mds_client.c 86.7 KB
Newer Older
1
#include <linux/ceph/ceph_debug.h>
S
Sage Weil 已提交
2

S
Sage Weil 已提交
3
#include <linux/fs.h>
S
Sage Weil 已提交
4
#include <linux/wait.h>
5
#include <linux/slab.h>
S
Sage Weil 已提交
6
#include <linux/sched.h>
7 8
#include <linux/debugfs.h>
#include <linux/seq_file.h>
S
Sage Weil 已提交
9 10

#include "super.h"
11 12 13 14 15 16 17
#include "mds_client.h"

#include <linux/ceph/messenger.h>
#include <linux/ceph/decode.h>
#include <linux/ceph/pagelist.h>
#include <linux/ceph/auth.h>
#include <linux/ceph/debugfs.h>
S
Sage Weil 已提交
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

/*
 * A cluster of MDS (metadata server) daemons is responsible for
 * managing the file system namespace (the directory hierarchy and
 * inodes) and for coordinating shared access to storage.  Metadata is
 * partitioning hierarchically across a number of servers, and that
 * partition varies over time as the cluster adjusts the distribution
 * in order to balance load.
 *
 * The MDS client is primarily responsible to managing synchronous
 * metadata requests for operations like open, unlink, and so forth.
 * If there is a MDS failure, we find out about it when we (possibly
 * request and) receive a new MDS map, and can resubmit affected
 * requests.
 *
 * For the most part, though, we take advantage of a lossless
 * communications channel to the MDS, and do not need to worry about
 * timing out or resubmitting requests.
 *
 * We maintain a stateful "session" with each MDS we interact with.
 * Within each session, we sent periodic heartbeat messages to ensure
 * any capabilities or leases we have been issues remain valid.  If
 * the session times out and goes stale, our leases and capabilities
 * are no longer valid.
 */

S
Sage Weil 已提交
44 45 46 47 48
struct ceph_reconnect_state {
	struct ceph_pagelist *pagelist;
	bool flock;
};

S
Sage Weil 已提交
49 50 51
static void __wake_requests(struct ceph_mds_client *mdsc,
			    struct list_head *head);

52
static const struct ceph_connection_operations mds_con_ops;
S
Sage Weil 已提交
53 54 55 56 57 58 59 60 61 62


/*
 * mds reply parsing
 */

/*
 * parse individual inode info
 */
static int parse_reply_info_in(void **p, void *end,
63 64
			       struct ceph_mds_reply_info_in *info,
			       int features)
S
Sage Weil 已提交
65 66 67 68 69 70 71 72 73 74 75 76 77
{
	int err = -EIO;

	info->in = *p;
	*p += sizeof(struct ceph_mds_reply_inode) +
		sizeof(*info->in->fragtree.splits) *
		le32_to_cpu(info->in->fragtree.nsplits);

	ceph_decode_32_safe(p, end, info->symlink_len, bad);
	ceph_decode_need(p, end, info->symlink_len, bad);
	info->symlink = *p;
	*p += info->symlink_len;

78 79 80 81 82 83
	if (features & CEPH_FEATURE_DIRLAYOUTHASH)
		ceph_decode_copy_safe(p, end, &info->dir_layout,
				      sizeof(info->dir_layout), bad);
	else
		memset(&info->dir_layout, 0, sizeof(info->dir_layout));

S
Sage Weil 已提交
84 85 86 87 88 89 90 91 92 93 94 95 96 97
	ceph_decode_32_safe(p, end, info->xattr_len, bad);
	ceph_decode_need(p, end, info->xattr_len, bad);
	info->xattr_data = *p;
	*p += info->xattr_len;
	return 0;
bad:
	return err;
}

/*
 * parse a normal reply, which may contain a (dir+)dentry and/or a
 * target inode.
 */
static int parse_reply_info_trace(void **p, void *end,
98 99
				  struct ceph_mds_reply_info_parsed *info,
				  int features)
S
Sage Weil 已提交
100 101 102 103
{
	int err;

	if (info->head->is_dentry) {
104
		err = parse_reply_info_in(p, end, &info->diri, features);
S
Sage Weil 已提交
105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124
		if (err < 0)
			goto out_bad;

		if (unlikely(*p + sizeof(*info->dirfrag) > end))
			goto bad;
		info->dirfrag = *p;
		*p += sizeof(*info->dirfrag) +
			sizeof(u32)*le32_to_cpu(info->dirfrag->ndist);
		if (unlikely(*p > end))
			goto bad;

		ceph_decode_32_safe(p, end, info->dname_len, bad);
		ceph_decode_need(p, end, info->dname_len, bad);
		info->dname = *p;
		*p += info->dname_len;
		info->dlease = *p;
		*p += sizeof(*info->dlease);
	}

	if (info->head->is_target) {
125
		err = parse_reply_info_in(p, end, &info->targeti, features);
S
Sage Weil 已提交
126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144
		if (err < 0)
			goto out_bad;
	}

	if (unlikely(*p != end))
		goto bad;
	return 0;

bad:
	err = -EIO;
out_bad:
	pr_err("problem parsing mds trace %d\n", err);
	return err;
}

/*
 * parse readdir results
 */
static int parse_reply_info_dir(void **p, void *end,
145 146
				struct ceph_mds_reply_info_parsed *info,
				int features)
S
Sage Weil 已提交
147 148 149 150 151 152 153 154 155 156 157 158 159
{
	u32 num, i = 0;
	int err;

	info->dir_dir = *p;
	if (*p + sizeof(*info->dir_dir) > end)
		goto bad;
	*p += sizeof(*info->dir_dir) +
		sizeof(u32)*le32_to_cpu(info->dir_dir->ndist);
	if (*p > end)
		goto bad;

	ceph_decode_need(p, end, sizeof(num) + 2, bad);
160 161 162
	num = ceph_decode_32(p);
	info->dir_end = ceph_decode_8(p);
	info->dir_complete = ceph_decode_8(p);
S
Sage Weil 已提交
163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183
	if (num == 0)
		goto done;

	/* alloc large array */
	info->dir_nr = num;
	info->dir_in = kcalloc(num, sizeof(*info->dir_in) +
			       sizeof(*info->dir_dname) +
			       sizeof(*info->dir_dname_len) +
			       sizeof(*info->dir_dlease),
			       GFP_NOFS);
	if (info->dir_in == NULL) {
		err = -ENOMEM;
		goto out_bad;
	}
	info->dir_dname = (void *)(info->dir_in + num);
	info->dir_dname_len = (void *)(info->dir_dname + num);
	info->dir_dlease = (void *)(info->dir_dname_len + num);

	while (num) {
		/* dentry */
		ceph_decode_need(p, end, sizeof(u32)*2, bad);
184
		info->dir_dname_len[i] = ceph_decode_32(p);
S
Sage Weil 已提交
185 186 187 188 189 190 191 192 193
		ceph_decode_need(p, end, info->dir_dname_len[i], bad);
		info->dir_dname[i] = *p;
		*p += info->dir_dname_len[i];
		dout("parsed dir dname '%.*s'\n", info->dir_dname_len[i],
		     info->dir_dname[i]);
		info->dir_dlease[i] = *p;
		*p += sizeof(struct ceph_mds_reply_lease);

		/* inode */
194
		err = parse_reply_info_in(p, end, &info->dir_in[i], features);
S
Sage Weil 已提交
195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212
		if (err < 0)
			goto out_bad;
		i++;
		num--;
	}

done:
	if (*p != end)
		goto bad;
	return 0;

bad:
	err = -EIO;
out_bad:
	pr_err("problem parsing dir contents %d\n", err);
	return err;
}

213 214 215 216
/*
 * parse fcntl F_GETLK results
 */
static int parse_reply_info_filelock(void **p, void *end,
217 218
				     struct ceph_mds_reply_info_parsed *info,
				     int features)
219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237
{
	if (*p + sizeof(*info->filelock_reply) > end)
		goto bad;

	info->filelock_reply = *p;
	*p += sizeof(*info->filelock_reply);

	if (unlikely(*p != end))
		goto bad;
	return 0;

bad:
	return -EIO;
}

/*
 * parse extra results
 */
static int parse_reply_info_extra(void **p, void *end,
238 239
				  struct ceph_mds_reply_info_parsed *info,
				  int features)
240 241
{
	if (info->head->op == CEPH_MDS_OP_GETFILELOCK)
242
		return parse_reply_info_filelock(p, end, info, features);
243
	else
244
		return parse_reply_info_dir(p, end, info, features);
245 246
}

S
Sage Weil 已提交
247 248 249 250
/*
 * parse entire mds reply
 */
static int parse_reply_info(struct ceph_msg *msg,
251 252
			    struct ceph_mds_reply_info_parsed *info,
			    int features)
S
Sage Weil 已提交
253 254 255 256 257 258 259 260 261 262 263 264
{
	void *p, *end;
	u32 len;
	int err;

	info->head = msg->front.iov_base;
	p = msg->front.iov_base + sizeof(struct ceph_mds_reply_head);
	end = p + msg->front.iov_len - sizeof(struct ceph_mds_reply_head);

	/* trace */
	ceph_decode_32_safe(&p, end, len, bad);
	if (len > 0) {
265
		err = parse_reply_info_trace(&p, p+len, info, features);
S
Sage Weil 已提交
266 267 268 269
		if (err < 0)
			goto out_bad;
	}

270
	/* extra */
S
Sage Weil 已提交
271 272
	ceph_decode_32_safe(&p, end, len, bad);
	if (len > 0) {
273
		err = parse_reply_info_extra(&p, p+len, info, features);
S
Sage Weil 已提交
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
		if (err < 0)
			goto out_bad;
	}

	/* snap blob */
	ceph_decode_32_safe(&p, end, len, bad);
	info->snapblob_len = len;
	info->snapblob = p;
	p += len;

	if (p != end)
		goto bad;
	return 0;

bad:
	err = -EIO;
out_bad:
	pr_err("mds parse_reply err %d\n", err);
	return err;
}

static void destroy_reply_info(struct ceph_mds_reply_info_parsed *info)
{
	kfree(info->dir_in);
}


/*
 * sessions
 */
static const char *session_state_name(int s)
{
	switch (s) {
	case CEPH_MDS_SESSION_NEW: return "new";
	case CEPH_MDS_SESSION_OPENING: return "opening";
	case CEPH_MDS_SESSION_OPEN: return "open";
	case CEPH_MDS_SESSION_HUNG: return "hung";
	case CEPH_MDS_SESSION_CLOSING: return "closing";
S
Sage Weil 已提交
312
	case CEPH_MDS_SESSION_RESTARTING: return "restarting";
S
Sage Weil 已提交
313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333
	case CEPH_MDS_SESSION_RECONNECTING: return "reconnecting";
	default: return "???";
	}
}

static struct ceph_mds_session *get_session(struct ceph_mds_session *s)
{
	if (atomic_inc_not_zero(&s->s_ref)) {
		dout("mdsc get_session %p %d -> %d\n", s,
		     atomic_read(&s->s_ref)-1, atomic_read(&s->s_ref));
		return s;
	} else {
		dout("mdsc get_session %p 0 -- FAIL", s);
		return NULL;
	}
}

void ceph_put_mds_session(struct ceph_mds_session *s)
{
	dout("mdsc put_session %p %d -> %d\n", s,
	     atomic_read(&s->s_ref), atomic_read(&s->s_ref)-1);
334 335
	if (atomic_dec_and_test(&s->s_ref)) {
		if (s->s_authorizer)
336 337 338
		     s->s_mdsc->fsc->client->monc.auth->ops->destroy_authorizer(
			     s->s_mdsc->fsc->client->monc.auth,
			     s->s_authorizer);
S
Sage Weil 已提交
339
		kfree(s);
340
	}
S
Sage Weil 已提交
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
}

/*
 * called under mdsc->mutex
 */
struct ceph_mds_session *__ceph_lookup_mds_session(struct ceph_mds_client *mdsc,
						   int mds)
{
	struct ceph_mds_session *session;

	if (mds >= mdsc->max_sessions || mdsc->sessions[mds] == NULL)
		return NULL;
	session = mdsc->sessions[mds];
	dout("lookup_mds_session %p %d\n", session,
	     atomic_read(&session->s_ref));
	get_session(session);
	return session;
}

static bool __have_session(struct ceph_mds_client *mdsc, int mds)
{
	if (mds >= mdsc->max_sessions)
		return false;
	return mdsc->sessions[mds];
}

367 368 369 370 371 372 373 374 375
static int __verify_registered_session(struct ceph_mds_client *mdsc,
				       struct ceph_mds_session *s)
{
	if (s->s_mds >= mdsc->max_sessions ||
	    mdsc->sessions[s->s_mds] != s)
		return -ENOENT;
	return 0;
}

S
Sage Weil 已提交
376 377 378 379 380 381 382 383 384 385
/*
 * create+register a new session for given mds.
 * called under mdsc->mutex.
 */
static struct ceph_mds_session *register_session(struct ceph_mds_client *mdsc,
						 int mds)
{
	struct ceph_mds_session *s;

	s = kzalloc(sizeof(*s), GFP_NOFS);
D
Dan Carpenter 已提交
386 387
	if (!s)
		return ERR_PTR(-ENOMEM);
S
Sage Weil 已提交
388 389 390 391 392 393 394
	s->s_mdsc = mdsc;
	s->s_mds = mds;
	s->s_state = CEPH_MDS_SESSION_NEW;
	s->s_ttl = 0;
	s->s_seq = 0;
	mutex_init(&s->s_mutex);

395
	ceph_con_init(mdsc->fsc->client->msgr, &s->s_con);
S
Sage Weil 已提交
396 397 398 399 400 401 402 403 404 405 406 407
	s->s_con.private = s;
	s->s_con.ops = &mds_con_ops;
	s->s_con.peer_name.type = CEPH_ENTITY_TYPE_MDS;
	s->s_con.peer_name.num = cpu_to_le64(mds);

	spin_lock_init(&s->s_cap_lock);
	s->s_cap_gen = 0;
	s->s_cap_ttl = 0;
	s->s_renew_requested = 0;
	s->s_renew_seq = 0;
	INIT_LIST_HEAD(&s->s_caps);
	s->s_nr_caps = 0;
408
	s->s_trim_caps = 0;
S
Sage Weil 已提交
409 410 411 412
	atomic_set(&s->s_ref, 1);
	INIT_LIST_HEAD(&s->s_waiting);
	INIT_LIST_HEAD(&s->s_unsafe);
	s->s_num_cap_releases = 0;
S
Sage Weil 已提交
413
	s->s_cap_iterator = NULL;
S
Sage Weil 已提交
414 415 416 417 418 419 420 421 422 423 424 425 426
	INIT_LIST_HEAD(&s->s_cap_releases);
	INIT_LIST_HEAD(&s->s_cap_releases_done);
	INIT_LIST_HEAD(&s->s_cap_flushing);
	INIT_LIST_HEAD(&s->s_cap_snaps_flushing);

	dout("register_session mds%d\n", mds);
	if (mds >= mdsc->max_sessions) {
		int newmax = 1 << get_count_order(mds+1);
		struct ceph_mds_session **sa;

		dout("register_session realloc to %d\n", newmax);
		sa = kcalloc(newmax, sizeof(void *), GFP_NOFS);
		if (sa == NULL)
427
			goto fail_realloc;
S
Sage Weil 已提交
428 429 430 431 432 433 434 435 436 437
		if (mdsc->sessions) {
			memcpy(sa, mdsc->sessions,
			       mdsc->max_sessions * sizeof(void *));
			kfree(mdsc->sessions);
		}
		mdsc->sessions = sa;
		mdsc->max_sessions = newmax;
	}
	mdsc->sessions[mds] = s;
	atomic_inc(&s->s_ref);  /* one ref to sessions[], one to caller */
438 439 440

	ceph_con_open(&s->s_con, ceph_mdsmap_get_addr(mdsc->mdsmap, mds));

S
Sage Weil 已提交
441
	return s;
442 443 444 445

fail_realloc:
	kfree(s);
	return ERR_PTR(-ENOMEM);
S
Sage Weil 已提交
446 447 448 449 450
}

/*
 * called under mdsc->mutex
 */
451
static void __unregister_session(struct ceph_mds_client *mdsc,
452
			       struct ceph_mds_session *s)
S
Sage Weil 已提交
453
{
454 455
	dout("__unregister_session mds%d %p\n", s->s_mds, s);
	BUG_ON(mdsc->sessions[s->s_mds] != s);
456 457 458
	mdsc->sessions[s->s_mds] = NULL;
	ceph_con_close(&s->s_con);
	ceph_put_mds_session(s);
S
Sage Weil 已提交
459 460 461 462 463 464 465 466 467 468 469 470 471 472 473
}

/*
 * drop session refs in request.
 *
 * should be last request ref, or hold mdsc->mutex
 */
static void put_request_session(struct ceph_mds_request *req)
{
	if (req->r_session) {
		ceph_put_mds_session(req->r_session);
		req->r_session = NULL;
	}
}

474
void ceph_mdsc_release_request(struct kref *kref)
S
Sage Weil 已提交
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
	struct ceph_mds_request *req = container_of(kref,
						    struct ceph_mds_request,
						    r_kref);
	if (req->r_request)
		ceph_msg_put(req->r_request);
	if (req->r_reply) {
		ceph_msg_put(req->r_reply);
		destroy_reply_info(&req->r_reply_info);
	}
	if (req->r_inode) {
		ceph_put_cap_refs(ceph_inode(req->r_inode),
				  CEPH_CAP_PIN);
		iput(req->r_inode);
	}
	if (req->r_locked_dir)
		ceph_put_cap_refs(ceph_inode(req->r_locked_dir),
				  CEPH_CAP_PIN);
	if (req->r_target_inode)
		iput(req->r_target_inode);
	if (req->r_dentry)
		dput(req->r_dentry);
	if (req->r_old_dentry) {
		ceph_put_cap_refs(
			ceph_inode(req->r_old_dentry->d_parent->d_inode),
			CEPH_CAP_PIN);
		dput(req->r_old_dentry);
S
Sage Weil 已提交
502
	}
503 504 505
	kfree(req->r_path1);
	kfree(req->r_path2);
	put_request_session(req);
506
	ceph_unreserve_caps(req->r_mdsc, &req->r_caps_reservation);
507
	kfree(req);
S
Sage Weil 已提交
508 509 510 511 512 513 514 515 516 517 518
}

/*
 * lookup session, bump ref if found.
 *
 * called under mdsc->mutex.
 */
static struct ceph_mds_request *__lookup_request(struct ceph_mds_client *mdsc,
					     u64 tid)
{
	struct ceph_mds_request *req;
S
Sage Weil 已提交
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
	struct rb_node *n = mdsc->request_tree.rb_node;

	while (n) {
		req = rb_entry(n, struct ceph_mds_request, r_node);
		if (tid < req->r_tid)
			n = n->rb_left;
		else if (tid > req->r_tid)
			n = n->rb_right;
		else {
			ceph_mdsc_get_request(req);
			return req;
		}
	}
	return NULL;
}

static void __insert_request(struct ceph_mds_client *mdsc,
			     struct ceph_mds_request *new)
{
	struct rb_node **p = &mdsc->request_tree.rb_node;
	struct rb_node *parent = NULL;
	struct ceph_mds_request *req = NULL;

	while (*p) {
		parent = *p;
		req = rb_entry(parent, struct ceph_mds_request, r_node);
		if (new->r_tid < req->r_tid)
			p = &(*p)->rb_left;
		else if (new->r_tid > req->r_tid)
			p = &(*p)->rb_right;
		else
			BUG();
	}

	rb_link_node(&new->r_node, parent, p);
	rb_insert_color(&new->r_node, &mdsc->request_tree);
S
Sage Weil 已提交
555 556 557 558 559 560 561 562 563 564 565 566 567 568
}

/*
 * Register an in-flight request, and assign a tid.  Link to directory
 * are modifying (if any).
 *
 * Called under mdsc->mutex.
 */
static void __register_request(struct ceph_mds_client *mdsc,
			       struct ceph_mds_request *req,
			       struct inode *dir)
{
	req->r_tid = ++mdsc->last_tid;
	if (req->r_num_caps)
569 570
		ceph_reserve_caps(mdsc, &req->r_caps_reservation,
				  req->r_num_caps);
S
Sage Weil 已提交
571 572
	dout("__register_request %p tid %lld\n", req, req->r_tid);
	ceph_mdsc_get_request(req);
S
Sage Weil 已提交
573
	__insert_request(mdsc, req);
S
Sage Weil 已提交
574

575 576 577
	req->r_uid = current_fsuid();
	req->r_gid = current_fsgid();

S
Sage Weil 已提交
578 579 580
	if (dir) {
		struct ceph_inode_info *ci = ceph_inode(dir);

581
		ihold(dir);
S
Sage Weil 已提交
582 583 584 585 586 587 588 589 590 591 592
		spin_lock(&ci->i_unsafe_lock);
		req->r_unsafe_dir = dir;
		list_add_tail(&req->r_unsafe_dir_item, &ci->i_unsafe_dirops);
		spin_unlock(&ci->i_unsafe_lock);
	}
}

static void __unregister_request(struct ceph_mds_client *mdsc,
				 struct ceph_mds_request *req)
{
	dout("__unregister_request %p tid %lld\n", req, req->r_tid);
S
Sage Weil 已提交
593
	rb_erase(&req->r_node, &mdsc->request_tree);
594
	RB_CLEAR_NODE(&req->r_node);
S
Sage Weil 已提交
595 596 597 598 599 600 601

	if (req->r_unsafe_dir) {
		struct ceph_inode_info *ci = ceph_inode(req->r_unsafe_dir);

		spin_lock(&ci->i_unsafe_lock);
		list_del_init(&req->r_unsafe_dir_item);
		spin_unlock(&ci->i_unsafe_lock);
602 603 604

		iput(req->r_unsafe_dir);
		req->r_unsafe_dir = NULL;
S
Sage Weil 已提交
605
	}
606 607

	ceph_mdsc_put_request(req);
S
Sage Weil 已提交
608 609 610 611 612 613 614 615 616 617
}

/*
 * Choose mds to send request to next.  If there is a hint set in the
 * request (e.g., due to a prior forward hint from the mds), use that.
 * Otherwise, consult frag tree and/or caps to identify the
 * appropriate mds.  If all else fails, choose randomly.
 *
 * Called under mdsc->mutex.
 */
618 619 620 621 622 623 624
struct dentry *get_nonsnap_parent(struct dentry *dentry)
{
	while (!IS_ROOT(dentry) && ceph_snap(dentry->d_inode) != CEPH_NOSNAP)
		dentry = dentry->d_parent;
	return dentry;
}

S
Sage Weil 已提交
625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654
static int __choose_mds(struct ceph_mds_client *mdsc,
			struct ceph_mds_request *req)
{
	struct inode *inode;
	struct ceph_inode_info *ci;
	struct ceph_cap *cap;
	int mode = req->r_direct_mode;
	int mds = -1;
	u32 hash = req->r_direct_hash;
	bool is_hash = req->r_direct_is_hash;

	/*
	 * is there a specific mds we should try?  ignore hint if we have
	 * no session and the mds is not up (active or recovering).
	 */
	if (req->r_resend_mds >= 0 &&
	    (__have_session(mdsc, req->r_resend_mds) ||
	     ceph_mdsmap_get_state(mdsc->mdsmap, req->r_resend_mds) > 0)) {
		dout("choose_mds using resend_mds mds%d\n",
		     req->r_resend_mds);
		return req->r_resend_mds;
	}

	if (mode == USE_RANDOM_MDS)
		goto random;

	inode = NULL;
	if (req->r_inode) {
		inode = req->r_inode;
	} else if (req->r_dentry) {
655 656
		struct inode *dir = req->r_dentry->d_parent->d_inode;

657
		if (dir->i_sb != mdsc->fsc->sb) {
658 659 660 661 662 663 664 665 666 667 668
			/* not this fs! */
			inode = req->r_dentry->d_inode;
		} else if (ceph_snap(dir) != CEPH_NOSNAP) {
			/* direct snapped/virtual snapdir requests
			 * based on parent dir inode */
			struct dentry *dn =
				get_nonsnap_parent(req->r_dentry->d_parent);
			inode = dn->d_inode;
			dout("__choose_mds using nonsnap parent %p\n", inode);
		} else if (req->r_dentry->d_inode) {
			/* dentry target */
S
Sage Weil 已提交
669 670
			inode = req->r_dentry->d_inode;
		} else {
671 672
			/* dir + name */
			inode = dir;
673
			hash = ceph_dentry_hash(req->r_dentry);
S
Sage Weil 已提交
674 675 676
			is_hash = true;
		}
	}
677

S
Sage Weil 已提交
678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699
	dout("__choose_mds %p is_hash=%d (%d) mode %d\n", inode, (int)is_hash,
	     (int)hash, mode);
	if (!inode)
		goto random;
	ci = ceph_inode(inode);

	if (is_hash && S_ISDIR(inode->i_mode)) {
		struct ceph_inode_frag frag;
		int found;

		ceph_choose_frag(ci, hash, &frag, &found);
		if (found) {
			if (mode == USE_ANY_MDS && frag.ndist > 0) {
				u8 r;

				/* choose a random replica */
				get_random_bytes(&r, 1);
				r %= frag.ndist;
				mds = frag.dist[r];
				dout("choose_mds %p %llx.%llx "
				     "frag %u mds%d (%d/%d)\n",
				     inode, ceph_vinop(inode),
700
				     frag.frag, mds,
S
Sage Weil 已提交
701
				     (int)r, frag.ndist);
702 703 704
				if (ceph_mdsmap_get_state(mdsc->mdsmap, mds) >=
				    CEPH_MDS_STATE_ACTIVE)
					return mds;
S
Sage Weil 已提交
705 706 707 708 709 710 711 712 713 714 715 716
			}

			/* since this file/dir wasn't known to be
			 * replicated, then we want to look for the
			 * authoritative mds. */
			mode = USE_AUTH_MDS;
			if (frag.mds >= 0) {
				/* choose auth mds */
				mds = frag.mds;
				dout("choose_mds %p %llx.%llx "
				     "frag %u mds%d (auth)\n",
				     inode, ceph_vinop(inode), frag.frag, mds);
717 718 719
				if (ceph_mdsmap_get_state(mdsc->mdsmap, mds) >=
				    CEPH_MDS_STATE_ACTIVE)
					return mds;
S
Sage Weil 已提交
720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755
			}
		}
	}

	spin_lock(&inode->i_lock);
	cap = NULL;
	if (mode == USE_AUTH_MDS)
		cap = ci->i_auth_cap;
	if (!cap && !RB_EMPTY_ROOT(&ci->i_caps))
		cap = rb_entry(rb_first(&ci->i_caps), struct ceph_cap, ci_node);
	if (!cap) {
		spin_unlock(&inode->i_lock);
		goto random;
	}
	mds = cap->session->s_mds;
	dout("choose_mds %p %llx.%llx mds%d (%scap %p)\n",
	     inode, ceph_vinop(inode), mds,
	     cap == ci->i_auth_cap ? "auth " : "", cap);
	spin_unlock(&inode->i_lock);
	return mds;

random:
	mds = ceph_mdsmap_get_random_mds(mdsc->mdsmap);
	dout("choose_mds chose random mds%d\n", mds);
	return mds;
}


/*
 * session messages
 */
static struct ceph_msg *create_session_msg(u32 op, u64 seq)
{
	struct ceph_msg *msg;
	struct ceph_mds_session_head *h;

756
	msg = ceph_msg_new(CEPH_MSG_CLIENT_SESSION, sizeof(*h), GFP_NOFS);
757
	if (!msg) {
S
Sage Weil 已提交
758
		pr_err("create_session_msg ENOMEM creating msg\n");
759
		return NULL;
S
Sage Weil 已提交
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
	}
	h = msg->front.iov_base;
	h->op = cpu_to_le32(op);
	h->seq = cpu_to_le64(seq);
	return msg;
}

/*
 * send session open request.
 *
 * called under mdsc->mutex
 */
static int __open_session(struct ceph_mds_client *mdsc,
			  struct ceph_mds_session *session)
{
	struct ceph_msg *msg;
	int mstate;
	int mds = session->s_mds;

	/* wait for mds to go active? */
	mstate = ceph_mdsmap_get_state(mdsc->mdsmap, mds);
	dout("open_session to mds%d (%s)\n", mds,
	     ceph_mds_state_name(mstate));
	session->s_state = CEPH_MDS_SESSION_OPENING;
	session->s_renew_requested = jiffies;

	/* send connect message */
	msg = create_session_msg(CEPH_SESSION_REQUEST_OPEN, session->s_seq);
788 789
	if (!msg)
		return -ENOMEM;
S
Sage Weil 已提交
790 791 792 793
	ceph_con_send(&session->s_con, msg);
	return 0;
}

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
/*
 * open sessions for any export targets for the given mds
 *
 * called under mdsc->mutex
 */
static void __open_export_target_sessions(struct ceph_mds_client *mdsc,
					  struct ceph_mds_session *session)
{
	struct ceph_mds_info *mi;
	struct ceph_mds_session *ts;
	int i, mds = session->s_mds;
	int target;

	if (mds >= mdsc->mdsmap->m_max_mds)
		return;
	mi = &mdsc->mdsmap->m_info[mds];
	dout("open_export_target_sessions for mds%d (%d targets)\n",
	     session->s_mds, mi->num_export_targets);

	for (i = 0; i < mi->num_export_targets; i++) {
		target = mi->export_targets[i];
		ts = __ceph_lookup_mds_session(mdsc, target);
		if (!ts) {
			ts = register_session(mdsc, target);
			if (IS_ERR(ts))
				return;
		}
		if (session->s_state == CEPH_MDS_SESSION_NEW ||
		    session->s_state == CEPH_MDS_SESSION_CLOSING)
			__open_session(mdsc, session);
		else
			dout(" mds%d target mds%d %p is %s\n", session->s_mds,
			     i, ts, session_state_name(ts->s_state));
		ceph_put_mds_session(ts);
	}
}

831 832 833 834 835 836 837 838
void ceph_mdsc_open_export_target_sessions(struct ceph_mds_client *mdsc,
					   struct ceph_mds_session *session)
{
	mutex_lock(&mdsc->mutex);
	__open_export_target_sessions(mdsc, session);
	mutex_unlock(&mdsc->mutex);
}

S
Sage Weil 已提交
839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866
/*
 * session caps
 */

/*
 * Free preallocated cap messages assigned to this session
 */
static void cleanup_cap_releases(struct ceph_mds_session *session)
{
	struct ceph_msg *msg;

	spin_lock(&session->s_cap_lock);
	while (!list_empty(&session->s_cap_releases)) {
		msg = list_first_entry(&session->s_cap_releases,
				       struct ceph_msg, list_head);
		list_del_init(&msg->list_head);
		ceph_msg_put(msg);
	}
	while (!list_empty(&session->s_cap_releases_done)) {
		msg = list_first_entry(&session->s_cap_releases_done,
				       struct ceph_msg, list_head);
		list_del_init(&msg->list_head);
		ceph_msg_put(msg);
	}
	spin_unlock(&session->s_cap_lock);
}

/*
S
Sage Weil 已提交
867 868
 * Helper to safely iterate over all caps associated with a session, with
 * special care taken to handle a racing __ceph_remove_cap().
S
Sage Weil 已提交
869
 *
S
Sage Weil 已提交
870
 * Caller must hold session s_mutex.
S
Sage Weil 已提交
871 872 873 874 875
 */
static int iterate_session_caps(struct ceph_mds_session *session,
				 int (*cb)(struct inode *, struct ceph_cap *,
					    void *), void *arg)
{
S
Sage Weil 已提交
876 877 878 879
	struct list_head *p;
	struct ceph_cap *cap;
	struct inode *inode, *last_inode = NULL;
	struct ceph_cap *old_cap = NULL;
S
Sage Weil 已提交
880 881 882 883
	int ret;

	dout("iterate_session_caps %p mds%d\n", session, session->s_mds);
	spin_lock(&session->s_cap_lock);
S
Sage Weil 已提交
884 885 886
	p = session->s_caps.next;
	while (p != &session->s_caps) {
		cap = list_entry(p, struct ceph_cap, session_caps);
S
Sage Weil 已提交
887
		inode = igrab(&cap->ci->vfs_inode);
S
Sage Weil 已提交
888 889
		if (!inode) {
			p = p->next;
S
Sage Weil 已提交
890
			continue;
S
Sage Weil 已提交
891 892
		}
		session->s_cap_iterator = cap;
S
Sage Weil 已提交
893
		spin_unlock(&session->s_cap_lock);
S
Sage Weil 已提交
894 895 896 897 898 899

		if (last_inode) {
			iput(last_inode);
			last_inode = NULL;
		}
		if (old_cap) {
900
			ceph_put_cap(session->s_mdsc, old_cap);
S
Sage Weil 已提交
901 902 903
			old_cap = NULL;
		}

S
Sage Weil 已提交
904
		ret = cb(inode, cap, arg);
S
Sage Weil 已提交
905 906
		last_inode = inode;

S
Sage Weil 已提交
907
		spin_lock(&session->s_cap_lock);
S
Sage Weil 已提交
908 909 910 911 912 913 914 915 916 917
		p = p->next;
		if (cap->ci == NULL) {
			dout("iterate_session_caps  finishing cap %p removal\n",
			     cap);
			BUG_ON(cap->session != session);
			list_del_init(&cap->session_caps);
			session->s_nr_caps--;
			cap->session = NULL;
			old_cap = cap;  /* put_cap it w/o locks held */
		}
918 919
		if (ret < 0)
			goto out;
S
Sage Weil 已提交
920
	}
921 922
	ret = 0;
out:
S
Sage Weil 已提交
923
	session->s_cap_iterator = NULL;
S
Sage Weil 已提交
924
	spin_unlock(&session->s_cap_lock);
S
Sage Weil 已提交
925 926 927 928

	if (last_inode)
		iput(last_inode);
	if (old_cap)
929
		ceph_put_cap(session->s_mdsc, old_cap);
S
Sage Weil 已提交
930

931
	return ret;
S
Sage Weil 已提交
932 933 934
}

static int remove_session_caps_cb(struct inode *inode, struct ceph_cap *cap,
935
				  void *arg)
S
Sage Weil 已提交
936 937
{
	struct ceph_inode_info *ci = ceph_inode(inode);
938 939
	int drop = 0;

S
Sage Weil 已提交
940 941
	dout("removing cap %p, ci is %p, inode is %p\n",
	     cap, ci, &ci->vfs_inode);
942 943 944 945
	spin_lock(&inode->i_lock);
	__ceph_remove_cap(cap);
	if (!__ceph_is_any_real_caps(ci)) {
		struct ceph_mds_client *mdsc =
946
			ceph_sb_to_client(inode->i_sb)->mdsc;
947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977

		spin_lock(&mdsc->cap_dirty_lock);
		if (!list_empty(&ci->i_dirty_item)) {
			pr_info(" dropping dirty %s state for %p %lld\n",
				ceph_cap_string(ci->i_dirty_caps),
				inode, ceph_ino(inode));
			ci->i_dirty_caps = 0;
			list_del_init(&ci->i_dirty_item);
			drop = 1;
		}
		if (!list_empty(&ci->i_flushing_item)) {
			pr_info(" dropping dirty+flushing %s state for %p %lld\n",
				ceph_cap_string(ci->i_flushing_caps),
				inode, ceph_ino(inode));
			ci->i_flushing_caps = 0;
			list_del_init(&ci->i_flushing_item);
			mdsc->num_cap_flushing--;
			drop = 1;
		}
		if (drop && ci->i_wrbuffer_ref) {
			pr_info(" dropping dirty data for %p %lld\n",
				inode, ceph_ino(inode));
			ci->i_wrbuffer_ref = 0;
			ci->i_wrbuffer_ref_head = 0;
			drop++;
		}
		spin_unlock(&mdsc->cap_dirty_lock);
	}
	spin_unlock(&inode->i_lock);
	while (drop--)
		iput(inode);
S
Sage Weil 已提交
978 979 980 981 982 983 984 985 986 987 988
	return 0;
}

/*
 * caller must hold session s_mutex
 */
static void remove_session_caps(struct ceph_mds_session *session)
{
	dout("remove_session_caps on %p\n", session);
	iterate_session_caps(session, remove_session_caps_cb, NULL);
	BUG_ON(session->s_nr_caps > 0);
989
	BUG_ON(!list_empty(&session->s_cap_flushing));
S
Sage Weil 已提交
990 991 992 993 994 995 996 997 998 999 1000 1001
	cleanup_cap_releases(session);
}

/*
 * wake up any threads waiting on this session's caps.  if the cap is
 * old (didn't get renewed on the client reconnect), remove it now.
 *
 * caller must hold s_mutex.
 */
static int wake_up_session_cb(struct inode *inode, struct ceph_cap *cap,
			      void *arg)
{
1002 1003
	struct ceph_inode_info *ci = ceph_inode(inode);

1004
	wake_up_all(&ci->i_cap_wq);
1005 1006 1007 1008 1009 1010
	if (arg) {
		spin_lock(&inode->i_lock);
		ci->i_wanted_max_size = 0;
		ci->i_requested_max_size = 0;
		spin_unlock(&inode->i_lock);
	}
S
Sage Weil 已提交
1011 1012 1013
	return 0;
}

1014 1015
static void wake_up_session_caps(struct ceph_mds_session *session,
				 int reconnect)
S
Sage Weil 已提交
1016 1017
{
	dout("wake_up_session_caps %p mds%d\n", session, session->s_mds);
1018 1019
	iterate_session_caps(session, wake_up_session_cb,
			     (void *)(unsigned long)reconnect);
S
Sage Weil 已提交
1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036
}

/*
 * Send periodic message to MDS renewing all currently held caps.  The
 * ack will reset the expiration for all caps from this session.
 *
 * caller holds s_mutex
 */
static int send_renew_caps(struct ceph_mds_client *mdsc,
			   struct ceph_mds_session *session)
{
	struct ceph_msg *msg;
	int state;

	if (time_after_eq(jiffies, session->s_cap_ttl) &&
	    time_after_eq(session->s_cap_ttl, session->s_renew_requested))
		pr_info("mds%d caps stale\n", session->s_mds);
1037
	session->s_renew_requested = jiffies;
S
Sage Weil 已提交
1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051

	/* do not try to renew caps until a recovering mds has reconnected
	 * with its clients. */
	state = ceph_mdsmap_get_state(mdsc->mdsmap, session->s_mds);
	if (state < CEPH_MDS_STATE_RECONNECT) {
		dout("send_renew_caps ignoring mds%d (%s)\n",
		     session->s_mds, ceph_mds_state_name(state));
		return 0;
	}

	dout("send_renew_caps to mds%d (%s)\n", session->s_mds,
		ceph_mds_state_name(state));
	msg = create_session_msg(CEPH_SESSION_REQUEST_RENEWCAPS,
				 ++session->s_renew_seq);
1052 1053
	if (!msg)
		return -ENOMEM;
S
Sage Weil 已提交
1054 1055 1056 1057 1058 1059
	ceph_con_send(&session->s_con, msg);
	return 0;
}

/*
 * Note new cap ttl, and any transition from stale -> not stale (fresh?).
1060 1061
 *
 * Called under session->s_mutex
S
Sage Weil 已提交
1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089
 */
static void renewed_caps(struct ceph_mds_client *mdsc,
			 struct ceph_mds_session *session, int is_renew)
{
	int was_stale;
	int wake = 0;

	spin_lock(&session->s_cap_lock);
	was_stale = is_renew && (session->s_cap_ttl == 0 ||
				 time_after_eq(jiffies, session->s_cap_ttl));

	session->s_cap_ttl = session->s_renew_requested +
		mdsc->mdsmap->m_session_timeout*HZ;

	if (was_stale) {
		if (time_before(jiffies, session->s_cap_ttl)) {
			pr_info("mds%d caps renewed\n", session->s_mds);
			wake = 1;
		} else {
			pr_info("mds%d caps still stale\n", session->s_mds);
		}
	}
	dout("renewed_caps mds%d ttl now %lu, was %s, now %s\n",
	     session->s_mds, session->s_cap_ttl, was_stale ? "stale" : "fresh",
	     time_before(jiffies, session->s_cap_ttl) ? "stale" : "fresh");
	spin_unlock(&session->s_cap_lock);

	if (wake)
1090
		wake_up_session_caps(session, 0);
S
Sage Weil 已提交
1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104
}

/*
 * send a session close request
 */
static int request_close_session(struct ceph_mds_client *mdsc,
				 struct ceph_mds_session *session)
{
	struct ceph_msg *msg;

	dout("request_close_session mds%d state %s seq %lld\n",
	     session->s_mds, session_state_name(session->s_state),
	     session->s_seq);
	msg = create_session_msg(CEPH_SESSION_REQUEST_CLOSE, session->s_seq);
1105 1106 1107 1108
	if (!msg)
		return -ENOMEM;
	ceph_con_send(&session->s_con, msg);
	return 0;
S
Sage Weil 已提交
1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157
}

/*
 * Called with s_mutex held.
 */
static int __close_session(struct ceph_mds_client *mdsc,
			 struct ceph_mds_session *session)
{
	if (session->s_state >= CEPH_MDS_SESSION_CLOSING)
		return 0;
	session->s_state = CEPH_MDS_SESSION_CLOSING;
	return request_close_session(mdsc, session);
}

/*
 * Trim old(er) caps.
 *
 * Because we can't cache an inode without one or more caps, we do
 * this indirectly: if a cap is unused, we prune its aliases, at which
 * point the inode will hopefully get dropped to.
 *
 * Yes, this is a bit sloppy.  Our only real goal here is to respond to
 * memory pressure from the MDS, though, so it needn't be perfect.
 */
static int trim_caps_cb(struct inode *inode, struct ceph_cap *cap, void *arg)
{
	struct ceph_mds_session *session = arg;
	struct ceph_inode_info *ci = ceph_inode(inode);
	int used, oissued, mine;

	if (session->s_trim_caps <= 0)
		return -1;

	spin_lock(&inode->i_lock);
	mine = cap->issued | cap->implemented;
	used = __ceph_caps_used(ci);
	oissued = __ceph_caps_issued_other(ci, cap);

	dout("trim_caps_cb %p cap %p mine %s oissued %s used %s\n",
	     inode, cap, ceph_cap_string(mine), ceph_cap_string(oissued),
	     ceph_cap_string(used));
	if (ci->i_dirty_caps)
		goto out;   /* dirty caps */
	if ((used & ~oissued) & mine)
		goto out;   /* we need these caps */

	session->s_trim_caps--;
	if (oissued) {
		/* we aren't the only cap.. just remove us */
S
Sage Weil 已提交
1158
		__ceph_remove_cap(cap);
S
Sage Weil 已提交
1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189
	} else {
		/* try to drop referring dentries */
		spin_unlock(&inode->i_lock);
		d_prune_aliases(inode);
		dout("trim_caps_cb %p cap %p  pruned, count now %d\n",
		     inode, cap, atomic_read(&inode->i_count));
		return 0;
	}

out:
	spin_unlock(&inode->i_lock);
	return 0;
}

/*
 * Trim session cap count down to some max number.
 */
static int trim_caps(struct ceph_mds_client *mdsc,
		     struct ceph_mds_session *session,
		     int max_caps)
{
	int trim_caps = session->s_nr_caps - max_caps;

	dout("trim_caps mds%d start: %d / %d, trim %d\n",
	     session->s_mds, session->s_nr_caps, max_caps, trim_caps);
	if (trim_caps > 0) {
		session->s_trim_caps = trim_caps;
		iterate_session_caps(session, trim_caps_cb, session);
		dout("trim_caps mds%d done: %d / %d, trimmed %d\n",
		     session->s_mds, session->s_nr_caps, max_caps,
			trim_caps - session->s_trim_caps);
1190
		session->s_trim_caps = 0;
S
Sage Weil 已提交
1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201
	}
	return 0;
}

/*
 * Allocate cap_release messages.  If there is a partially full message
 * in the queue, try to allocate enough to cover it's remainder, so that
 * we can send it immediately.
 *
 * Called under s_mutex.
 */
1202
int ceph_add_cap_releases(struct ceph_mds_client *mdsc,
S
Sage Weil 已提交
1203
			  struct ceph_mds_session *session)
S
Sage Weil 已提交
1204
{
S
Sage Weil 已提交
1205
	struct ceph_msg *msg, *partial = NULL;
S
Sage Weil 已提交
1206 1207
	struct ceph_mds_cap_release *head;
	int err = -ENOMEM;
1208
	int extra = mdsc->fsc->mount_options->cap_release_safety;
S
Sage Weil 已提交
1209
	int num;
S
Sage Weil 已提交
1210

S
Sage Weil 已提交
1211 1212
	dout("add_cap_releases %p mds%d extra %d\n", session, session->s_mds,
	     extra);
S
Sage Weil 已提交
1213 1214 1215 1216 1217 1218 1219 1220

	spin_lock(&session->s_cap_lock);

	if (!list_empty(&session->s_cap_releases)) {
		msg = list_first_entry(&session->s_cap_releases,
				       struct ceph_msg,
				 list_head);
		head = msg->front.iov_base;
S
Sage Weil 已提交
1221 1222 1223 1224 1225 1226 1227
		num = le32_to_cpu(head->num);
		if (num) {
			dout(" partial %p with (%d/%d)\n", msg, num,
			     (int)CEPH_CAPS_PER_RELEASE);
			extra += CEPH_CAPS_PER_RELEASE - num;
			partial = msg;
		}
S
Sage Weil 已提交
1228 1229 1230
	}
	while (session->s_num_cap_releases < session->s_nr_caps + extra) {
		spin_unlock(&session->s_cap_lock);
1231 1232
		msg = ceph_msg_new(CEPH_MSG_CLIENT_CAPRELEASE, PAGE_CACHE_SIZE,
				   GFP_NOFS);
S
Sage Weil 已提交
1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244
		if (!msg)
			goto out_unlocked;
		dout("add_cap_releases %p msg %p now %d\n", session, msg,
		     (int)msg->front.iov_len);
		head = msg->front.iov_base;
		head->num = cpu_to_le32(0);
		msg->front.iov_len = sizeof(*head);
		spin_lock(&session->s_cap_lock);
		list_add(&msg->list_head, &session->s_cap_releases);
		session->s_num_cap_releases += CEPH_CAPS_PER_RELEASE;
	}

S
Sage Weil 已提交
1245 1246 1247 1248 1249 1250 1251 1252
	if (partial) {
		head = partial->front.iov_base;
		num = le32_to_cpu(head->num);
		dout(" queueing partial %p with %d/%d\n", partial, num,
		     (int)CEPH_CAPS_PER_RELEASE);
		list_move_tail(&partial->list_head,
			       &session->s_cap_releases_done);
		session->s_num_cap_releases -= CEPH_CAPS_PER_RELEASE - num;
S
Sage Weil 已提交
1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312
	}
	err = 0;
	spin_unlock(&session->s_cap_lock);
out_unlocked:
	return err;
}

/*
 * flush all dirty inode data to disk.
 *
 * returns true if we've flushed through want_flush_seq
 */
static int check_cap_flush(struct ceph_mds_client *mdsc, u64 want_flush_seq)
{
	int mds, ret = 1;

	dout("check_cap_flush want %lld\n", want_flush_seq);
	mutex_lock(&mdsc->mutex);
	for (mds = 0; ret && mds < mdsc->max_sessions; mds++) {
		struct ceph_mds_session *session = mdsc->sessions[mds];

		if (!session)
			continue;
		get_session(session);
		mutex_unlock(&mdsc->mutex);

		mutex_lock(&session->s_mutex);
		if (!list_empty(&session->s_cap_flushing)) {
			struct ceph_inode_info *ci =
				list_entry(session->s_cap_flushing.next,
					   struct ceph_inode_info,
					   i_flushing_item);
			struct inode *inode = &ci->vfs_inode;

			spin_lock(&inode->i_lock);
			if (ci->i_cap_flush_seq <= want_flush_seq) {
				dout("check_cap_flush still flushing %p "
				     "seq %lld <= %lld to mds%d\n", inode,
				     ci->i_cap_flush_seq, want_flush_seq,
				     session->s_mds);
				ret = 0;
			}
			spin_unlock(&inode->i_lock);
		}
		mutex_unlock(&session->s_mutex);
		ceph_put_mds_session(session);

		if (!ret)
			return ret;
		mutex_lock(&mdsc->mutex);
	}

	mutex_unlock(&mdsc->mutex);
	dout("check_cap_flush ok, flushed thru %lld\n", want_flush_seq);
	return ret;
}

/*
 * called under s_mutex
 */
1313 1314
void ceph_send_cap_releases(struct ceph_mds_client *mdsc,
			    struct ceph_mds_session *session)
S
Sage Weil 已提交
1315 1316 1317 1318
{
	struct ceph_msg *msg;

	dout("send_cap_releases mds%d\n", session->s_mds);
1319 1320
	spin_lock(&session->s_cap_lock);
	while (!list_empty(&session->s_cap_releases_done)) {
S
Sage Weil 已提交
1321 1322 1323 1324 1325 1326 1327
		msg = list_first_entry(&session->s_cap_releases_done,
				 struct ceph_msg, list_head);
		list_del_init(&msg->list_head);
		spin_unlock(&session->s_cap_lock);
		msg->hdr.front_len = cpu_to_le32(msg->front.iov_len);
		dout("send_cap_releases mds%d %p\n", session->s_mds, msg);
		ceph_con_send(&session->s_con, msg);
1328
		spin_lock(&session->s_cap_lock);
S
Sage Weil 已提交
1329 1330 1331 1332
	}
	spin_unlock(&session->s_cap_lock);
}

1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370
static void discard_cap_releases(struct ceph_mds_client *mdsc,
				 struct ceph_mds_session *session)
{
	struct ceph_msg *msg;
	struct ceph_mds_cap_release *head;
	unsigned num;

	dout("discard_cap_releases mds%d\n", session->s_mds);
	spin_lock(&session->s_cap_lock);

	/* zero out the in-progress message */
	msg = list_first_entry(&session->s_cap_releases,
			       struct ceph_msg, list_head);
	head = msg->front.iov_base;
	num = le32_to_cpu(head->num);
	dout("discard_cap_releases mds%d %p %u\n", session->s_mds, msg, num);
	head->num = cpu_to_le32(0);
	session->s_num_cap_releases += num;

	/* requeue completed messages */
	while (!list_empty(&session->s_cap_releases_done)) {
		msg = list_first_entry(&session->s_cap_releases_done,
				 struct ceph_msg, list_head);
		list_del_init(&msg->list_head);

		head = msg->front.iov_base;
		num = le32_to_cpu(head->num);
		dout("discard_cap_releases mds%d %p %u\n", session->s_mds, msg,
		     num);
		session->s_num_cap_releases += num;
		head->num = cpu_to_le32(0);
		msg->front.iov_len = sizeof(*head);
		list_add(&msg->list_head, &session->s_cap_releases);
	}

	spin_unlock(&session->s_cap_lock);
}

S
Sage Weil 已提交
1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385
/*
 * requests
 */

/*
 * Create an mds request.
 */
struct ceph_mds_request *
ceph_mdsc_create_request(struct ceph_mds_client *mdsc, int op, int mode)
{
	struct ceph_mds_request *req = kzalloc(sizeof(*req), GFP_NOFS);

	if (!req)
		return ERR_PTR(-ENOMEM);

1386
	mutex_init(&req->r_fill_mutex);
1387
	req->r_mdsc = mdsc;
S
Sage Weil 已提交
1388 1389 1390 1391
	req->r_started = jiffies;
	req->r_resend_mds = -1;
	INIT_LIST_HEAD(&req->r_unsafe_dir_item);
	req->r_fmode = -1;
1392
	kref_init(&req->r_kref);
S
Sage Weil 已提交
1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403
	INIT_LIST_HEAD(&req->r_wait);
	init_completion(&req->r_completion);
	init_completion(&req->r_safe_completion);
	INIT_LIST_HEAD(&req->r_unsafe_item);

	req->r_op = op;
	req->r_direct_mode = mode;
	return req;
}

/*
S
Sage Weil 已提交
1404
 * return oldest (lowest) request, tid in request tree, 0 if none.
S
Sage Weil 已提交
1405 1406 1407
 *
 * called under mdsc->mutex.
 */
S
Sage Weil 已提交
1408 1409 1410 1411 1412 1413 1414 1415
static struct ceph_mds_request *__get_oldest_req(struct ceph_mds_client *mdsc)
{
	if (RB_EMPTY_ROOT(&mdsc->request_tree))
		return NULL;
	return rb_entry(rb_first(&mdsc->request_tree),
			struct ceph_mds_request, r_node);
}

S
Sage Weil 已提交
1416 1417
static u64 __get_oldest_tid(struct ceph_mds_client *mdsc)
{
S
Sage Weil 已提交
1418 1419 1420 1421 1422
	struct ceph_mds_request *req = __get_oldest_req(mdsc);

	if (req)
		return req->r_tid;
	return 0;
S
Sage Weil 已提交
1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457
}

/*
 * Build a dentry's path.  Allocate on heap; caller must kfree.  Based
 * on build_path_from_dentry in fs/cifs/dir.c.
 *
 * If @stop_on_nosnap, generate path relative to the first non-snapped
 * inode.
 *
 * Encode hidden .snap dirs as a double /, i.e.
 *   foo/.snap/bar -> foo//bar
 */
char *ceph_mdsc_build_path(struct dentry *dentry, int *plen, u64 *base,
			   int stop_on_nosnap)
{
	struct dentry *temp;
	char *path;
	int len, pos;

	if (dentry == NULL)
		return ERR_PTR(-EINVAL);

retry:
	len = 0;
	for (temp = dentry; !IS_ROOT(temp);) {
		struct inode *inode = temp->d_inode;
		if (inode && ceph_snap(inode) == CEPH_SNAPDIR)
			len++;  /* slash only */
		else if (stop_on_nosnap && inode &&
			 ceph_snap(inode) == CEPH_NOSNAP)
			break;
		else
			len += 1 + temp->d_name.len;
		temp = temp->d_parent;
		if (temp == NULL) {
1458
			pr_err("build_path corrupt dentry %p\n", dentry);
S
Sage Weil 已提交
1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473
			return ERR_PTR(-EINVAL);
		}
	}
	if (len)
		len--;  /* no leading '/' */

	path = kmalloc(len+1, GFP_NOFS);
	if (path == NULL)
		return ERR_PTR(-ENOMEM);
	pos = len;
	path[pos] = 0;	/* trailing null */
	for (temp = dentry; !IS_ROOT(temp) && pos != 0; ) {
		struct inode *inode = temp->d_inode;

		if (inode && ceph_snap(inode) == CEPH_SNAPDIR) {
S
Sage Weil 已提交
1474
			dout("build_path path+%d: %p SNAPDIR\n",
S
Sage Weil 已提交
1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489
			     pos, temp);
		} else if (stop_on_nosnap && inode &&
			   ceph_snap(inode) == CEPH_NOSNAP) {
			break;
		} else {
			pos -= temp->d_name.len;
			if (pos < 0)
				break;
			strncpy(path + pos, temp->d_name.name,
				temp->d_name.len);
		}
		if (pos)
			path[--pos] = '/';
		temp = temp->d_parent;
		if (temp == NULL) {
S
Sage Weil 已提交
1490
			pr_err("build_path corrupt dentry\n");
S
Sage Weil 已提交
1491 1492 1493 1494 1495
			kfree(path);
			return ERR_PTR(-EINVAL);
		}
	}
	if (pos != 0) {
S
Sage Weil 已提交
1496
		pr_err("build_path did not end path lookup where "
S
Sage Weil 已提交
1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507
		       "expected, namelen is %d, pos is %d\n", len, pos);
		/* presumably this is only possible if racing with a
		   rename of one of the parent directories (we can not
		   lock the dentries above us to prevent this, but
		   retrying should be harmless) */
		kfree(path);
		goto retry;
	}

	*base = ceph_ino(temp->d_inode);
	*plen = len;
S
Sage Weil 已提交
1508
	dout("build_path on %p %d built %llx '%.*s'\n",
N
Nick Piggin 已提交
1509
	     dentry, dentry->d_count, *base, len, path);
S
Sage Weil 已提交
1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619
	return path;
}

static int build_dentry_path(struct dentry *dentry,
			     const char **ppath, int *ppathlen, u64 *pino,
			     int *pfreepath)
{
	char *path;

	if (ceph_snap(dentry->d_parent->d_inode) == CEPH_NOSNAP) {
		*pino = ceph_ino(dentry->d_parent->d_inode);
		*ppath = dentry->d_name.name;
		*ppathlen = dentry->d_name.len;
		return 0;
	}
	path = ceph_mdsc_build_path(dentry, ppathlen, pino, 1);
	if (IS_ERR(path))
		return PTR_ERR(path);
	*ppath = path;
	*pfreepath = 1;
	return 0;
}

static int build_inode_path(struct inode *inode,
			    const char **ppath, int *ppathlen, u64 *pino,
			    int *pfreepath)
{
	struct dentry *dentry;
	char *path;

	if (ceph_snap(inode) == CEPH_NOSNAP) {
		*pino = ceph_ino(inode);
		*ppathlen = 0;
		return 0;
	}
	dentry = d_find_alias(inode);
	path = ceph_mdsc_build_path(dentry, ppathlen, pino, 1);
	dput(dentry);
	if (IS_ERR(path))
		return PTR_ERR(path);
	*ppath = path;
	*pfreepath = 1;
	return 0;
}

/*
 * request arguments may be specified via an inode *, a dentry *, or
 * an explicit ino+path.
 */
static int set_request_path_attr(struct inode *rinode, struct dentry *rdentry,
				  const char *rpath, u64 rino,
				  const char **ppath, int *pathlen,
				  u64 *ino, int *freepath)
{
	int r = 0;

	if (rinode) {
		r = build_inode_path(rinode, ppath, pathlen, ino, freepath);
		dout(" inode %p %llx.%llx\n", rinode, ceph_ino(rinode),
		     ceph_snap(rinode));
	} else if (rdentry) {
		r = build_dentry_path(rdentry, ppath, pathlen, ino, freepath);
		dout(" dentry %p %llx/%.*s\n", rdentry, *ino, *pathlen,
		     *ppath);
	} else if (rpath) {
		*ino = rino;
		*ppath = rpath;
		*pathlen = strlen(rpath);
		dout(" path %.*s\n", *pathlen, rpath);
	}

	return r;
}

/*
 * called under mdsc->mutex
 */
static struct ceph_msg *create_request_message(struct ceph_mds_client *mdsc,
					       struct ceph_mds_request *req,
					       int mds)
{
	struct ceph_msg *msg;
	struct ceph_mds_request_head *head;
	const char *path1 = NULL;
	const char *path2 = NULL;
	u64 ino1 = 0, ino2 = 0;
	int pathlen1 = 0, pathlen2 = 0;
	int freepath1 = 0, freepath2 = 0;
	int len;
	u16 releases;
	void *p, *end;
	int ret;

	ret = set_request_path_attr(req->r_inode, req->r_dentry,
			      req->r_path1, req->r_ino1.ino,
			      &path1, &pathlen1, &ino1, &freepath1);
	if (ret < 0) {
		msg = ERR_PTR(ret);
		goto out;
	}

	ret = set_request_path_attr(NULL, req->r_old_dentry,
			      req->r_path2, req->r_ino2.ino,
			      &path2, &pathlen2, &ino2, &freepath2);
	if (ret < 0) {
		msg = ERR_PTR(ret);
		goto out_free1;
	}

	len = sizeof(*head) +
1620
		pathlen1 + pathlen2 + 2*(1 + sizeof(u32) + sizeof(u64));
S
Sage Weil 已提交
1621 1622 1623 1624 1625 1626 1627 1628 1629 1630

	/* calculate (max) length for cap releases */
	len += sizeof(struct ceph_mds_request_release) *
		(!!req->r_inode_drop + !!req->r_dentry_drop +
		 !!req->r_old_inode_drop + !!req->r_old_dentry_drop);
	if (req->r_dentry_drop)
		len += req->r_dentry->d_name.len;
	if (req->r_old_dentry_drop)
		len += req->r_old_dentry->d_name.len;

1631
	msg = ceph_msg_new(CEPH_MSG_CLIENT_REQUEST, len, GFP_NOFS);
1632 1633
	if (!msg) {
		msg = ERR_PTR(-ENOMEM);
S
Sage Weil 已提交
1634
		goto out_free2;
1635
	}
S
Sage Weil 已提交
1636

1637 1638
	msg->hdr.tid = cpu_to_le64(req->r_tid);

S
Sage Weil 已提交
1639 1640 1641 1642 1643 1644
	head = msg->front.iov_base;
	p = msg->front.iov_base + sizeof(*head);
	end = msg->front.iov_base + msg->front.iov_len;

	head->mdsmap_epoch = cpu_to_le32(mdsc->mdsmap->m_epoch);
	head->op = cpu_to_le32(req->r_op);
1645 1646
	head->caller_uid = cpu_to_le32(req->r_uid);
	head->caller_gid = cpu_to_le32(req->r_gid);
S
Sage Weil 已提交
1647 1648 1649 1650 1651
	head->args = req->r_args;

	ceph_encode_filepath(&p, end, ino1, path1);
	ceph_encode_filepath(&p, end, ino2, path2);

1652 1653 1654
	/* make note of release offset, in case we need to replay */
	req->r_request_release_offset = p - msg->front.iov_base;

S
Sage Weil 已提交
1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701
	/* cap releases */
	releases = 0;
	if (req->r_inode_drop)
		releases += ceph_encode_inode_release(&p,
		      req->r_inode ? req->r_inode : req->r_dentry->d_inode,
		      mds, req->r_inode_drop, req->r_inode_unless, 0);
	if (req->r_dentry_drop)
		releases += ceph_encode_dentry_release(&p, req->r_dentry,
		       mds, req->r_dentry_drop, req->r_dentry_unless);
	if (req->r_old_dentry_drop)
		releases += ceph_encode_dentry_release(&p, req->r_old_dentry,
		       mds, req->r_old_dentry_drop, req->r_old_dentry_unless);
	if (req->r_old_inode_drop)
		releases += ceph_encode_inode_release(&p,
		      req->r_old_dentry->d_inode,
		      mds, req->r_old_inode_drop, req->r_old_inode_unless, 0);
	head->num_releases = cpu_to_le16(releases);

	BUG_ON(p > end);
	msg->front.iov_len = p - msg->front.iov_base;
	msg->hdr.front_len = cpu_to_le32(msg->front.iov_len);

	msg->pages = req->r_pages;
	msg->nr_pages = req->r_num_pages;
	msg->hdr.data_len = cpu_to_le32(req->r_data_len);
	msg->hdr.data_off = cpu_to_le16(0);

out_free2:
	if (freepath2)
		kfree((char *)path2);
out_free1:
	if (freepath1)
		kfree((char *)path1);
out:
	return msg;
}

/*
 * called under mdsc->mutex if error, under no mutex if
 * success.
 */
static void complete_request(struct ceph_mds_client *mdsc,
			     struct ceph_mds_request *req)
{
	if (req->r_callback)
		req->r_callback(mdsc, req);
	else
1702
		complete_all(&req->r_completion);
S
Sage Weil 已提交
1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716
}

/*
 * called under mdsc->mutex
 */
static int __prepare_send_request(struct ceph_mds_client *mdsc,
				  struct ceph_mds_request *req,
				  int mds)
{
	struct ceph_mds_request_head *rhead;
	struct ceph_msg *msg;
	int flags = 0;

	req->r_attempts++;
1717 1718 1719 1720 1721 1722 1723 1724 1725
	if (req->r_inode) {
		struct ceph_cap *cap =
			ceph_get_cap_for_mds(ceph_inode(req->r_inode), mds);

		if (cap)
			req->r_sent_on_mseq = cap->mseq;
		else
			req->r_sent_on_mseq = -1;
	}
S
Sage Weil 已提交
1726 1727 1728
	dout("prepare_send_request %p tid %lld %s (attempt %d)\n", req,
	     req->r_tid, ceph_mds_op_name(req->r_op), req->r_attempts);

1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746
	if (req->r_got_unsafe) {
		/*
		 * Replay.  Do not regenerate message (and rebuild
		 * paths, etc.); just use the original message.
		 * Rebuilding paths will break for renames because
		 * d_move mangles the src name.
		 */
		msg = req->r_request;
		rhead = msg->front.iov_base;

		flags = le32_to_cpu(rhead->flags);
		flags |= CEPH_MDS_FLAG_REPLAY;
		rhead->flags = cpu_to_le32(flags);

		if (req->r_target_inode)
			rhead->ino = cpu_to_le64(ceph_ino(req->r_target_inode));

		rhead->num_retry = req->r_attempts - 1;
1747 1748 1749 1750 1751

		/* remove cap/dentry releases from message */
		rhead->num_releases = 0;
		msg->hdr.front_len = cpu_to_le32(req->r_request_release_offset);
		msg->front.iov_len = req->r_request_release_offset;
1752 1753 1754
		return 0;
	}

S
Sage Weil 已提交
1755 1756 1757 1758 1759 1760
	if (req->r_request) {
		ceph_msg_put(req->r_request);
		req->r_request = NULL;
	}
	msg = create_request_message(mdsc, req, mds);
	if (IS_ERR(msg)) {
1761
		req->r_err = PTR_ERR(msg);
S
Sage Weil 已提交
1762
		complete_request(mdsc, req);
1763
		return PTR_ERR(msg);
S
Sage Weil 已提交
1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775
	}
	req->r_request = msg;

	rhead = msg->front.iov_base;
	rhead->oldest_client_tid = cpu_to_le64(__get_oldest_tid(mdsc));
	if (req->r_got_unsafe)
		flags |= CEPH_MDS_FLAG_REPLAY;
	if (req->r_locked_dir)
		flags |= CEPH_MDS_FLAG_WANT_DENTRY;
	rhead->flags = cpu_to_le32(flags);
	rhead->num_fwd = req->r_num_fwd;
	rhead->num_retry = req->r_attempts - 1;
1776
	rhead->ino = 0;
S
Sage Weil 已提交
1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791

	dout(" r_locked_dir = %p\n", req->r_locked_dir);
	return 0;
}

/*
 * send request, or put it on the appropriate wait list.
 */
static int __do_request(struct ceph_mds_client *mdsc,
			struct ceph_mds_request *req)
{
	struct ceph_mds_session *session = NULL;
	int mds = -1;
	int err = -EAGAIN;

1792
	if (req->r_err || req->r_got_result)
S
Sage Weil 已提交
1793 1794 1795 1796 1797 1798 1799 1800 1801
		goto out;

	if (req->r_timeout &&
	    time_after_eq(jiffies, req->r_started + req->r_timeout)) {
		dout("do_request timed out\n");
		err = -EIO;
		goto finish;
	}

1802 1803
	put_request_session(req);

S
Sage Weil 已提交
1804 1805 1806 1807 1808 1809 1810 1811 1812 1813
	mds = __choose_mds(mdsc, req);
	if (mds < 0 ||
	    ceph_mdsmap_get_state(mdsc->mdsmap, mds) < CEPH_MDS_STATE_ACTIVE) {
		dout("do_request no mds or not active, waiting for map\n");
		list_add(&req->r_wait, &mdsc->waiting_for_map);
		goto out;
	}

	/* get, open session */
	session = __ceph_lookup_mds_session(mdsc, mds);
1814
	if (!session) {
S
Sage Weil 已提交
1815
		session = register_session(mdsc, mds);
1816 1817 1818 1819 1820
		if (IS_ERR(session)) {
			err = PTR_ERR(session);
			goto finish;
		}
	}
1821 1822
	req->r_session = get_session(session);

S
Sage Weil 已提交
1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851
	dout("do_request mds%d session %p state %s\n", mds, session,
	     session_state_name(session->s_state));
	if (session->s_state != CEPH_MDS_SESSION_OPEN &&
	    session->s_state != CEPH_MDS_SESSION_HUNG) {
		if (session->s_state == CEPH_MDS_SESSION_NEW ||
		    session->s_state == CEPH_MDS_SESSION_CLOSING)
			__open_session(mdsc, session);
		list_add(&req->r_wait, &session->s_waiting);
		goto out_session;
	}

	/* send request */
	req->r_resend_mds = -1;   /* forget any previous mds hint */

	if (req->r_request_started == 0)   /* note request start time */
		req->r_request_started = jiffies;

	err = __prepare_send_request(mdsc, req, mds);
	if (!err) {
		ceph_msg_get(req->r_request);
		ceph_con_send(&session->s_con, req->r_request);
	}

out_session:
	ceph_put_mds_session(session);
out:
	return err;

finish:
1852
	req->r_err = err;
S
Sage Weil 已提交
1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872
	complete_request(mdsc, req);
	goto out;
}

/*
 * called under mdsc->mutex
 */
static void __wake_requests(struct ceph_mds_client *mdsc,
			    struct list_head *head)
{
	struct ceph_mds_request *req, *nreq;

	list_for_each_entry_safe(req, nreq, head, r_wait) {
		list_del_init(&req->r_wait);
		__do_request(mdsc, req);
	}
}

/*
 * Wake up threads with requests pending for @mds, so that they can
1873
 * resubmit their requests to a possibly different mds.
S
Sage Weil 已提交
1874
 */
1875
static void kick_requests(struct ceph_mds_client *mdsc, int mds)
S
Sage Weil 已提交
1876
{
S
Sage Weil 已提交
1877 1878
	struct ceph_mds_request *req;
	struct rb_node *p;
S
Sage Weil 已提交
1879 1880

	dout("kick_requests mds%d\n", mds);
S
Sage Weil 已提交
1881 1882 1883 1884 1885 1886 1887 1888
	for (p = rb_first(&mdsc->request_tree); p; p = rb_next(p)) {
		req = rb_entry(p, struct ceph_mds_request, r_node);
		if (req->r_got_unsafe)
			continue;
		if (req->r_session &&
		    req->r_session->s_mds == mds) {
			dout(" kicking tid %llu\n", req->r_tid);
			__do_request(mdsc, req);
S
Sage Weil 已提交
1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929
		}
	}
}

void ceph_mdsc_submit_request(struct ceph_mds_client *mdsc,
			      struct ceph_mds_request *req)
{
	dout("submit_request on %p\n", req);
	mutex_lock(&mdsc->mutex);
	__register_request(mdsc, req, NULL);
	__do_request(mdsc, req);
	mutex_unlock(&mdsc->mutex);
}

/*
 * Synchrously perform an mds request.  Take care of all of the
 * session setup, forwarding, retry details.
 */
int ceph_mdsc_do_request(struct ceph_mds_client *mdsc,
			 struct inode *dir,
			 struct ceph_mds_request *req)
{
	int err;

	dout("do_request on %p\n", req);

	/* take CAP_PIN refs for r_inode, r_locked_dir, r_old_dentry */
	if (req->r_inode)
		ceph_get_cap_refs(ceph_inode(req->r_inode), CEPH_CAP_PIN);
	if (req->r_locked_dir)
		ceph_get_cap_refs(ceph_inode(req->r_locked_dir), CEPH_CAP_PIN);
	if (req->r_old_dentry)
		ceph_get_cap_refs(
			ceph_inode(req->r_old_dentry->d_parent->d_inode),
			CEPH_CAP_PIN);

	/* issue */
	mutex_lock(&mdsc->mutex);
	__register_request(mdsc, req, dir);
	__do_request(mdsc, req);

1930 1931 1932 1933 1934
	if (req->r_err) {
		err = req->r_err;
		__unregister_request(mdsc, req);
		dout("do_request early error %d\n", err);
		goto out;
S
Sage Weil 已提交
1935 1936
	}

1937 1938 1939 1940
	/* wait */
	mutex_unlock(&mdsc->mutex);
	dout("do_request waiting\n");
	if (req->r_timeout) {
1941
		err = (long)wait_for_completion_killable_timeout(
1942 1943 1944 1945
			&req->r_completion, req->r_timeout);
		if (err == 0)
			err = -EIO;
	} else {
1946
		err = wait_for_completion_killable(&req->r_completion);
1947 1948 1949
	}
	dout("do_request waited, got %d\n", err);
	mutex_lock(&mdsc->mutex);
1950

1951 1952 1953 1954 1955
	/* only abort if we didn't race with a real reply */
	if (req->r_got_result) {
		err = le32_to_cpu(req->r_reply_info.head->result);
	} else if (err < 0) {
		dout("aborted request %lld with %d\n", req->r_tid, err);
1956 1957 1958 1959 1960 1961 1962

		/*
		 * ensure we aren't running concurrently with
		 * ceph_fill_trace or ceph_readdir_prepopulate, which
		 * rely on locks (dir mutex) held by our caller.
		 */
		mutex_lock(&req->r_fill_mutex);
1963 1964
		req->r_err = err;
		req->r_aborted = true;
1965
		mutex_unlock(&req->r_fill_mutex);
1966

1967
		if (req->r_locked_dir &&
1968 1969
		    (req->r_op & CEPH_MDS_OP_WRITE))
			ceph_invalidate_dir_request(req);
S
Sage Weil 已提交
1970
	} else {
1971
		err = req->r_err;
S
Sage Weil 已提交
1972 1973
	}

1974 1975
out:
	mutex_unlock(&mdsc->mutex);
S
Sage Weil 已提交
1976 1977 1978 1979
	dout("do_request %p done, result %d\n", req, err);
	return err;
}

1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000
/*
 * Invalidate dir I_COMPLETE, dentry lease state on an aborted MDS
 * namespace request.
 */
void ceph_invalidate_dir_request(struct ceph_mds_request *req)
{
	struct inode *inode = req->r_locked_dir;
	struct ceph_inode_info *ci = ceph_inode(inode);

	dout("invalidate_dir_request %p (I_COMPLETE, lease(s))\n", inode);
	spin_lock(&inode->i_lock);
	ci->i_ceph_flags &= ~CEPH_I_COMPLETE;
	ci->i_release_count++;
	spin_unlock(&inode->i_lock);

	if (req->r_dentry)
		ceph_invalidate_dentry_lease(req->r_dentry);
	if (req->r_old_dentry)
		ceph_invalidate_dentry_lease(req->r_old_dentry);
}

S
Sage Weil 已提交
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
/*
 * Handle mds reply.
 *
 * We take the session mutex and parse and process the reply immediately.
 * This preserves the logical ordering of replies, capabilities, etc., sent
 * by the MDS as they are applied to our local cache.
 */
static void handle_reply(struct ceph_mds_session *session, struct ceph_msg *msg)
{
	struct ceph_mds_client *mdsc = session->s_mdsc;
	struct ceph_mds_request *req;
	struct ceph_mds_reply_head *head = msg->front.iov_base;
	struct ceph_mds_reply_info_parsed *rinfo;  /* parsed reply info */
	u64 tid;
	int err, result;
2016
	int mds = session->s_mds;
S
Sage Weil 已提交
2017 2018 2019

	if (msg->front.iov_len < sizeof(*head)) {
		pr_err("mdsc_handle_reply got corrupt (short) reply\n");
2020
		ceph_msg_dump(msg);
S
Sage Weil 已提交
2021 2022 2023 2024
		return;
	}

	/* get request, session */
2025
	tid = le64_to_cpu(msg->hdr.tid);
S
Sage Weil 已提交
2026 2027 2028 2029 2030 2031 2032 2033 2034 2035
	mutex_lock(&mdsc->mutex);
	req = __lookup_request(mdsc, tid);
	if (!req) {
		dout("handle_reply on unknown tid %llu\n", tid);
		mutex_unlock(&mdsc->mutex);
		return;
	}
	dout("handle_reply %p\n", req);

	/* correct session? */
S
Sage Weil 已提交
2036
	if (req->r_session != session) {
S
Sage Weil 已提交
2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051
		pr_err("mdsc_handle_reply got %llu on session mds%d"
		       " not mds%d\n", tid, session->s_mds,
		       req->r_session ? req->r_session->s_mds : -1);
		mutex_unlock(&mdsc->mutex);
		goto out;
	}

	/* dup? */
	if ((req->r_got_unsafe && !head->safe) ||
	    (req->r_got_safe && head->safe)) {
		pr_warning("got a dup %s reply on %llu from mds%d\n",
			   head->safe ? "safe" : "unsafe", tid, mds);
		mutex_unlock(&mdsc->mutex);
		goto out;
	}
2052 2053 2054 2055 2056 2057
	if (req->r_got_safe && !head->safe) {
		pr_warning("got unsafe after safe on %llu from mds%d\n",
			   tid, mds);
		mutex_unlock(&mdsc->mutex);
		goto out;
	}
S
Sage Weil 已提交
2058 2059 2060 2061

	result = le32_to_cpu(head->result);

	/*
2062 2063 2064 2065 2066
	 * Handle an ESTALE
	 * if we're not talking to the authority, send to them
	 * if the authority has changed while we weren't looking,
	 * send to new authority
	 * Otherwise we just have to return an ESTALE
S
Sage Weil 已提交
2067 2068
	 */
	if (result == -ESTALE) {
2069
		dout("got ESTALE on request %llu", req->r_tid);
S
Sage Weil 已提交
2070 2071 2072
		if (!req->r_inode) {
			/* do nothing; not an authority problem */
		} else if (req->r_direct_mode != USE_AUTH_MDS) {
2073 2074
			dout("not using auth, setting for that now");
			req->r_direct_mode = USE_AUTH_MDS;
S
Sage Weil 已提交
2075 2076 2077
			__do_request(mdsc, req);
			mutex_unlock(&mdsc->mutex);
			goto out;
2078 2079
		} else  {
			struct ceph_inode_info *ci = ceph_inode(req->r_inode);
S
Sage Weil 已提交
2080 2081 2082 2083 2084
			struct ceph_cap *cap = NULL;

			if (req->r_session)
				cap = ceph_get_cap_for_mds(ci,
						   req->r_session->s_mds);
2085 2086 2087 2088 2089 2090 2091 2092 2093

			dout("already using auth");
			if ((!cap || cap != ci->i_auth_cap) ||
			    (cap->mseq != req->r_sent_on_mseq)) {
				dout("but cap changed, so resending");
				__do_request(mdsc, req);
				mutex_unlock(&mdsc->mutex);
				goto out;
			}
S
Sage Weil 已提交
2094
		}
2095
		dout("have to return ESTALE on request %llu", req->r_tid);
S
Sage Weil 已提交
2096 2097
	}

2098

S
Sage Weil 已提交
2099 2100 2101
	if (head->safe) {
		req->r_got_safe = true;
		__unregister_request(mdsc, req);
2102
		complete_all(&req->r_safe_completion);
S
Sage Weil 已提交
2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115

		if (req->r_got_unsafe) {
			/*
			 * We already handled the unsafe response, now do the
			 * cleanup.  No need to examine the response; the MDS
			 * doesn't include any result info in the safe
			 * response.  And even if it did, there is nothing
			 * useful we could do with a revised return value.
			 */
			dout("got safe reply %llu, mds%d\n", tid, mds);
			list_del_init(&req->r_unsafe_item);

			/* last unsafe request during umount? */
S
Sage Weil 已提交
2116
			if (mdsc->stopping && !__get_oldest_req(mdsc))
2117
				complete_all(&mdsc->safe_umount_waiters);
S
Sage Weil 已提交
2118 2119 2120
			mutex_unlock(&mdsc->mutex);
			goto out;
		}
2121
	} else {
S
Sage Weil 已提交
2122 2123 2124 2125 2126 2127
		req->r_got_unsafe = true;
		list_add_tail(&req->r_unsafe_item, &req->r_session->s_unsafe);
	}

	dout("handle_reply tid %lld result %d\n", tid, result);
	rinfo = &req->r_reply_info;
2128
	err = parse_reply_info(msg, rinfo, session->s_con.peer_features);
S
Sage Weil 已提交
2129 2130 2131 2132
	mutex_unlock(&mdsc->mutex);

	mutex_lock(&session->s_mutex);
	if (err < 0) {
2133
		pr_err("mdsc_handle_reply got corrupt reply mds%d(tid:%lld)\n", mds, tid);
2134
		ceph_msg_dump(msg);
S
Sage Weil 已提交
2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149
		goto out_err;
	}

	/* snap trace */
	if (rinfo->snapblob_len) {
		down_write(&mdsc->snap_rwsem);
		ceph_update_snap_trace(mdsc, rinfo->snapblob,
			       rinfo->snapblob + rinfo->snapblob_len,
			       le32_to_cpu(head->op) == CEPH_MDS_OP_RMSNAP);
		downgrade_write(&mdsc->snap_rwsem);
	} else {
		down_read(&mdsc->snap_rwsem);
	}

	/* insert trace into our cache */
2150
	mutex_lock(&req->r_fill_mutex);
2151
	err = ceph_fill_trace(mdsc->fsc->sb, req, req->r_session);
S
Sage Weil 已提交
2152
	if (err == 0) {
2153 2154
		if (result == 0 && req->r_op != CEPH_MDS_OP_GETFILELOCK &&
		    rinfo->dir_nr)
S
Sage Weil 已提交
2155
			ceph_readdir_prepopulate(req, req->r_session);
2156
		ceph_unreserve_caps(mdsc, &req->r_caps_reservation);
S
Sage Weil 已提交
2157
	}
2158
	mutex_unlock(&req->r_fill_mutex);
S
Sage Weil 已提交
2159 2160 2161

	up_read(&mdsc->snap_rwsem);
out_err:
2162 2163 2164 2165 2166 2167 2168 2169 2170
	mutex_lock(&mdsc->mutex);
	if (!req->r_aborted) {
		if (err) {
			req->r_err = err;
		} else {
			req->r_reply = msg;
			ceph_msg_get(msg);
			req->r_got_result = true;
		}
S
Sage Weil 已提交
2171
	} else {
2172
		dout("reply arrived after request %lld was aborted\n", tid);
S
Sage Weil 已提交
2173
	}
2174
	mutex_unlock(&mdsc->mutex);
S
Sage Weil 已提交
2175

S
Sage Weil 已提交
2176
	ceph_add_cap_releases(mdsc, req->r_session);
S
Sage Weil 已提交
2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190
	mutex_unlock(&session->s_mutex);

	/* kick calling process */
	complete_request(mdsc, req);
out:
	ceph_mdsc_put_request(req);
	return;
}



/*
 * handle mds notification that our request has been forwarded.
 */
2191 2192 2193
static void handle_forward(struct ceph_mds_client *mdsc,
			   struct ceph_mds_session *session,
			   struct ceph_msg *msg)
S
Sage Weil 已提交
2194 2195
{
	struct ceph_mds_request *req;
2196
	u64 tid = le64_to_cpu(msg->hdr.tid);
S
Sage Weil 已提交
2197 2198 2199 2200 2201 2202
	u32 next_mds;
	u32 fwd_seq;
	int err = -EINVAL;
	void *p = msg->front.iov_base;
	void *end = p + msg->front.iov_len;

2203
	ceph_decode_need(&p, end, 2*sizeof(u32), bad);
2204 2205
	next_mds = ceph_decode_32(&p);
	fwd_seq = ceph_decode_32(&p);
S
Sage Weil 已提交
2206 2207 2208 2209

	mutex_lock(&mdsc->mutex);
	req = __lookup_request(mdsc, tid);
	if (!req) {
2210
		dout("forward tid %llu to mds%d - req dne\n", tid, next_mds);
S
Sage Weil 已提交
2211 2212 2213
		goto out;  /* dup reply? */
	}

2214 2215 2216 2217 2218
	if (req->r_aborted) {
		dout("forward tid %llu aborted, unregistering\n", tid);
		__unregister_request(mdsc, req);
	} else if (fwd_seq <= req->r_num_fwd) {
		dout("forward tid %llu to mds%d - old seq %d <= %d\n",
S
Sage Weil 已提交
2219 2220 2221
		     tid, next_mds, req->r_num_fwd, fwd_seq);
	} else {
		/* resend. forward race not possible; mds would drop */
2222 2223 2224
		dout("forward tid %llu to mds%d (we resend)\n", tid, next_mds);
		BUG_ON(req->r_err);
		BUG_ON(req->r_got_result);
S
Sage Weil 已提交
2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247
		req->r_num_fwd = fwd_seq;
		req->r_resend_mds = next_mds;
		put_request_session(req);
		__do_request(mdsc, req);
	}
	ceph_mdsc_put_request(req);
out:
	mutex_unlock(&mdsc->mutex);
	return;

bad:
	pr_err("mdsc_handle_forward decode error err=%d\n", err);
}

/*
 * handle a mds session control message
 */
static void handle_session(struct ceph_mds_session *session,
			   struct ceph_msg *msg)
{
	struct ceph_mds_client *mdsc = session->s_mdsc;
	u32 op;
	u64 seq;
2248
	int mds = session->s_mds;
S
Sage Weil 已提交
2249 2250 2251 2252 2253 2254 2255 2256 2257 2258
	struct ceph_mds_session_head *h = msg->front.iov_base;
	int wake = 0;

	/* decode */
	if (msg->front.iov_len != sizeof(*h))
		goto bad;
	op = le32_to_cpu(h->op);
	seq = le64_to_cpu(h->seq);

	mutex_lock(&mdsc->mutex);
2259 2260
	if (op == CEPH_SESSION_CLOSE)
		__unregister_session(mdsc, session);
S
Sage Weil 已提交
2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277
	/* FIXME: this ttl calculation is generous */
	session->s_ttl = jiffies + HZ*mdsc->mdsmap->m_session_autoclose;
	mutex_unlock(&mdsc->mutex);

	mutex_lock(&session->s_mutex);

	dout("handle_session mds%d %s %p state %s seq %llu\n",
	     mds, ceph_session_op_name(op), session,
	     session_state_name(session->s_state), seq);

	if (session->s_state == CEPH_MDS_SESSION_HUNG) {
		session->s_state = CEPH_MDS_SESSION_OPEN;
		pr_info("mds%d came back\n", session->s_mds);
	}

	switch (op) {
	case CEPH_SESSION_OPEN:
2278 2279
		if (session->s_state == CEPH_MDS_SESSION_RECONNECTING)
			pr_info("mds%d reconnect success\n", session->s_mds);
S
Sage Weil 已提交
2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292
		session->s_state = CEPH_MDS_SESSION_OPEN;
		renewed_caps(mdsc, session, 0);
		wake = 1;
		if (mdsc->stopping)
			__close_session(mdsc, session);
		break;

	case CEPH_SESSION_RENEWCAPS:
		if (session->s_renew_seq == seq)
			renewed_caps(mdsc, session, 1);
		break;

	case CEPH_SESSION_CLOSE:
2293 2294
		if (session->s_state == CEPH_MDS_SESSION_RECONNECTING)
			pr_info("mds%d reconnect denied\n", session->s_mds);
S
Sage Weil 已提交
2295 2296
		remove_session_caps(session);
		wake = 1; /* for good measure */
2297
		wake_up_all(&mdsc->session_close_wq);
2298
		kick_requests(mdsc, mds);
S
Sage Weil 已提交
2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330
		break;

	case CEPH_SESSION_STALE:
		pr_info("mds%d caps went stale, renewing\n",
			session->s_mds);
		spin_lock(&session->s_cap_lock);
		session->s_cap_gen++;
		session->s_cap_ttl = 0;
		spin_unlock(&session->s_cap_lock);
		send_renew_caps(mdsc, session);
		break;

	case CEPH_SESSION_RECALL_STATE:
		trim_caps(mdsc, session, le32_to_cpu(h->max_caps));
		break;

	default:
		pr_err("mdsc_handle_session bad op %d mds%d\n", op, mds);
		WARN_ON(1);
	}

	mutex_unlock(&session->s_mutex);
	if (wake) {
		mutex_lock(&mdsc->mutex);
		__wake_requests(mdsc, &session->s_waiting);
		mutex_unlock(&mdsc->mutex);
	}
	return;

bad:
	pr_err("mdsc_handle_session corrupt message mds%d len %d\n", mds,
	       (int)msg->front.iov_len);
2331
	ceph_msg_dump(msg);
S
Sage Weil 已提交
2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363
	return;
}


/*
 * called under session->mutex.
 */
static void replay_unsafe_requests(struct ceph_mds_client *mdsc,
				   struct ceph_mds_session *session)
{
	struct ceph_mds_request *req, *nreq;
	int err;

	dout("replay_unsafe_requests mds%d\n", session->s_mds);

	mutex_lock(&mdsc->mutex);
	list_for_each_entry_safe(req, nreq, &session->s_unsafe, r_unsafe_item) {
		err = __prepare_send_request(mdsc, req, session->s_mds);
		if (!err) {
			ceph_msg_get(req->r_request);
			ceph_con_send(&session->s_con, req->r_request);
		}
	}
	mutex_unlock(&mdsc->mutex);
}

/*
 * Encode information about a cap for a reconnect with the MDS.
 */
static int encode_caps_cb(struct inode *inode, struct ceph_cap *cap,
			  void *arg)
{
S
Sage Weil 已提交
2364 2365 2366 2367 2368
	union {
		struct ceph_mds_cap_reconnect v2;
		struct ceph_mds_cap_reconnect_v1 v1;
	} rec;
	size_t reclen;
S
Sage Weil 已提交
2369
	struct ceph_inode_info *ci;
S
Sage Weil 已提交
2370 2371
	struct ceph_reconnect_state *recon_state = arg;
	struct ceph_pagelist *pagelist = recon_state->pagelist;
S
Sage Weil 已提交
2372 2373 2374 2375 2376 2377 2378 2379 2380 2381
	char *path;
	int pathlen, err;
	u64 pathbase;
	struct dentry *dentry;

	ci = cap->ci;

	dout(" adding %p ino %llx.%llx cap %p %lld %s\n",
	     inode, ceph_vinop(inode), cap, cap->cap_id,
	     ceph_cap_string(cap->issued));
2382 2383 2384
	err = ceph_pagelist_encode_64(pagelist, ceph_ino(inode));
	if (err)
		return err;
S
Sage Weil 已提交
2385 2386 2387 2388 2389 2390

	dentry = d_find_alias(inode);
	if (dentry) {
		path = ceph_mdsc_build_path(dentry, &pathlen, &pathbase, 0);
		if (IS_ERR(path)) {
			err = PTR_ERR(path);
2391
			goto out_dput;
S
Sage Weil 已提交
2392 2393 2394 2395 2396
		}
	} else {
		path = NULL;
		pathlen = 0;
	}
2397 2398
	err = ceph_pagelist_encode_string(pagelist, path, pathlen);
	if (err)
2399
		goto out_free;
S
Sage Weil 已提交
2400 2401 2402 2403

	spin_lock(&inode->i_lock);
	cap->seq = 0;        /* reset cap seq */
	cap->issue_seq = 0;  /* and issue_seq */
S
Sage Weil 已提交
2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423

	if (recon_state->flock) {
		rec.v2.cap_id = cpu_to_le64(cap->cap_id);
		rec.v2.wanted = cpu_to_le32(__ceph_caps_wanted(ci));
		rec.v2.issued = cpu_to_le32(cap->issued);
		rec.v2.snaprealm = cpu_to_le64(ci->i_snap_realm->ino);
		rec.v2.pathbase = cpu_to_le64(pathbase);
		rec.v2.flock_len = 0;
		reclen = sizeof(rec.v2);
	} else {
		rec.v1.cap_id = cpu_to_le64(cap->cap_id);
		rec.v1.wanted = cpu_to_le32(__ceph_caps_wanted(ci));
		rec.v1.issued = cpu_to_le32(cap->issued);
		rec.v1.size = cpu_to_le64(inode->i_size);
		ceph_encode_timespec(&rec.v1.mtime, &inode->i_mtime);
		ceph_encode_timespec(&rec.v1.atime, &inode->i_atime);
		rec.v1.snaprealm = cpu_to_le64(ci->i_snap_realm->ino);
		rec.v1.pathbase = cpu_to_le64(pathbase);
		reclen = sizeof(rec.v1);
	}
S
Sage Weil 已提交
2424 2425
	spin_unlock(&inode->i_lock);

G
Greg Farnum 已提交
2426 2427
	if (recon_state->flock) {
		int num_fcntl_locks, num_flock_locks;
2428 2429 2430 2431
		struct ceph_pagelist_cursor trunc_point;

		ceph_pagelist_set_cursor(pagelist, &trunc_point);
		do {
S
Sage Weil 已提交
2432
			lock_flocks();
2433 2434 2435 2436 2437
			ceph_count_locks(inode, &num_fcntl_locks,
					 &num_flock_locks);
			rec.v2.flock_len = (2*sizeof(u32) +
					    (num_fcntl_locks+num_flock_locks) *
					    sizeof(struct ceph_filelock));
S
Sage Weil 已提交
2438
			unlock_flocks();
2439 2440 2441 2442 2443 2444 2445 2446 2447 2448

			/* pre-alloc pagelist */
			ceph_pagelist_truncate(pagelist, &trunc_point);
			err = ceph_pagelist_append(pagelist, &rec, reclen);
			if (!err)
				err = ceph_pagelist_reserve(pagelist,
							    rec.v2.flock_len);

			/* encode locks */
			if (!err) {
S
Sage Weil 已提交
2449
				lock_flocks();
2450 2451 2452 2453
				err = ceph_encode_locks(inode,
							pagelist,
							num_fcntl_locks,
							num_flock_locks);
S
Sage Weil 已提交
2454
				unlock_flocks();
2455 2456
			}
		} while (err == -ENOSPC);
2457 2458
	} else {
		err = ceph_pagelist_append(pagelist, &rec, reclen);
G
Greg Farnum 已提交
2459
	}
2460

2461
out_free:
S
Sage Weil 已提交
2462
	kfree(path);
2463
out_dput:
S
Sage Weil 已提交
2464
	dput(dentry);
2465
	return err;
S
Sage Weil 已提交
2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480
}


/*
 * If an MDS fails and recovers, clients need to reconnect in order to
 * reestablish shared state.  This includes all caps issued through
 * this session _and_ the snap_realm hierarchy.  Because it's not
 * clear which snap realms the mds cares about, we send everything we
 * know about.. that ensures we'll then get any new info the
 * recovering MDS might have.
 *
 * This is a relatively heavyweight operation, but it's rare.
 *
 * called with mdsc->mutex held.
 */
2481 2482
static void send_mds_reconnect(struct ceph_mds_client *mdsc,
			       struct ceph_mds_session *session)
S
Sage Weil 已提交
2483 2484
{
	struct ceph_msg *reply;
S
Sage Weil 已提交
2485
	struct rb_node *p;
2486
	int mds = session->s_mds;
2487
	int err = -ENOMEM;
2488
	struct ceph_pagelist *pagelist;
S
Sage Weil 已提交
2489
	struct ceph_reconnect_state recon_state;
S
Sage Weil 已提交
2490

2491
	pr_info("mds%d reconnect start\n", mds);
S
Sage Weil 已提交
2492

2493 2494 2495 2496 2497
	pagelist = kmalloc(sizeof(*pagelist), GFP_NOFS);
	if (!pagelist)
		goto fail_nopagelist;
	ceph_pagelist_init(pagelist);

2498
	reply = ceph_msg_new(CEPH_MSG_CLIENT_RECONNECT, 0, GFP_NOFS);
2499
	if (!reply)
2500 2501
		goto fail_nomsg;

2502 2503 2504
	mutex_lock(&session->s_mutex);
	session->s_state = CEPH_MDS_SESSION_RECONNECTING;
	session->s_seq = 0;
S
Sage Weil 已提交
2505

2506 2507
	ceph_con_open(&session->s_con,
		      ceph_mdsmap_get_addr(mdsc->mdsmap, mds));
S
Sage Weil 已提交
2508

2509 2510
	/* replay unsafe requests */
	replay_unsafe_requests(mdsc, session);
S
Sage Weil 已提交
2511 2512 2513 2514 2515 2516

	down_read(&mdsc->snap_rwsem);

	dout("session %p state %s\n", session,
	     session_state_name(session->s_state));

2517 2518 2519
	/* drop old cap expires; we're about to reestablish that state */
	discard_cap_releases(mdsc, session);

S
Sage Weil 已提交
2520
	/* traverse this session's caps */
2521 2522 2523
	err = ceph_pagelist_encode_32(pagelist, session->s_nr_caps);
	if (err)
		goto fail;
S
Sage Weil 已提交
2524 2525 2526 2527

	recon_state.pagelist = pagelist;
	recon_state.flock = session->s_con.peer_features & CEPH_FEATURE_FLOCK;
	err = iterate_session_caps(session, encode_caps_cb, &recon_state);
S
Sage Weil 已提交
2528
	if (err < 0)
2529
		goto fail;
S
Sage Weil 已提交
2530 2531 2532 2533 2534 2535

	/*
	 * snaprealms.  we provide mds with the ino, seq (version), and
	 * parent for all of our realms.  If the mds has any newer info,
	 * it will tell us.
	 */
S
Sage Weil 已提交
2536 2537 2538
	for (p = rb_first(&mdsc->snap_realms); p; p = rb_next(p)) {
		struct ceph_snap_realm *realm =
			rb_entry(p, struct ceph_snap_realm, node);
2539
		struct ceph_mds_snaprealm_reconnect sr_rec;
S
Sage Weil 已提交
2540 2541 2542

		dout(" adding snap realm %llx seq %lld parent %llx\n",
		     realm->ino, realm->seq, realm->parent_ino);
2543 2544 2545 2546 2547 2548
		sr_rec.ino = cpu_to_le64(realm->ino);
		sr_rec.seq = cpu_to_le64(realm->seq);
		sr_rec.parent = cpu_to_le64(realm->parent_ino);
		err = ceph_pagelist_append(pagelist, &sr_rec, sizeof(sr_rec));
		if (err)
			goto fail;
S
Sage Weil 已提交
2549 2550
	}

2551
	reply->pagelist = pagelist;
S
Sage Weil 已提交
2552 2553
	if (recon_state.flock)
		reply->hdr.version = cpu_to_le16(2);
2554 2555
	reply->hdr.data_len = cpu_to_le32(pagelist->length);
	reply->nr_pages = calc_pages_for(0, pagelist->length);
S
Sage Weil 已提交
2556 2557
	ceph_con_send(&session->s_con, reply);

2558 2559 2560 2561 2562 2563
	mutex_unlock(&session->s_mutex);

	mutex_lock(&mdsc->mutex);
	__wake_requests(mdsc, &session->s_waiting);
	mutex_unlock(&mdsc->mutex);

S
Sage Weil 已提交
2564 2565 2566
	up_read(&mdsc->snap_rwsem);
	return;

2567
fail:
S
Sage Weil 已提交
2568
	ceph_msg_put(reply);
2569 2570
	up_read(&mdsc->snap_rwsem);
	mutex_unlock(&session->s_mutex);
2571 2572 2573 2574
fail_nomsg:
	ceph_pagelist_release(pagelist);
	kfree(pagelist);
fail_nopagelist:
2575 2576
	pr_err("error %d preparing reconnect for mds%d\n", err, mds);
	return;
S
Sage Weil 已提交
2577 2578 2579 2580 2581 2582 2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603
}


/*
 * compare old and new mdsmaps, kicking requests
 * and closing out old connections as necessary
 *
 * called under mdsc->mutex.
 */
static void check_new_map(struct ceph_mds_client *mdsc,
			  struct ceph_mdsmap *newmap,
			  struct ceph_mdsmap *oldmap)
{
	int i;
	int oldstate, newstate;
	struct ceph_mds_session *s;

	dout("check_new_map new %u old %u\n",
	     newmap->m_epoch, oldmap->m_epoch);

	for (i = 0; i < oldmap->m_max_mds && i < mdsc->max_sessions; i++) {
		if (mdsc->sessions[i] == NULL)
			continue;
		s = mdsc->sessions[i];
		oldstate = ceph_mdsmap_get_state(oldmap, i);
		newstate = ceph_mdsmap_get_state(newmap, i);

2604
		dout("check_new_map mds%d state %s%s -> %s%s (session %s)\n",
S
Sage Weil 已提交
2605
		     i, ceph_mds_state_name(oldstate),
2606
		     ceph_mdsmap_is_laggy(oldmap, i) ? " (laggy)" : "",
S
Sage Weil 已提交
2607
		     ceph_mds_state_name(newstate),
2608
		     ceph_mdsmap_is_laggy(newmap, i) ? " (laggy)" : "",
S
Sage Weil 已提交
2609 2610 2611 2612 2613 2614 2615 2616 2617
		     session_state_name(s->s_state));

		if (memcmp(ceph_mdsmap_get_addr(oldmap, i),
			   ceph_mdsmap_get_addr(newmap, i),
			   sizeof(struct ceph_entity_addr))) {
			if (s->s_state == CEPH_MDS_SESSION_OPENING) {
				/* the session never opened, just close it
				 * out now */
				__wake_requests(mdsc, &s->s_waiting);
2618
				__unregister_session(mdsc, s);
S
Sage Weil 已提交
2619 2620 2621 2622 2623 2624 2625 2626 2627 2628 2629
			} else {
				/* just close it */
				mutex_unlock(&mdsc->mutex);
				mutex_lock(&s->s_mutex);
				mutex_lock(&mdsc->mutex);
				ceph_con_close(&s->s_con);
				mutex_unlock(&s->s_mutex);
				s->s_state = CEPH_MDS_SESSION_RESTARTING;
			}

			/* kick any requests waiting on the recovering mds */
2630
			kick_requests(mdsc, i);
S
Sage Weil 已提交
2631 2632 2633 2634 2635 2636 2637 2638
		} else if (oldstate == newstate) {
			continue;  /* nothing new with this mds */
		}

		/*
		 * send reconnect?
		 */
		if (s->s_state == CEPH_MDS_SESSION_RESTARTING &&
2639 2640 2641 2642 2643
		    newstate >= CEPH_MDS_STATE_RECONNECT) {
			mutex_unlock(&mdsc->mutex);
			send_mds_reconnect(mdsc, s);
			mutex_lock(&mdsc->mutex);
		}
S
Sage Weil 已提交
2644 2645

		/*
2646
		 * kick request on any mds that has gone active.
S
Sage Weil 已提交
2647 2648 2649
		 */
		if (oldstate < CEPH_MDS_STATE_ACTIVE &&
		    newstate >= CEPH_MDS_STATE_ACTIVE) {
2650 2651 2652 2653
			if (oldstate != CEPH_MDS_STATE_CREATING &&
			    oldstate != CEPH_MDS_STATE_STARTING)
				pr_info("mds%d recovery completed\n", s->s_mds);
			kick_requests(mdsc, i);
S
Sage Weil 已提交
2654
			ceph_kick_flushing_caps(mdsc, s);
2655
			wake_up_session_caps(s, 1);
S
Sage Weil 已提交
2656 2657
		}
	}
2658 2659 2660 2661 2662 2663 2664 2665 2666 2667 2668 2669 2670 2671 2672

	for (i = 0; i < newmap->m_max_mds && i < mdsc->max_sessions; i++) {
		s = mdsc->sessions[i];
		if (!s)
			continue;
		if (!ceph_mdsmap_is_laggy(newmap, i))
			continue;
		if (s->s_state == CEPH_MDS_SESSION_OPEN ||
		    s->s_state == CEPH_MDS_SESSION_HUNG ||
		    s->s_state == CEPH_MDS_SESSION_CLOSING) {
			dout(" connecting to export targets of laggy mds%d\n",
			     i);
			__open_export_target_sessions(mdsc, s);
		}
	}
S
Sage Weil 已提交
2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 2683 2684 2685 2686 2687 2688 2689 2690 2691
}



/*
 * leases
 */

/*
 * caller must hold session s_mutex, dentry->d_lock
 */
void __ceph_mdsc_drop_dentry_lease(struct dentry *dentry)
{
	struct ceph_dentry_info *di = ceph_dentry(dentry);

	ceph_put_mds_session(di->lease_session);
	di->lease_session = NULL;
}

2692 2693 2694
static void handle_lease(struct ceph_mds_client *mdsc,
			 struct ceph_mds_session *session,
			 struct ceph_msg *msg)
S
Sage Weil 已提交
2695
{
2696
	struct super_block *sb = mdsc->fsc->sb;
S
Sage Weil 已提交
2697 2698 2699
	struct inode *inode;
	struct dentry *parent, *dentry;
	struct ceph_dentry_info *di;
2700
	int mds = session->s_mds;
S
Sage Weil 已提交
2701
	struct ceph_mds_lease *h = msg->front.iov_base;
2702
	u32 seq;
S
Sage Weil 已提交
2703 2704 2705 2706 2707 2708 2709 2710 2711 2712 2713 2714 2715
	struct ceph_vino vino;
	int mask;
	struct qstr dname;
	int release = 0;

	dout("handle_lease from mds%d\n", mds);

	/* decode */
	if (msg->front.iov_len < sizeof(*h) + sizeof(u32))
		goto bad;
	vino.ino = le64_to_cpu(h->ino);
	vino.snap = CEPH_NOSNAP;
	mask = le16_to_cpu(h->mask);
2716
	seq = le32_to_cpu(h->seq);
S
Sage Weil 已提交
2717 2718 2719 2720 2721 2722 2723 2724 2725 2726
	dname.name = (void *)h + sizeof(*h) + sizeof(u32);
	dname.len = msg->front.iov_len - sizeof(*h) - sizeof(u32);
	if (dname.len != get_unaligned_le32(h+1))
		goto bad;

	mutex_lock(&session->s_mutex);
	session->s_seq++;

	/* lookup inode */
	inode = ceph_find_inode(sb, vino);
2727 2728 2729
	dout("handle_lease %s, mask %d, ino %llx %p %.*s\n",
	     ceph_lease_op_name(h->action), mask, vino.ino, inode,
	     dname.len, dname.name);
S
Sage Weil 已提交
2730 2731 2732 2733 2734 2735 2736 2737 2738 2739 2740 2741 2742 2743 2744 2745 2746 2747 2748 2749 2750 2751 2752
	if (inode == NULL) {
		dout("handle_lease no inode %llx\n", vino.ino);
		goto release;
	}

	/* dentry */
	parent = d_find_alias(inode);
	if (!parent) {
		dout("no parent dentry on inode %p\n", inode);
		WARN_ON(1);
		goto release;  /* hrm... */
	}
	dname.hash = full_name_hash(dname.name, dname.len);
	dentry = d_lookup(parent, &dname);
	dput(parent);
	if (!dentry)
		goto release;

	spin_lock(&dentry->d_lock);
	di = ceph_dentry(dentry);
	switch (h->action) {
	case CEPH_MDS_LEASE_REVOKE:
		if (di && di->lease_session == session) {
2753 2754
			if (ceph_seq_cmp(di->lease_seq, seq) > 0)
				h->seq = cpu_to_le32(di->lease_seq);
S
Sage Weil 已提交
2755 2756 2757 2758 2759 2760 2761 2762 2763 2764 2765 2766 2767
			__ceph_mdsc_drop_dentry_lease(dentry);
		}
		release = 1;
		break;

	case CEPH_MDS_LEASE_RENEW:
		if (di && di->lease_session == session &&
		    di->lease_gen == session->s_cap_gen &&
		    di->lease_renew_from &&
		    di->lease_renew_after == 0) {
			unsigned long duration =
				le32_to_cpu(h->duration_ms) * HZ / 1000;

2768
			di->lease_seq = seq;
S
Sage Weil 已提交
2769 2770 2771 2772 2773 2774 2775 2776 2777 2778 2779 2780 2781 2782 2783 2784 2785 2786 2787 2788 2789 2790 2791 2792 2793 2794
			dentry->d_time = di->lease_renew_from + duration;
			di->lease_renew_after = di->lease_renew_from +
				(duration >> 1);
			di->lease_renew_from = 0;
		}
		break;
	}
	spin_unlock(&dentry->d_lock);
	dput(dentry);

	if (!release)
		goto out;

release:
	/* let's just reuse the same message */
	h->action = CEPH_MDS_LEASE_REVOKE_ACK;
	ceph_msg_get(msg);
	ceph_con_send(&session->s_con, msg);

out:
	iput(inode);
	mutex_unlock(&session->s_mutex);
	return;

bad:
	pr_err("corrupt lease message\n");
2795
	ceph_msg_dump(msg);
S
Sage Weil 已提交
2796 2797 2798 2799 2800 2801 2802 2803 2804 2805 2806 2807 2808 2809 2810 2811 2812
}

void ceph_mdsc_lease_send_msg(struct ceph_mds_session *session,
			      struct inode *inode,
			      struct dentry *dentry, char action,
			      u32 seq)
{
	struct ceph_msg *msg;
	struct ceph_mds_lease *lease;
	int len = sizeof(*lease) + sizeof(u32);
	int dnamelen = 0;

	dout("lease_send_msg inode %p dentry %p %s to mds%d\n",
	     inode, dentry, ceph_lease_op_name(action), session->s_mds);
	dnamelen = dentry->d_name.len;
	len += dnamelen;

2813
	msg = ceph_msg_new(CEPH_MSG_CLIENT_LEASE, len, GFP_NOFS);
2814
	if (!msg)
S
Sage Weil 已提交
2815 2816 2817
		return;
	lease = msg->front.iov_base;
	lease->action = action;
S
Sage Weil 已提交
2818
	lease->mask = cpu_to_le16(1);
S
Sage Weil 已提交
2819 2820 2821 2822 2823 2824 2825 2826 2827 2828 2829 2830 2831 2832 2833 2834 2835 2836 2837 2838 2839 2840 2841 2842 2843 2844 2845 2846 2847
	lease->ino = cpu_to_le64(ceph_vino(inode).ino);
	lease->first = lease->last = cpu_to_le64(ceph_vino(inode).snap);
	lease->seq = cpu_to_le32(seq);
	put_unaligned_le32(dnamelen, lease + 1);
	memcpy((void *)(lease + 1) + 4, dentry->d_name.name, dnamelen);

	/*
	 * if this is a preemptive lease RELEASE, no need to
	 * flush request stream, since the actual request will
	 * soon follow.
	 */
	msg->more_to_follow = (action == CEPH_MDS_LEASE_RELEASE);

	ceph_con_send(&session->s_con, msg);
}

/*
 * Preemptively release a lease we expect to invalidate anyway.
 * Pass @inode always, @dentry is optional.
 */
void ceph_mdsc_lease_release(struct ceph_mds_client *mdsc, struct inode *inode,
			     struct dentry *dentry, int mask)
{
	struct ceph_dentry_info *di;
	struct ceph_mds_session *session;
	u32 seq;

	BUG_ON(inode == NULL);
	BUG_ON(dentry == NULL);
S
Sage Weil 已提交
2848
	BUG_ON(mask == 0);
S
Sage Weil 已提交
2849 2850 2851 2852 2853 2854 2855 2856 2857 2858 2859 2860 2861 2862 2863 2864 2865 2866 2867 2868 2869 2870 2871 2872 2873 2874 2875 2876 2877 2878 2879 2880 2881 2882 2883 2884 2885 2886 2887 2888 2889 2890 2891 2892 2893 2894 2895 2896 2897 2898 2899 2900 2901 2902 2903 2904 2905 2906 2907 2908 2909 2910 2911 2912 2913 2914 2915 2916 2917 2918 2919

	/* is dentry lease valid? */
	spin_lock(&dentry->d_lock);
	di = ceph_dentry(dentry);
	if (!di || !di->lease_session ||
	    di->lease_session->s_mds < 0 ||
	    di->lease_gen != di->lease_session->s_cap_gen ||
	    !time_before(jiffies, dentry->d_time)) {
		dout("lease_release inode %p dentry %p -- "
		     "no lease on %d\n",
		     inode, dentry, mask);
		spin_unlock(&dentry->d_lock);
		return;
	}

	/* we do have a lease on this dentry; note mds and seq */
	session = ceph_get_mds_session(di->lease_session);
	seq = di->lease_seq;
	__ceph_mdsc_drop_dentry_lease(dentry);
	spin_unlock(&dentry->d_lock);

	dout("lease_release inode %p dentry %p mask %d to mds%d\n",
	     inode, dentry, mask, session->s_mds);
	ceph_mdsc_lease_send_msg(session, inode, dentry,
				 CEPH_MDS_LEASE_RELEASE, seq);
	ceph_put_mds_session(session);
}

/*
 * drop all leases (and dentry refs) in preparation for umount
 */
static void drop_leases(struct ceph_mds_client *mdsc)
{
	int i;

	dout("drop_leases\n");
	mutex_lock(&mdsc->mutex);
	for (i = 0; i < mdsc->max_sessions; i++) {
		struct ceph_mds_session *s = __ceph_lookup_mds_session(mdsc, i);
		if (!s)
			continue;
		mutex_unlock(&mdsc->mutex);
		mutex_lock(&s->s_mutex);
		mutex_unlock(&s->s_mutex);
		ceph_put_mds_session(s);
		mutex_lock(&mdsc->mutex);
	}
	mutex_unlock(&mdsc->mutex);
}



/*
 * delayed work -- periodically trim expired leases, renew caps with mds
 */
static void schedule_delayed(struct ceph_mds_client *mdsc)
{
	int delay = 5;
	unsigned hz = round_jiffies_relative(HZ * delay);
	schedule_delayed_work(&mdsc->delayed_work, hz);
}

static void delayed_work(struct work_struct *work)
{
	int i;
	struct ceph_mds_client *mdsc =
		container_of(work, struct ceph_mds_client, delayed_work.work);
	int renew_interval;
	int renew_caps;

	dout("mdsc delayed_work\n");
2920
	ceph_check_delayed_caps(mdsc);
S
Sage Weil 已提交
2921 2922 2923 2924 2925 2926 2927 2928 2929 2930 2931 2932 2933 2934 2935 2936 2937 2938 2939 2940 2941 2942 2943 2944 2945 2946 2947 2948 2949 2950 2951 2952 2953 2954 2955 2956 2957

	mutex_lock(&mdsc->mutex);
	renew_interval = mdsc->mdsmap->m_session_timeout >> 2;
	renew_caps = time_after_eq(jiffies, HZ*renew_interval +
				   mdsc->last_renew_caps);
	if (renew_caps)
		mdsc->last_renew_caps = jiffies;

	for (i = 0; i < mdsc->max_sessions; i++) {
		struct ceph_mds_session *s = __ceph_lookup_mds_session(mdsc, i);
		if (s == NULL)
			continue;
		if (s->s_state == CEPH_MDS_SESSION_CLOSING) {
			dout("resending session close request for mds%d\n",
			     s->s_mds);
			request_close_session(mdsc, s);
			ceph_put_mds_session(s);
			continue;
		}
		if (s->s_ttl && time_after(jiffies, s->s_ttl)) {
			if (s->s_state == CEPH_MDS_SESSION_OPEN) {
				s->s_state = CEPH_MDS_SESSION_HUNG;
				pr_info("mds%d hung\n", s->s_mds);
			}
		}
		if (s->s_state < CEPH_MDS_SESSION_OPEN) {
			/* this mds is failed or recovering, just wait */
			ceph_put_mds_session(s);
			continue;
		}
		mutex_unlock(&mdsc->mutex);

		mutex_lock(&s->s_mutex);
		if (renew_caps)
			send_renew_caps(mdsc, s);
		else
			ceph_con_keepalive(&s->s_con);
S
Sage Weil 已提交
2958
		ceph_add_cap_releases(mdsc, s);
2959 2960
		if (s->s_state == CEPH_MDS_SESSION_OPEN ||
		    s->s_state == CEPH_MDS_SESSION_HUNG)
2961
			ceph_send_cap_releases(mdsc, s);
S
Sage Weil 已提交
2962 2963 2964 2965 2966 2967 2968 2969 2970 2971
		mutex_unlock(&s->s_mutex);
		ceph_put_mds_session(s);

		mutex_lock(&mdsc->mutex);
	}
	mutex_unlock(&mdsc->mutex);

	schedule_delayed(mdsc);
}

2972
int ceph_mdsc_init(struct ceph_fs_client *fsc)
S
Sage Weil 已提交
2973 2974

{
2975 2976 2977 2978 2979 2980 2981
	struct ceph_mds_client *mdsc;

	mdsc = kzalloc(sizeof(struct ceph_mds_client), GFP_NOFS);
	if (!mdsc)
		return -ENOMEM;
	mdsc->fsc = fsc;
	fsc->mdsc = mdsc;
S
Sage Weil 已提交
2982 2983
	mutex_init(&mdsc->mutex);
	mdsc->mdsmap = kzalloc(sizeof(*mdsc->mdsmap), GFP_NOFS);
C
Cheng Renquan 已提交
2984 2985 2986
	if (mdsc->mdsmap == NULL)
		return -ENOMEM;

S
Sage Weil 已提交
2987
	init_completion(&mdsc->safe_umount_waiters);
2988
	init_waitqueue_head(&mdsc->session_close_wq);
S
Sage Weil 已提交
2989 2990 2991 2992 2993
	INIT_LIST_HEAD(&mdsc->waiting_for_map);
	mdsc->sessions = NULL;
	mdsc->max_sessions = 0;
	mdsc->stopping = 0;
	init_rwsem(&mdsc->snap_rwsem);
S
Sage Weil 已提交
2994
	mdsc->snap_realms = RB_ROOT;
S
Sage Weil 已提交
2995 2996 2997
	INIT_LIST_HEAD(&mdsc->snap_empty);
	spin_lock_init(&mdsc->snap_empty_lock);
	mdsc->last_tid = 0;
S
Sage Weil 已提交
2998
	mdsc->request_tree = RB_ROOT;
S
Sage Weil 已提交
2999 3000 3001 3002 3003 3004 3005 3006
	INIT_DELAYED_WORK(&mdsc->delayed_work, delayed_work);
	mdsc->last_renew_caps = jiffies;
	INIT_LIST_HEAD(&mdsc->cap_delay_list);
	spin_lock_init(&mdsc->cap_delay_lock);
	INIT_LIST_HEAD(&mdsc->snap_flush_list);
	spin_lock_init(&mdsc->snap_flush_lock);
	mdsc->cap_flush_seq = 0;
	INIT_LIST_HEAD(&mdsc->cap_dirty);
S
Sage Weil 已提交
3007
	INIT_LIST_HEAD(&mdsc->cap_dirty_migrating);
S
Sage Weil 已提交
3008 3009 3010 3011 3012
	mdsc->num_cap_flushing = 0;
	spin_lock_init(&mdsc->cap_dirty_lock);
	init_waitqueue_head(&mdsc->cap_flushing_wq);
	spin_lock_init(&mdsc->dentry_lru_lock);
	INIT_LIST_HEAD(&mdsc->dentry_lru);
C
Cheng Renquan 已提交
3013

3014
	ceph_caps_init(mdsc);
3015
	ceph_adjust_min_caps(mdsc, fsc->min_caps);
3016

3017
	return 0;
S
Sage Weil 已提交
3018 3019 3020 3021 3022 3023 3024 3025 3026
}

/*
 * Wait for safe replies on open mds requests.  If we time out, drop
 * all requests from the tree to avoid dangling dentry refs.
 */
static void wait_requests(struct ceph_mds_client *mdsc)
{
	struct ceph_mds_request *req;
3027
	struct ceph_fs_client *fsc = mdsc->fsc;
S
Sage Weil 已提交
3028 3029

	mutex_lock(&mdsc->mutex);
S
Sage Weil 已提交
3030
	if (__get_oldest_req(mdsc)) {
S
Sage Weil 已提交
3031
		mutex_unlock(&mdsc->mutex);
S
Sage Weil 已提交
3032

S
Sage Weil 已提交
3033 3034
		dout("wait_requests waiting for requests\n");
		wait_for_completion_timeout(&mdsc->safe_umount_waiters,
3035
				    fsc->client->options->mount_timeout * HZ);
S
Sage Weil 已提交
3036 3037

		/* tear down remaining requests */
S
Sage Weil 已提交
3038 3039
		mutex_lock(&mdsc->mutex);
		while ((req = __get_oldest_req(mdsc))) {
S
Sage Weil 已提交
3040 3041
			dout("wait_requests timed out on tid %llu\n",
			     req->r_tid);
S
Sage Weil 已提交
3042
			__unregister_request(mdsc, req);
S
Sage Weil 已提交
3043 3044 3045 3046 3047 3048 3049 3050 3051 3052 3053 3054 3055 3056 3057 3058
		}
	}
	mutex_unlock(&mdsc->mutex);
	dout("wait_requests done\n");
}

/*
 * called before mount is ro, and before dentries are torn down.
 * (hmm, does this still race with new lookups?)
 */
void ceph_mdsc_pre_umount(struct ceph_mds_client *mdsc)
{
	dout("pre_umount\n");
	mdsc->stopping = 1;

	drop_leases(mdsc);
3059
	ceph_flush_dirty_caps(mdsc);
S
Sage Weil 已提交
3060
	wait_requests(mdsc);
3061 3062 3063 3064 3065 3066

	/*
	 * wait for reply handlers to drop their request refs and
	 * their inode/dcache refs
	 */
	ceph_msgr_flush();
S
Sage Weil 已提交
3067 3068 3069 3070 3071 3072 3073
}

/*
 * wait for all write mds requests to flush.
 */
static void wait_unsafe_requests(struct ceph_mds_client *mdsc, u64 want_tid)
{
3074
	struct ceph_mds_request *req = NULL, *nextreq;
S
Sage Weil 已提交
3075
	struct rb_node *n;
S
Sage Weil 已提交
3076 3077 3078

	mutex_lock(&mdsc->mutex);
	dout("wait_unsafe_requests want %lld\n", want_tid);
3079
restart:
S
Sage Weil 已提交
3080 3081
	req = __get_oldest_req(mdsc);
	while (req && req->r_tid <= want_tid) {
3082 3083 3084 3085 3086 3087
		/* find next request */
		n = rb_next(&req->r_node);
		if (n)
			nextreq = rb_entry(n, struct ceph_mds_request, r_node);
		else
			nextreq = NULL;
S
Sage Weil 已提交
3088 3089 3090
		if ((req->r_op & CEPH_MDS_OP_WRITE)) {
			/* write op */
			ceph_mdsc_get_request(req);
3091 3092
			if (nextreq)
				ceph_mdsc_get_request(nextreq);
S
Sage Weil 已提交
3093 3094 3095 3096 3097 3098
			mutex_unlock(&mdsc->mutex);
			dout("wait_unsafe_requests  wait on %llu (want %llu)\n",
			     req->r_tid, want_tid);
			wait_for_completion(&req->r_safe_completion);
			mutex_lock(&mdsc->mutex);
			ceph_mdsc_put_request(req);
3099 3100 3101 3102 3103 3104 3105 3106
			if (!nextreq)
				break;  /* next dne before, so we're done! */
			if (RB_EMPTY_NODE(&nextreq->r_node)) {
				/* next request was removed from tree */
				ceph_mdsc_put_request(nextreq);
				goto restart;
			}
			ceph_mdsc_put_request(nextreq);  /* won't go away */
S
Sage Weil 已提交
3107
		}
3108
		req = nextreq;
S
Sage Weil 已提交
3109 3110 3111 3112 3113 3114 3115 3116 3117
	}
	mutex_unlock(&mdsc->mutex);
	dout("wait_unsafe_requests done\n");
}

void ceph_mdsc_sync(struct ceph_mds_client *mdsc)
{
	u64 want_tid, want_flush;

3118
	if (mdsc->fsc->mount_state == CEPH_MOUNT_SHUTDOWN)
3119 3120
		return;

S
Sage Weil 已提交
3121 3122 3123 3124 3125 3126 3127
	dout("sync\n");
	mutex_lock(&mdsc->mutex);
	want_tid = mdsc->last_tid;
	want_flush = mdsc->cap_flush_seq;
	mutex_unlock(&mdsc->mutex);
	dout("sync want tid %lld flush_seq %lld\n", want_tid, want_flush);

3128
	ceph_flush_dirty_caps(mdsc);
S
Sage Weil 已提交
3129 3130 3131 3132 3133

	wait_unsafe_requests(mdsc, want_tid);
	wait_event(mdsc->cap_flushing_wq, check_cap_flush(mdsc, want_flush));
}

3134 3135 3136 3137 3138 3139 3140
/*
 * true if all sessions are closed, or we force unmount
 */
bool done_closing_sessions(struct ceph_mds_client *mdsc)
{
	int i, n = 0;

3141
	if (mdsc->fsc->mount_state == CEPH_MOUNT_SHUTDOWN)
3142 3143 3144 3145 3146 3147 3148 3149 3150
		return true;

	mutex_lock(&mdsc->mutex);
	for (i = 0; i < mdsc->max_sessions; i++)
		if (mdsc->sessions[i])
			n++;
	mutex_unlock(&mdsc->mutex);
	return n == 0;
}
S
Sage Weil 已提交
3151 3152 3153 3154 3155 3156 3157 3158

/*
 * called after sb is ro.
 */
void ceph_mdsc_close_sessions(struct ceph_mds_client *mdsc)
{
	struct ceph_mds_session *session;
	int i;
3159 3160
	struct ceph_fs_client *fsc = mdsc->fsc;
	unsigned long timeout = fsc->client->options->mount_timeout * HZ;
S
Sage Weil 已提交
3161 3162 3163 3164

	dout("close_sessions\n");

	/* close sessions */
3165 3166 3167 3168 3169
	mutex_lock(&mdsc->mutex);
	for (i = 0; i < mdsc->max_sessions; i++) {
		session = __ceph_lookup_mds_session(mdsc, i);
		if (!session)
			continue;
S
Sage Weil 已提交
3170
		mutex_unlock(&mdsc->mutex);
3171 3172 3173 3174
		mutex_lock(&session->s_mutex);
		__close_session(mdsc, session);
		mutex_unlock(&session->s_mutex);
		ceph_put_mds_session(session);
S
Sage Weil 已提交
3175 3176
		mutex_lock(&mdsc->mutex);
	}
3177 3178 3179 3180 3181
	mutex_unlock(&mdsc->mutex);

	dout("waiting for sessions to close\n");
	wait_event_timeout(mdsc->session_close_wq, done_closing_sessions(mdsc),
			   timeout);
S
Sage Weil 已提交
3182 3183

	/* tear down remaining sessions */
3184
	mutex_lock(&mdsc->mutex);
S
Sage Weil 已提交
3185 3186 3187
	for (i = 0; i < mdsc->max_sessions; i++) {
		if (mdsc->sessions[i]) {
			session = get_session(mdsc->sessions[i]);
3188
			__unregister_session(mdsc, session);
S
Sage Weil 已提交
3189 3190 3191 3192 3193 3194 3195 3196 3197 3198 3199 3200 3201 3202 3203 3204 3205 3206
			mutex_unlock(&mdsc->mutex);
			mutex_lock(&session->s_mutex);
			remove_session_caps(session);
			mutex_unlock(&session->s_mutex);
			ceph_put_mds_session(session);
			mutex_lock(&mdsc->mutex);
		}
	}
	WARN_ON(!list_empty(&mdsc->cap_delay_list));
	mutex_unlock(&mdsc->mutex);

	ceph_cleanup_empty_realms(mdsc);

	cancel_delayed_work_sync(&mdsc->delayed_work); /* cancel timer */

	dout("stopped\n");
}

3207
static void ceph_mdsc_stop(struct ceph_mds_client *mdsc)
S
Sage Weil 已提交
3208 3209 3210 3211 3212 3213
{
	dout("stop\n");
	cancel_delayed_work_sync(&mdsc->delayed_work); /* cancel timer */
	if (mdsc->mdsmap)
		ceph_mdsmap_destroy(mdsc->mdsmap);
	kfree(mdsc->sessions);
3214
	ceph_caps_finalize(mdsc);
S
Sage Weil 已提交
3215 3216
}

3217 3218 3219 3220
void ceph_mdsc_destroy(struct ceph_fs_client *fsc)
{
	struct ceph_mds_client *mdsc = fsc->mdsc;

3221
	dout("mdsc_destroy %p\n", mdsc);
3222
	ceph_mdsc_stop(mdsc);
3223 3224 3225 3226

	/* flush out any connection work with references to us */
	ceph_msgr_flush();

3227 3228
	fsc->mdsc = NULL;
	kfree(mdsc);
3229
	dout("mdsc_destroy %p done\n", mdsc);
3230 3231
}

S
Sage Weil 已提交
3232 3233 3234 3235 3236 3237 3238 3239 3240 3241 3242 3243 3244 3245 3246 3247

/*
 * handle mds map update.
 */
void ceph_mdsc_handle_map(struct ceph_mds_client *mdsc, struct ceph_msg *msg)
{
	u32 epoch;
	u32 maplen;
	void *p = msg->front.iov_base;
	void *end = p + msg->front.iov_len;
	struct ceph_mdsmap *newmap, *oldmap;
	struct ceph_fsid fsid;
	int err = -EINVAL;

	ceph_decode_need(&p, end, sizeof(fsid)+2*sizeof(u32), bad);
	ceph_decode_copy(&p, &fsid, sizeof(fsid));
3248
	if (ceph_check_fsid(mdsc->fsc->client, &fsid) < 0)
3249
		return;
3250 3251
	epoch = ceph_decode_32(&p);
	maplen = ceph_decode_32(&p);
S
Sage Weil 已提交
3252 3253 3254
	dout("handle_map epoch %u len %d\n", epoch, (int)maplen);

	/* do we need it? */
3255
	ceph_monc_got_mdsmap(&mdsc->fsc->client->monc, epoch);
S
Sage Weil 已提交
3256 3257 3258 3259 3260 3261 3262 3263 3264 3265 3266 3267 3268 3269 3270 3271 3272 3273 3274 3275 3276 3277 3278
	mutex_lock(&mdsc->mutex);
	if (mdsc->mdsmap && epoch <= mdsc->mdsmap->m_epoch) {
		dout("handle_map epoch %u <= our %u\n",
		     epoch, mdsc->mdsmap->m_epoch);
		mutex_unlock(&mdsc->mutex);
		return;
	}

	newmap = ceph_mdsmap_decode(&p, end);
	if (IS_ERR(newmap)) {
		err = PTR_ERR(newmap);
		goto bad_unlock;
	}

	/* swap into place */
	if (mdsc->mdsmap) {
		oldmap = mdsc->mdsmap;
		mdsc->mdsmap = newmap;
		check_new_map(mdsc, newmap, oldmap);
		ceph_mdsmap_destroy(oldmap);
	} else {
		mdsc->mdsmap = newmap;  /* first mds map */
	}
3279
	mdsc->fsc->sb->s_maxbytes = mdsc->mdsmap->m_max_file_size;
S
Sage Weil 已提交
3280 3281 3282 3283 3284 3285 3286 3287 3288 3289 3290 3291 3292 3293 3294 3295 3296 3297 3298

	__wake_requests(mdsc, &mdsc->waiting_for_map);

	mutex_unlock(&mdsc->mutex);
	schedule_delayed(mdsc);
	return;

bad_unlock:
	mutex_unlock(&mdsc->mutex);
bad:
	pr_err("error decoding mdsmap %d\n", err);
	return;
}

static struct ceph_connection *con_get(struct ceph_connection *con)
{
	struct ceph_mds_session *s = con->private;

	if (get_session(s)) {
3299
		dout("mdsc con_get %p ok (%d)\n", s, atomic_read(&s->s_ref));
S
Sage Weil 已提交
3300 3301 3302 3303 3304 3305 3306 3307 3308 3309
		return con;
	}
	dout("mdsc con_get %p FAIL\n", s);
	return NULL;
}

static void con_put(struct ceph_connection *con)
{
	struct ceph_mds_session *s = con->private;

3310
	dout("mdsc con_put %p (%d)\n", s, atomic_read(&s->s_ref) - 1);
S
Sage Weil 已提交
3311 3312 3313 3314 3315 3316 3317 3318 3319 3320
	ceph_put_mds_session(s);
}

/*
 * if the client is unresponsive for long enough, the mds will kill
 * the session entirely.
 */
static void peer_reset(struct ceph_connection *con)
{
	struct ceph_mds_session *s = con->private;
3321
	struct ceph_mds_client *mdsc = s->s_mdsc;
S
Sage Weil 已提交
3322

3323 3324
	pr_warning("mds%d closed our session\n", s->s_mds);
	send_mds_reconnect(mdsc, s);
S
Sage Weil 已提交
3325 3326 3327 3328 3329 3330 3331 3332
}

static void dispatch(struct ceph_connection *con, struct ceph_msg *msg)
{
	struct ceph_mds_session *s = con->private;
	struct ceph_mds_client *mdsc = s->s_mdsc;
	int type = le16_to_cpu(msg->hdr.type);

3333 3334 3335 3336 3337 3338 3339
	mutex_lock(&mdsc->mutex);
	if (__verify_registered_session(mdsc, s) < 0) {
		mutex_unlock(&mdsc->mutex);
		goto out;
	}
	mutex_unlock(&mdsc->mutex);

S
Sage Weil 已提交
3340 3341 3342 3343 3344 3345 3346 3347 3348 3349 3350
	switch (type) {
	case CEPH_MSG_MDS_MAP:
		ceph_mdsc_handle_map(mdsc, msg);
		break;
	case CEPH_MSG_CLIENT_SESSION:
		handle_session(s, msg);
		break;
	case CEPH_MSG_CLIENT_REPLY:
		handle_reply(s, msg);
		break;
	case CEPH_MSG_CLIENT_REQUEST_FORWARD:
3351
		handle_forward(mdsc, s, msg);
S
Sage Weil 已提交
3352 3353 3354 3355 3356
		break;
	case CEPH_MSG_CLIENT_CAPS:
		ceph_handle_caps(s, msg);
		break;
	case CEPH_MSG_CLIENT_SNAP:
3357
		ceph_handle_snap(mdsc, s, msg);
S
Sage Weil 已提交
3358 3359
		break;
	case CEPH_MSG_CLIENT_LEASE:
3360
		handle_lease(mdsc, s, msg);
S
Sage Weil 已提交
3361 3362 3363 3364 3365 3366
		break;

	default:
		pr_err("received unknown message type %d %s\n", type,
		       ceph_msg_type_name(type));
	}
3367
out:
S
Sage Weil 已提交
3368 3369 3370
	ceph_msg_put(msg);
}

3371 3372 3373 3374 3375 3376 3377 3378 3379
/*
 * authentication
 */
static int get_authorizer(struct ceph_connection *con,
			  void **buf, int *len, int *proto,
			  void **reply_buf, int *reply_len, int force_new)
{
	struct ceph_mds_session *s = con->private;
	struct ceph_mds_client *mdsc = s->s_mdsc;
3380
	struct ceph_auth_client *ac = mdsc->fsc->client->monc.auth;
3381 3382 3383 3384 3385 3386 3387 3388 3389 3390 3391 3392 3393 3394 3395 3396 3397 3398 3399 3400 3401 3402 3403 3404 3405 3406 3407 3408 3409 3410 3411 3412 3413
	int ret = 0;

	if (force_new && s->s_authorizer) {
		ac->ops->destroy_authorizer(ac, s->s_authorizer);
		s->s_authorizer = NULL;
	}
	if (s->s_authorizer == NULL) {
		if (ac->ops->create_authorizer) {
			ret = ac->ops->create_authorizer(
				ac, CEPH_ENTITY_TYPE_MDS,
				&s->s_authorizer,
				&s->s_authorizer_buf,
				&s->s_authorizer_buf_len,
				&s->s_authorizer_reply_buf,
				&s->s_authorizer_reply_buf_len);
			if (ret)
				return ret;
		}
	}

	*proto = ac->protocol;
	*buf = s->s_authorizer_buf;
	*len = s->s_authorizer_buf_len;
	*reply_buf = s->s_authorizer_reply_buf;
	*reply_len = s->s_authorizer_reply_buf_len;
	return 0;
}


static int verify_authorizer_reply(struct ceph_connection *con, int len)
{
	struct ceph_mds_session *s = con->private;
	struct ceph_mds_client *mdsc = s->s_mdsc;
3414
	struct ceph_auth_client *ac = mdsc->fsc->client->monc.auth;
3415 3416 3417 3418

	return ac->ops->verify_authorizer_reply(ac, s->s_authorizer, len);
}

3419 3420 3421 3422
static int invalidate_authorizer(struct ceph_connection *con)
{
	struct ceph_mds_session *s = con->private;
	struct ceph_mds_client *mdsc = s->s_mdsc;
3423
	struct ceph_auth_client *ac = mdsc->fsc->client->monc.auth;
3424 3425 3426 3427

	if (ac->ops->invalidate_authorizer)
		ac->ops->invalidate_authorizer(ac, CEPH_ENTITY_TYPE_MDS);

3428
	return ceph_monc_validate_auth(&mdsc->fsc->client->monc);
3429 3430
}

3431
static const struct ceph_connection_operations mds_con_ops = {
S
Sage Weil 已提交
3432 3433 3434
	.get = con_get,
	.put = con_put,
	.dispatch = dispatch,
3435 3436
	.get_authorizer = get_authorizer,
	.verify_authorizer_reply = verify_authorizer_reply,
3437
	.invalidate_authorizer = invalidate_authorizer,
S
Sage Weil 已提交
3438 3439 3440 3441
	.peer_reset = peer_reset,
};

/* eof */