;
S
Sage Weil 已提交
938

939
	/* ignore the rest */
S
Sage Weil 已提交
940 941
	*p = end;

942
	dout("full osdmap epoch %d max_osd %d\n", map->epoch, map->max_osd);
943
	return 0;
S
Sage Weil 已提交
944

945 946
e_inval:
	err = -EINVAL;
S
Sage Weil 已提交
947
bad:
948 949 950 951 952
	pr_err("corrupt full osdmap (%d) epoch %d off %d (%p of %p-%p)\n",
	       err, epoch, (int)(*p - start), *p, start, end);
	print_hex_dump(KERN_DEBUG, "osdmap: ",
		       DUMP_PREFIX_OFFSET, 16, 1,
		       start, end - start, true);
953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977
	return err;
}

/*
 * Allocate and decode a full map.
 */
struct ceph_osdmap *ceph_osdmap_decode(void **p, void *end)
{
	struct ceph_osdmap *map;
	int ret;

	map = kzalloc(sizeof(*map), GFP_NOFS);
	if (!map)
		return ERR_PTR(-ENOMEM);

	map->pg_temp = RB_ROOT;
	mutex_init(&map->crush_scratch_mutex);

	ret = osdmap_decode(p, end, map);
	if (ret) {
		ceph_osdmap_destroy(map);
		return ERR_PTR(ret);
	}

	return map;
S
Sage Weil 已提交
978 979 980 981 982 983 984 985 986 987 988 989 990
}

/*
 * decode and apply an incremental map update.
 */
struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end,
					     struct ceph_osdmap *map,
					     struct ceph_messenger *msgr)
{
	struct crush_map *newcrush = NULL;
	struct ceph_fsid fsid;
	u32 epoch = 0;
	struct ceph_timespec modified;
991 992 993 994
	s32 len;
	u64 pool;
	__s64 new_pool_max;
	__s32 new_flags, max;
S
Sage Weil 已提交
995
	void *start = *p;
996
	int err;
997
	u8 struct_v;
S
Sage Weil 已提交
998

999 1000
	dout("%s %p to %p len %d\n", __func__, *p, end, (int)(end - *p));

1001 1002 1003
	err = get_osdmap_client_data_v(p, end, "inc", &struct_v);
	if (err)
		goto bad;
S
Sage Weil 已提交
1004

1005 1006 1007
	/* fsid, epoch, modified, new_pool_max, new_flags */
	ceph_decode_need(p, end, sizeof(fsid) + sizeof(u32) + sizeof(modified) +
			 sizeof(u64) + sizeof(u32), e_inval);
S
Sage Weil 已提交
1008
	ceph_decode_copy(p, &fsid, sizeof(fsid));
1009
	epoch = ceph_decode_32(p);
S
Sage Weil 已提交
1010 1011
	BUG_ON(epoch != map->epoch+1);
	ceph_decode_copy(p, &modified, sizeof(modified));
1012
	new_pool_max = ceph_decode_64(p);
1013
	new_flags = ceph_decode_32(p);
S
Sage Weil 已提交
1014 1015

	/* full map? */
1016
	ceph_decode_32_safe(p, end, len, e_inval);
S
Sage Weil 已提交
1017 1018 1019
	if (len > 0) {
		dout("apply_incremental full map len %d, %p to %p\n",
		     len, *p, end);
1020
		return ceph_osdmap_decode(p, min(*p+len, end));
S
Sage Weil 已提交
1021 1022 1023
	}

	/* new crush? */
1024
	ceph_decode_32_safe(p, end, len, e_inval);
S
Sage Weil 已提交
1025 1026
	if (len > 0) {
		newcrush = crush_decode(*p, min(*p+len, end));
1027 1028 1029 1030 1031
		if (IS_ERR(newcrush)) {
			err = PTR_ERR(newcrush);
			newcrush = NULL;
			goto bad;
		}
S
Sage Weil 已提交
1032
		*p += len;
S
Sage Weil 已提交
1033 1034 1035 1036 1037
	}

	/* new flags? */
	if (new_flags >= 0)
		map->flags = new_flags;
1038 1039
	if (new_pool_max >= 0)
		map->pool_max = new_pool_max;
S
Sage Weil 已提交
1040 1041

	/* new max? */
1042
	ceph_decode_32_safe(p, end, max, e_inval);
S
Sage Weil 已提交
1043 1044
	if (max >= 0) {
		err = osdmap_set_max_osd(map, max);
1045
		if (err)
S
Sage Weil 已提交
1046 1047 1048 1049
			goto bad;
	}

	map->epoch++;
1050
	map->modified = modified;
S
Sage Weil 已提交
1051 1052 1053 1054 1055 1056 1057
	if (newcrush) {
		if (map->crush)
			crush_destroy(map->crush);
		map->crush = newcrush;
		newcrush = NULL;
	}

1058 1059 1060 1061
	/* new_pools */
	err = decode_new_pools(p, end, map);
	if (err)
		goto bad;
1062

1063 1064
	/* new_pool_names */
	err = decode_pool_names(p, end, map);
1065 1066
	if (err)
		goto bad;
S
Sage Weil 已提交
1067

1068
	/* old_pool */
1069
	ceph_decode_32_safe(p, end, len, e_inval);
1070 1071 1072
	while (len--) {
		struct ceph_pg_pool_info *pi;

1073
		ceph_decode_64_safe(p, end, pool, e_inval);
1074
		pi = __lookup_pg_pool(&map->pg_pools, pool);
1075 1076
		if (pi)
			__remove_pg_pool(&map->pg_pools, pi);
1077
	}
S
Sage Weil 已提交
1078 1079

	/* new_up */
1080
	ceph_decode_32_safe(p, end, len, e_inval);
S
Sage Weil 已提交
1081 1082 1083
	while (len--) {
		u32 osd;
		struct ceph_entity_addr addr;
1084 1085
		ceph_decode_32_safe(p, end, osd, e_inval);
		ceph_decode_copy_safe(p, end, &addr, sizeof(addr), e_inval);
1086
		ceph_decode_addr(&addr);
S
Sage Weil 已提交
1087 1088 1089 1090 1091 1092
		pr_info("osd%d up\n", osd);
		BUG_ON(osd >= map->max_osd);
		map->osd_state[osd] |= CEPH_OSD_UP;
		map->osd_addr[osd] = addr;
	}

1093
	/* new_state */
1094
	ceph_decode_32_safe(p, end, len, e_inval);
S
Sage Weil 已提交
1095 1096
	while (len--) {
		u32 osd;
1097
		u8 xorstate;
1098
		ceph_decode_32_safe(p, end, osd, e_inval);
1099
		xorstate = **(u8 **)p;
S
Sage Weil 已提交
1100
		(*p)++;  /* clean flag */
1101 1102 1103 1104
		if (xorstate == 0)
			xorstate = CEPH_OSD_UP;
		if (xorstate & CEPH_OSD_UP)
			pr_info("osd%d down\n", osd);
S
Sage Weil 已提交
1105
		if (osd < map->max_osd)
1106
			map->osd_state[osd] ^= xorstate;
S
Sage Weil 已提交
1107 1108 1109
	}

	/* new_weight */
1110
	ceph_decode_32_safe(p, end, len, e_inval);
S
Sage Weil 已提交
1111 1112
	while (len--) {
		u32 osd, off;
1113
		ceph_decode_need(p, end, sizeof(u32)*2, e_inval);
1114 1115
		osd = ceph_decode_32(p);
		off = ceph_decode_32(p);
S
Sage Weil 已提交
1116 1117 1118 1119 1120 1121 1122 1123
		pr_info("osd%d weight 0x%x %s\n", osd, off,
		     off == CEPH_OSD_IN ? "(in)" :
		     (off == CEPH_OSD_OUT ? "(out)" : ""));
		if (osd < map->max_osd)
			map->osd_weight[osd] = off;
	}

	/* new_pg_temp */
1124 1125 1126
	err = decode_new_pg_temp(p, end, map);
	if (err)
		goto bad;
S
Sage Weil 已提交
1127 1128 1129

	/* ignore the rest */
	*p = end;
1130 1131

	dout("inc osdmap epoch %d max_osd %d\n", map->epoch, map->max_osd);
S
Sage Weil 已提交
1132 1133
	return map;

1134 1135
e_inval:
	err = -EINVAL;
S
Sage Weil 已提交
1136
bad:
1137 1138
	pr_err("corrupt inc osdmap (%d) epoch %d off %d (%p of %p-%p)\n",
	       err, epoch, (int)(*p - start), *p, start, end);
1139 1140 1141
	print_hex_dump(KERN_DEBUG, "osdmap: ",
		       DUMP_PREFIX_OFFSET, 16, 1,
		       start, end - start, true);
S
Sage Weil 已提交
1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157
	if (newcrush)
		crush_destroy(newcrush);
	return ERR_PTR(err);
}




/*
 * calculate file layout from given offset, length.
 * fill in correct oid, logical length, and object extent
 * offset, length.
 *
 * for now, we write only a single su, until we can
 * pass a stride back to the caller.
 */
S
Sage Weil 已提交
1158
int ceph_calc_file_object_mapping(struct ceph_file_layout *layout,
1159
				   u64 off, u64 len,
1160
				   u64 *ono,
S
Sage Weil 已提交
1161 1162 1163 1164 1165 1166 1167
				   u64 *oxoff, u64 *oxlen)
{
	u32 osize = le32_to_cpu(layout->fl_object_size);
	u32 su = le32_to_cpu(layout->fl_stripe_unit);
	u32 sc = le32_to_cpu(layout->fl_stripe_count);
	u32 bl, stripeno, stripepos, objsetno;
	u32 su_per_object;
1168
	u64 t, su_offset;
S
Sage Weil 已提交
1169

1170
	dout("mapping %llu~%llu  osize %u fl_su %u\n", off, len,
S
Sage Weil 已提交
1171
	     osize, su);
S
Sage Weil 已提交
1172 1173
	if (su == 0 || sc == 0)
		goto invalid;
1174
	su_per_object = osize / su;
S
Sage Weil 已提交
1175 1176
	if (su_per_object == 0)
		goto invalid;
S
Sage Weil 已提交
1177 1178 1179
	dout("osize %u / su %u = su_per_object %u\n", osize, su,
	     su_per_object);

S
Sage Weil 已提交
1180 1181 1182
	if ((su & ~PAGE_MASK) != 0)
		goto invalid;

S
Sage Weil 已提交
1183 1184 1185 1186 1187 1188 1189 1190 1191 1192
	/* bl = *off / su; */
	t = off;
	do_div(t, su);
	bl = t;
	dout("off %llu / su %u = bl %u\n", off, su, bl);

	stripeno = bl / sc;
	stripepos = bl % sc;
	objsetno = stripeno / su_per_object;

1193
	*ono = objsetno * sc + stripepos;
1194
	dout("objset %u * sc %u = ono %u\n", objsetno, sc, (unsigned int)*ono);
1195 1196

	/* *oxoff = *off % layout->fl_stripe_unit;  # offset in su */
S
Sage Weil 已提交
1197
	t = off;
1198 1199 1200 1201 1202
	su_offset = do_div(t, su);
	*oxoff = su_offset + (stripeno % su_per_object) * su;

	/*
	 * Calculate the length of the extent being written to the selected
1203
	 * object. This is the minimum of the full length requested (len) or
1204 1205
	 * the remainder of the current stripe being written to.
	 */
1206
	*oxlen = min_t(u64, len, su - su_offset);
S
Sage Weil 已提交
1207 1208

	dout(" obj extent %llu~%llu\n", *oxoff, *oxlen);
S
Sage Weil 已提交
1209 1210 1211 1212 1213 1214 1215 1216
	return 0;

invalid:
	dout(" invalid layout\n");
	*ono = 0;
	*oxoff = 0;
	*oxlen = 0;
	return -EINVAL;
S
Sage Weil 已提交
1217
}
1218
EXPORT_SYMBOL(ceph_calc_file_object_mapping);
S
Sage Weil 已提交
1219 1220

/*
1221 1222 1223
 * Calculate mapping of a (oloc, oid) pair to a PG.  Should only be
 * called with target's (oloc, oid), since tiering isn't taken into
 * account.
S
Sage Weil 已提交
1224
 */
1225 1226 1227 1228
int ceph_oloc_oid_to_pg(struct ceph_osdmap *osdmap,
			struct ceph_object_locator *oloc,
			struct ceph_object_id *oid,
			struct ceph_pg *pg_out)
S
Sage Weil 已提交
1229
{
1230
	struct ceph_pg_pool_info *pi;
S
Sage Weil 已提交
1231

1232 1233
	pi = __lookup_pg_pool(&osdmap->pg_pools, oloc->pool);
	if (!pi)
1234
		return -EIO;
S
Sage Weil 已提交
1235

1236 1237 1238 1239 1240 1241
	pg_out->pool = oloc->pool;
	pg_out->seed = ceph_str_hash(pi->object_hash, oid->name,
				     oid->name_len);

	dout("%s '%.*s' pgid %llu.%x\n", __func__, oid->name_len, oid->name,
	     pg_out->pool, pg_out->seed);
S
Sage Weil 已提交
1242 1243
	return 0;
}
1244
EXPORT_SYMBOL(ceph_oloc_oid_to_pg);
S
Sage Weil 已提交
1245

1246 1247 1248
static int do_crush(struct ceph_osdmap *map, int ruleno, int x,
		    int *result, int result_max,
		    const __u32 *weight, int weight_max)
1249
{
1250 1251 1252 1253 1254 1255 1256 1257
	int r;

	BUG_ON(result_max > CEPH_PG_MAX_SIZE);

	mutex_lock(&map->crush_scratch_mutex);
	r = crush_do_rule(map->crush, ruleno, x, result, result_max,
			  weight, weight_max, map->crush_scratch_ary);
	mutex_unlock(&map->crush_scratch_mutex);
1258

1259
	return r;
1260 1261
}

S
Sage Weil 已提交
1262 1263 1264 1265
/*
 * Calculate raw osd vector for the given pgid.  Return pointer to osd
 * array, or NULL on failure.
 */
1266
static int *calc_pg_raw(struct ceph_osdmap *osdmap, struct ceph_pg pgid,
S
Sage Weil 已提交
1267 1268 1269 1270 1271
			int *osds, int *num)
{
	struct ceph_pg_mapping *pg;
	struct ceph_pg_pool_info *pool;
	int ruleno;
1272 1273
	int r;
	u32 pps;
S
Sage Weil 已提交
1274

1275
	pool = __lookup_pg_pool(&osdmap->pg_pools, pgid.pool);
1276 1277 1278
	if (!pool)
		return NULL;

S
Sage Weil 已提交
1279
	/* pg_temp? */
1280
	pgid.seed = ceph_stable_mod(pgid.seed, pool->pg_num,
1281
				    pool->pg_num_mask);
1282 1283 1284 1285
	pg = __lookup_pg_mapping(&osdmap->pg_temp, pgid);
	if (pg) {
		*num = pg->len;
		return pg->osds;
S
Sage Weil 已提交
1286 1287 1288
	}

	/* crush */
1289 1290
	ruleno = crush_find_rule(osdmap->crush, pool->crush_ruleset,
				 pool->type, pool->size);
S
Sage Weil 已提交
1291
	if (ruleno < 0) {
1292 1293
		pr_err("no crush rule pool %lld ruleset %d type %d size %d\n",
		       pgid.pool, pool->crush_ruleset, pool->type,
1294
		       pool->size);
S
Sage Weil 已提交
1295 1296 1297
		return NULL;
	}

1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314
	if (pool->flags & CEPH_POOL_FLAG_HASHPSPOOL) {
		/* hash pool id and seed sothat pool PGs do not overlap */
		pps = crush_hash32_2(CRUSH_HASH_RJENKINS1,
				     ceph_stable_mod(pgid.seed, pool->pgp_num,
						     pool->pgp_num_mask),
				     pgid.pool);
	} else {
		/*
		 * legacy ehavior: add ps and pool together.  this is
		 * not a great approach because the PGs from each pool
		 * will overlap on top of each other: 0.5 == 1.4 ==
		 * 2.3 == ...
		 */
		pps = ceph_stable_mod(pgid.seed, pool->pgp_num,
				      pool->pgp_num_mask) +
			(unsigned)pgid.pool;
	}
1315 1316
	r = do_crush(osdmap, ruleno, pps, osds, min_t(int, pool->size, *num),
		     osdmap->osd_weight, osdmap->max_osd);
S
Sage Weil 已提交
1317
	if (r < 0) {
1318 1319
		pr_err("error %d from crush rule: pool %lld ruleset %d type %d"
		       " size %d\n", r, pgid.pool, pool->crush_ruleset,
1320
		       pool->type, pool->size);
S
Sage Weil 已提交
1321 1322 1323
		return NULL;
	}
	*num = r;
S
Sage Weil 已提交
1324 1325 1326
	return osds;
}

1327 1328 1329
/*
 * Return acting set for given pgid.
 */
1330
int ceph_calc_pg_acting(struct ceph_osdmap *osdmap, struct ceph_pg pgid,
1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347
			int *acting)
{
	int rawosds[CEPH_PG_MAX_SIZE], *osds;
	int i, o, num = CEPH_PG_MAX_SIZE;

	osds = calc_pg_raw(osdmap, pgid, rawosds, &num);
	if (!osds)
		return -1;

	/* primary is first up osd */
	o = 0;
	for (i = 0; i < num; i++)
		if (ceph_osd_is_up(osdmap, osds[i]))
			acting[o++] = osds[i];
	return o;
}

S
Sage Weil 已提交
1348 1349 1350
/*
 * Return primary osd for given pgid, or -1 if none.
 */
1351
int ceph_calc_pg_primary(struct ceph_osdmap *osdmap, struct ceph_pg pgid)
S
Sage Weil 已提交
1352
{
1353 1354
	int rawosds[CEPH_PG_MAX_SIZE], *osds;
	int i, num = CEPH_PG_MAX_SIZE;
S
Sage Weil 已提交
1355 1356 1357 1358 1359 1360 1361

	osds = calc_pg_raw(osdmap, pgid, rawosds, &num);
	if (!osds)
		return -1;

	/* primary is first up osd */
	for (i = 0; i < num; i++)
1362
		if (ceph_osd_is_up(osdmap, osds[i]))
S
Sage Weil 已提交
1363 1364 1365
			return osds[i];
	return -1;
}
1366
EXPORT_SYMBOL(ceph_calc_pg_primary);