glock.c 50.2 KB
Newer Older
D
David Teigland 已提交
1 2
/*
 * Copyright (C) Sistina Software, Inc.  1997-2003 All rights reserved.
3
 * Copyright (C) 2004-2006 Red Hat, Inc.  All rights reserved.
D
David Teigland 已提交
4 5 6
 *
 * This copyrighted material is made available to anyone wishing to use,
 * modify, copy, or redistribute it subject to the terms and conditions
7
 * of the GNU General Public License version 2.
D
David Teigland 已提交
8 9 10 11 12 13 14 15 16 17
 */

#include <linux/sched.h>
#include <linux/slab.h>
#include <linux/spinlock.h>
#include <linux/completion.h>
#include <linux/buffer_head.h>
#include <linux/delay.h>
#include <linux/sort.h>
#include <linux/jhash.h>
S
Steven Whitehouse 已提交
18
#include <linux/kallsyms.h>
19
#include <linux/gfs2_ondisk.h>
20
#include <linux/list.h>
21
#include <linux/lm_interface.h>
D
David Teigland 已提交
22 23 24
#include <asm/uaccess.h>

#include "gfs2.h"
25
#include "incore.h"
D
David Teigland 已提交
26 27 28 29 30 31 32 33
#include "glock.h"
#include "glops.h"
#include "inode.h"
#include "lm.h"
#include "lops.h"
#include "meta_io.h"
#include "quota.h"
#include "super.h"
34
#include "util.h"
D
David Teigland 已提交
35 36 37 38 39 40

struct greedy {
	struct gfs2_holder gr_gh;
	struct work_struct gr_work;
};

41
struct gfs2_gl_hash_bucket {
42
        struct hlist_head hb_list;
43 44
};

D
David Teigland 已提交
45 46
typedef void (*glock_examiner) (struct gfs2_glock * gl);

47
static int gfs2_dump_lockstate(struct gfs2_sbd *sdp);
48
static int dump_glock(struct gfs2_glock *gl);
49
static int dump_inode(struct gfs2_inode *ip);
50

51
#define GFS2_GL_HASH_SHIFT      15
52 53 54
#define GFS2_GL_HASH_SIZE       (1 << GFS2_GL_HASH_SHIFT)
#define GFS2_GL_HASH_MASK       (GFS2_GL_HASH_SIZE - 1)

55
static struct gfs2_gl_hash_bucket gl_hash_table[GFS2_GL_HASH_SIZE];
56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95

/*
 * Despite what you might think, the numbers below are not arbitrary :-)
 * They are taken from the ipv4 routing hash code, which is well tested
 * and thus should be nearly optimal. Later on we might tweek the numbers
 * but for now this should be fine.
 *
 * The reason for putting the locks in a separate array from the list heads
 * is that we can have fewer locks than list heads and save memory. We use
 * the same hash function for both, but with a different hash mask.
 */
#if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_SPINLOCK) || \
	defined(CONFIG_PROVE_LOCKING)

#ifdef CONFIG_LOCKDEP
# define GL_HASH_LOCK_SZ        256
#else
# if NR_CPUS >= 32
#  define GL_HASH_LOCK_SZ       4096
# elif NR_CPUS >= 16
#  define GL_HASH_LOCK_SZ       2048
# elif NR_CPUS >= 8
#  define GL_HASH_LOCK_SZ       1024
# elif NR_CPUS >= 4
#  define GL_HASH_LOCK_SZ       512
# else
#  define GL_HASH_LOCK_SZ       256
# endif
#endif

/* We never want more locks than chains */
#if GFS2_GL_HASH_SIZE < GL_HASH_LOCK_SZ
# undef GL_HASH_LOCK_SZ
# define GL_HASH_LOCK_SZ GFS2_GL_HASH_SIZE
#endif

static rwlock_t gl_hash_locks[GL_HASH_LOCK_SZ];

static inline rwlock_t *gl_lock_addr(unsigned int x)
{
96
	return &gl_hash_locks[x & (GL_HASH_LOCK_SZ-1)];
97 98 99 100 101 102 103
}
#else /* not SMP, so no spinlocks required */
static inline rwlock_t *gl_lock_addr(x)
{
	return NULL;
}
#endif
104

D
David Teigland 已提交
105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138
/**
 * relaxed_state_ok - is a requested lock compatible with the current lock mode?
 * @actual: the current state of the lock
 * @requested: the lock state that was requested by the caller
 * @flags: the modifier flags passed in by the caller
 *
 * Returns: 1 if the locks are compatible, 0 otherwise
 */

static inline int relaxed_state_ok(unsigned int actual, unsigned requested,
				   int flags)
{
	if (actual == requested)
		return 1;

	if (flags & GL_EXACT)
		return 0;

	if (actual == LM_ST_EXCLUSIVE && requested == LM_ST_SHARED)
		return 1;

	if (actual != LM_ST_UNLOCKED && (flags & LM_FLAG_ANY))
		return 1;

	return 0;
}

/**
 * gl_hash() - Turn glock number into hash bucket number
 * @lock: The glock number
 *
 * Returns: The number of the corresponding hash bucket
 */

139 140
static unsigned int gl_hash(const struct gfs2_sbd *sdp,
			    const struct lm_lockname *name)
D
David Teigland 已提交
141 142 143
{
	unsigned int h;

144
	h = jhash(&name->ln_number, sizeof(u64), 0);
D
David Teigland 已提交
145
	h = jhash(&name->ln_type, sizeof(unsigned int), h);
146
	h = jhash(&sdp, sizeof(struct gfs2_sbd *), h);
D
David Teigland 已提交
147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180
	h &= GFS2_GL_HASH_MASK;

	return h;
}

/**
 * glock_free() - Perform a few checks and then release struct gfs2_glock
 * @gl: The glock to release
 *
 * Also calls lock module to release its internal structure for this glock.
 *
 */

static void glock_free(struct gfs2_glock *gl)
{
	struct gfs2_sbd *sdp = gl->gl_sbd;
	struct inode *aspace = gl->gl_aspace;

	gfs2_lm_put_lock(sdp, gl->gl_lock);

	if (aspace)
		gfs2_aspace_put(aspace);

	kmem_cache_free(gfs2_glock_cachep, gl);
}

/**
 * gfs2_glock_hold() - increment reference count on glock
 * @gl: The glock to hold
 *
 */

void gfs2_glock_hold(struct gfs2_glock *gl)
{
181
	atomic_inc(&gl->gl_ref);
D
David Teigland 已提交
182 183 184 185 186 187 188 189 190 191 192
}

/**
 * gfs2_glock_put() - Decrement reference count on glock
 * @gl: The glock to put
 *
 */

int gfs2_glock_put(struct gfs2_glock *gl)
{
	int rv = 0;
193
	struct gfs2_sbd *sdp = gl->gl_sbd;
D
David Teigland 已提交
194

195
	write_lock(gl_lock_addr(gl->gl_hash));
196
	if (atomic_dec_and_test(&gl->gl_ref)) {
197
		hlist_del(&gl->gl_list);
198
		write_unlock(gl_lock_addr(gl->gl_hash));
199
		BUG_ON(spin_is_locked(&gl->gl_spin));
200 201 202 203 204 205
		gfs2_assert(sdp, gl->gl_state == LM_ST_UNLOCKED);
		gfs2_assert(sdp, list_empty(&gl->gl_reclaim));
		gfs2_assert(sdp, list_empty(&gl->gl_holders));
		gfs2_assert(sdp, list_empty(&gl->gl_waiters1));
		gfs2_assert(sdp, list_empty(&gl->gl_waiters2));
		gfs2_assert(sdp, list_empty(&gl->gl_waiters3));
D
David Teigland 已提交
206 207 208 209
		glock_free(gl);
		rv = 1;
		goto out;
	}
210
	write_unlock(gl_lock_addr(gl->gl_hash));
211
out:
D
David Teigland 已提交
212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246
	return rv;
}

/**
 * queue_empty - check to see if a glock's queue is empty
 * @gl: the glock
 * @head: the head of the queue to check
 *
 * This function protects the list in the event that a process already
 * has a holder on the list and is adding a second holder for itself.
 * The glmutex lock is what generally prevents processes from working
 * on the same glock at once, but the special case of adding a second
 * holder for yourself ("recursive" locking) doesn't involve locking
 * glmutex, making the spin lock necessary.
 *
 * Returns: 1 if the queue is empty
 */

static inline int queue_empty(struct gfs2_glock *gl, struct list_head *head)
{
	int empty;
	spin_lock(&gl->gl_spin);
	empty = list_empty(head);
	spin_unlock(&gl->gl_spin);
	return empty;
}

/**
 * search_bucket() - Find struct gfs2_glock by lock number
 * @bucket: the bucket to search
 * @name: The lock name
 *
 * Returns: NULL, or the struct gfs2_glock with the requested number
 */

247
static struct gfs2_glock *search_bucket(unsigned int hash,
248
					const struct gfs2_sbd *sdp,
249
					const struct lm_lockname *name)
D
David Teigland 已提交
250 251
{
	struct gfs2_glock *gl;
252
	struct hlist_node *h;
D
David Teigland 已提交
253

254
	hlist_for_each_entry(gl, h, &gl_hash_table[hash].hb_list, gl_list) {
D
David Teigland 已提交
255 256
		if (!lm_name_equal(&gl->gl_name, name))
			continue;
257 258
		if (gl->gl_sbd != sdp)
			continue;
D
David Teigland 已提交
259

260
		atomic_inc(&gl->gl_ref);
D
David Teigland 已提交
261 262 263 264 265 266 267 268 269 270 271 272 273 274 275

		return gl;
	}

	return NULL;
}

/**
 * gfs2_glock_find() - Find glock by lock number
 * @sdp: The GFS2 superblock
 * @name: The lock name
 *
 * Returns: NULL, or the struct gfs2_glock with the requested number
 */

276
static struct gfs2_glock *gfs2_glock_find(const struct gfs2_sbd *sdp,
277
					  const struct lm_lockname *name)
D
David Teigland 已提交
278
{
279
	unsigned int hash = gl_hash(sdp, name);
D
David Teigland 已提交
280 281
	struct gfs2_glock *gl;

282
	read_lock(gl_lock_addr(hash));
283
	gl = search_bucket(hash, sdp, name);
284
	read_unlock(gl_lock_addr(hash));
D
David Teigland 已提交
285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301

	return gl;
}

/**
 * gfs2_glock_get() - Get a glock, or create one if one doesn't exist
 * @sdp: The GFS2 superblock
 * @number: the lock number
 * @glops: The glock_operations to use
 * @create: If 0, don't create the glock if it doesn't exist
 * @glp: the glock is returned here
 *
 * This does not lock a glock, just finds/creates structures for one.
 *
 * Returns: errno
 */

302
int gfs2_glock_get(struct gfs2_sbd *sdp, u64 number,
303
		   const struct gfs2_glock_operations *glops, int create,
D
David Teigland 已提交
304 305
		   struct gfs2_glock **glp)
{
306
	struct lm_lockname name = { .ln_number = number, .ln_type = glops->go_type };
D
David Teigland 已提交
307
	struct gfs2_glock *gl, *tmp;
308
	unsigned int hash = gl_hash(sdp, &name);
D
David Teigland 已提交
309 310
	int error;

311
	read_lock(gl_lock_addr(hash));
312
	gl = search_bucket(hash, sdp, &name);
313
	read_unlock(gl_lock_addr(hash));
D
David Teigland 已提交
314 315 316 317 318 319 320 321 322 323

	if (gl || !create) {
		*glp = gl;
		return 0;
	}

	gl = kmem_cache_alloc(gfs2_glock_cachep, GFP_KERNEL);
	if (!gl)
		return -ENOMEM;

324
	gl->gl_flags = 0;
D
David Teigland 已提交
325
	gl->gl_name = name;
326
	atomic_set(&gl->gl_ref, 1);
D
David Teigland 已提交
327
	gl->gl_state = LM_ST_UNLOCKED;
328
	gl->gl_hash = hash;
329 330
	gl->gl_owner = NULL;
	gl->gl_ip = 0;
D
David Teigland 已提交
331
	gl->gl_ops = glops;
332 333 334 335 336
	gl->gl_req_gh = NULL;
	gl->gl_req_bh = NULL;
	gl->gl_vn = 0;
	gl->gl_stamp = jiffies;
	gl->gl_object = NULL;
D
David Teigland 已提交
337
	gl->gl_sbd = sdp;
338
	gl->gl_aspace = NULL;
D
David Teigland 已提交
339 340 341 342
	lops_init_le(&gl->gl_le, &gfs2_glock_lops);

	/* If this glock protects actual on-disk data or metadata blocks,
	   create a VFS inode to manage the pages/buffers holding them. */
S
Steven Whitehouse 已提交
343
	if (glops == &gfs2_inode_glops || glops == &gfs2_rgrp_glops) {
D
David Teigland 已提交
344 345 346 347 348 349 350 351 352 353 354
		gl->gl_aspace = gfs2_aspace_get(sdp);
		if (!gl->gl_aspace) {
			error = -ENOMEM;
			goto fail;
		}
	}

	error = gfs2_lm_get_lock(sdp, &name, &gl->gl_lock);
	if (error)
		goto fail_aspace;

355
	write_lock(gl_lock_addr(hash));
356
	tmp = search_bucket(hash, sdp, &name);
D
David Teigland 已提交
357
	if (tmp) {
358
		write_unlock(gl_lock_addr(hash));
D
David Teigland 已提交
359 360 361
		glock_free(gl);
		gl = tmp;
	} else {
362
		hlist_add_head(&gl->gl_list, &gl_hash_table[hash].hb_list);
363
		write_unlock(gl_lock_addr(hash));
D
David Teigland 已提交
364 365 366 367 368 369
	}

	*glp = gl;

	return 0;

370
fail_aspace:
D
David Teigland 已提交
371 372
	if (gl->gl_aspace)
		gfs2_aspace_put(gl->gl_aspace);
373
fail:
374
	kmem_cache_free(gfs2_glock_cachep, gl);
D
David Teigland 已提交
375 376 377 378 379 380 381 382 383 384 385 386
	return error;
}

/**
 * gfs2_holder_init - initialize a struct gfs2_holder in the default way
 * @gl: the glock
 * @state: the state we're requesting
 * @flags: the modifier flags
 * @gh: the holder structure
 *
 */

387
void gfs2_holder_init(struct gfs2_glock *gl, unsigned int state, unsigned flags,
D
David Teigland 已提交
388 389 390 391
		      struct gfs2_holder *gh)
{
	INIT_LIST_HEAD(&gh->gh_list);
	gh->gh_gl = gl;
S
Steven Whitehouse 已提交
392
	gh->gh_ip = (unsigned long)__builtin_return_address(0);
393
	gh->gh_owner = current;
D
David Teigland 已提交
394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415
	gh->gh_state = state;
	gh->gh_flags = flags;
	gh->gh_error = 0;
	gh->gh_iflags = 0;
	init_completion(&gh->gh_wait);

	if (gh->gh_state == LM_ST_EXCLUSIVE)
		gh->gh_flags |= GL_LOCAL_EXCL;

	gfs2_glock_hold(gl);
}

/**
 * gfs2_holder_reinit - reinitialize a struct gfs2_holder so we can requeue it
 * @state: the state we're requesting
 * @flags: the modifier flags
 * @gh: the holder structure
 *
 * Don't mess with the glock.
 *
 */

416
void gfs2_holder_reinit(unsigned int state, unsigned flags, struct gfs2_holder *gh)
D
David Teigland 已提交
417 418
{
	gh->gh_state = state;
419
	gh->gh_flags = flags;
D
David Teigland 已提交
420 421 422 423
	if (gh->gh_state == LM_ST_EXCLUSIVE)
		gh->gh_flags |= GL_LOCAL_EXCL;

	gh->gh_iflags &= 1 << HIF_ALLOCED;
S
Steven Whitehouse 已提交
424
	gh->gh_ip = (unsigned long)__builtin_return_address(0);
D
David Teigland 已提交
425 426 427 428 429 430 431 432 433 434 435 436
}

/**
 * gfs2_holder_uninit - uninitialize a holder structure (drop glock reference)
 * @gh: the holder structure
 *
 */

void gfs2_holder_uninit(struct gfs2_holder *gh)
{
	gfs2_glock_put(gh->gh_gl);
	gh->gh_gl = NULL;
S
Steven Whitehouse 已提交
437
	gh->gh_ip = 0;
D
David Teigland 已提交
438 439 440 441 442 443 444
}

/**
 * gfs2_holder_get - get a struct gfs2_holder structure
 * @gl: the glock
 * @state: the state we're requesting
 * @flags: the modifier flags
445
 * @gfp_flags:
D
David Teigland 已提交
446 447 448 449 450 451 452 453
 *
 * Figure out how big an impact this function has.  Either:
 * 1) Replace it with a cache of structures hanging off the struct gfs2_sbd
 * 2) Leave it like it is
 *
 * Returns: the holder structure, NULL on ENOMEM
 */

454 455 456
static struct gfs2_holder *gfs2_holder_get(struct gfs2_glock *gl,
					   unsigned int state,
					   int flags, gfp_t gfp_flags)
D
David Teigland 已提交
457 458 459 460 461 462 463 464 465
{
	struct gfs2_holder *gh;

	gh = kmalloc(sizeof(struct gfs2_holder), gfp_flags);
	if (!gh)
		return NULL;

	gfs2_holder_init(gl, state, flags, gh);
	set_bit(HIF_ALLOCED, &gh->gh_iflags);
S
Steven Whitehouse 已提交
466
	gh->gh_ip = (unsigned long)__builtin_return_address(0);
D
David Teigland 已提交
467 468 469 470 471 472 473 474 475
	return gh;
}

/**
 * gfs2_holder_put - get rid of a struct gfs2_holder structure
 * @gh: the holder structure
 *
 */

476
static void gfs2_holder_put(struct gfs2_holder *gh)
D
David Teigland 已提交
477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513
{
	gfs2_holder_uninit(gh);
	kfree(gh);
}

/**
 * rq_mutex - process a mutex request in the queue
 * @gh: the glock holder
 *
 * Returns: 1 if the queue is blocked
 */

static int rq_mutex(struct gfs2_holder *gh)
{
	struct gfs2_glock *gl = gh->gh_gl;

	list_del_init(&gh->gh_list);
	/*  gh->gh_error never examined.  */
	set_bit(GLF_LOCK, &gl->gl_flags);
	complete(&gh->gh_wait);

	return 1;
}

/**
 * rq_promote - process a promote request in the queue
 * @gh: the glock holder
 *
 * Acquire a new inter-node lock, or change a lock state to more restrictive.
 *
 * Returns: 1 if the queue is blocked
 */

static int rq_promote(struct gfs2_holder *gh)
{
	struct gfs2_glock *gl = gh->gh_gl;
	struct gfs2_sbd *sdp = gl->gl_sbd;
514
	const struct gfs2_glock_operations *glops = gl->gl_ops;
D
David Teigland 已提交
515 516 517 518 519 520 521 522 523 524 525 526 527 528

	if (!relaxed_state_ok(gl->gl_state, gh->gh_state, gh->gh_flags)) {
		if (list_empty(&gl->gl_holders)) {
			gl->gl_req_gh = gh;
			set_bit(GLF_LOCK, &gl->gl_flags);
			spin_unlock(&gl->gl_spin);

			if (atomic_read(&sdp->sd_reclaim_count) >
			    gfs2_tune_get(sdp, gt_reclaim_limit) &&
			    !(gh->gh_flags & LM_FLAG_PRIORITY)) {
				gfs2_reclaim_glock(sdp);
				gfs2_reclaim_glock(sdp);
			}

529
			glops->go_xmote_th(gl, gh->gh_state, gh->gh_flags);
D
David Teigland 已提交
530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566
			spin_lock(&gl->gl_spin);
		}
		return 1;
	}

	if (list_empty(&gl->gl_holders)) {
		set_bit(HIF_FIRST, &gh->gh_iflags);
		set_bit(GLF_LOCK, &gl->gl_flags);
	} else {
		struct gfs2_holder *next_gh;
		if (gh->gh_flags & GL_LOCAL_EXCL)
			return 1;
		next_gh = list_entry(gl->gl_holders.next, struct gfs2_holder,
				     gh_list);
		if (next_gh->gh_flags & GL_LOCAL_EXCL)
			 return 1;
	}

	list_move_tail(&gh->gh_list, &gl->gl_holders);
	gh->gh_error = 0;
	set_bit(HIF_HOLDER, &gh->gh_iflags);

	complete(&gh->gh_wait);

	return 0;
}

/**
 * rq_demote - process a demote request in the queue
 * @gh: the glock holder
 *
 * Returns: 1 if the queue is blocked
 */

static int rq_demote(struct gfs2_holder *gh)
{
	struct gfs2_glock *gl = gh->gh_gl;
567
	const struct gfs2_glock_operations *glops = gl->gl_ops;
D
David Teigland 已提交
568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616

	if (!list_empty(&gl->gl_holders))
		return 1;

	if (gl->gl_state == gh->gh_state || gl->gl_state == LM_ST_UNLOCKED) {
		list_del_init(&gh->gh_list);
		gh->gh_error = 0;
		spin_unlock(&gl->gl_spin);
		if (test_bit(HIF_DEALLOC, &gh->gh_iflags))
			gfs2_holder_put(gh);
		else
			complete(&gh->gh_wait);
		spin_lock(&gl->gl_spin);
	} else {
		gl->gl_req_gh = gh;
		set_bit(GLF_LOCK, &gl->gl_flags);
		spin_unlock(&gl->gl_spin);

		if (gh->gh_state == LM_ST_UNLOCKED ||
		    gl->gl_state != LM_ST_EXCLUSIVE)
			glops->go_drop_th(gl);
		else
			glops->go_xmote_th(gl, gh->gh_state, gh->gh_flags);

		spin_lock(&gl->gl_spin);
	}

	return 0;
}

/**
 * rq_greedy - process a queued request to drop greedy status
 * @gh: the glock holder
 *
 * Returns: 1 if the queue is blocked
 */

static int rq_greedy(struct gfs2_holder *gh)
{
	struct gfs2_glock *gl = gh->gh_gl;

	list_del_init(&gh->gh_list);
	/*  gh->gh_error never examined.  */
	clear_bit(GLF_GREEDY, &gl->gl_flags);
	spin_unlock(&gl->gl_spin);

	gfs2_holder_uninit(gh);
	kfree(container_of(gh, struct greedy, gr_gh));

617
	spin_lock(&gl->gl_spin);
D
David Teigland 已提交
618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680

	return 0;
}

/**
 * run_queue - process holder structures on a glock
 * @gl: the glock
 *
 */
static void run_queue(struct gfs2_glock *gl)
{
	struct gfs2_holder *gh;
	int blocked = 1;

	for (;;) {
		if (test_bit(GLF_LOCK, &gl->gl_flags))
			break;

		if (!list_empty(&gl->gl_waiters1)) {
			gh = list_entry(gl->gl_waiters1.next,
					struct gfs2_holder, gh_list);

			if (test_bit(HIF_MUTEX, &gh->gh_iflags))
				blocked = rq_mutex(gh);
			else
				gfs2_assert_warn(gl->gl_sbd, 0);

		} else if (!list_empty(&gl->gl_waiters2) &&
			   !test_bit(GLF_SKIP_WAITERS2, &gl->gl_flags)) {
			gh = list_entry(gl->gl_waiters2.next,
					struct gfs2_holder, gh_list);

			if (test_bit(HIF_DEMOTE, &gh->gh_iflags))
				blocked = rq_demote(gh);
			else if (test_bit(HIF_GREEDY, &gh->gh_iflags))
				blocked = rq_greedy(gh);
			else
				gfs2_assert_warn(gl->gl_sbd, 0);

		} else if (!list_empty(&gl->gl_waiters3)) {
			gh = list_entry(gl->gl_waiters3.next,
					struct gfs2_holder, gh_list);

			if (test_bit(HIF_PROMOTE, &gh->gh_iflags))
				blocked = rq_promote(gh);
			else
				gfs2_assert_warn(gl->gl_sbd, 0);

		} else
			break;

		if (blocked)
			break;
	}
}

/**
 * gfs2_glmutex_lock - acquire a local lock on a glock
 * @gl: the glock
 *
 * Gives caller exclusive access to manipulate a glock structure.
 */

681
static void gfs2_glmutex_lock(struct gfs2_glock *gl)
D
David Teigland 已提交
682 683 684 685 686 687 688
{
	struct gfs2_holder gh;

	gfs2_holder_init(gl, 0, 0, &gh);
	set_bit(HIF_MUTEX, &gh.gh_iflags);

	spin_lock(&gl->gl_spin);
689
	if (test_and_set_bit(GLF_LOCK, &gl->gl_flags)) {
D
David Teigland 已提交
690
		list_add_tail(&gh.gh_list, &gl->gl_waiters1);
691
	} else {
692 693
		gl->gl_owner = current;
		gl->gl_ip = (unsigned long)__builtin_return_address(0);
D
David Teigland 已提交
694
		complete(&gh.gh_wait);
695
	}
D
David Teigland 已提交
696 697 698 699 700 701 702 703 704 705 706 707 708
	spin_unlock(&gl->gl_spin);

	wait_for_completion(&gh.gh_wait);
	gfs2_holder_uninit(&gh);
}

/**
 * gfs2_glmutex_trylock - try to acquire a local lock on a glock
 * @gl: the glock
 *
 * Returns: 1 if the glock is acquired
 */

709
static int gfs2_glmutex_trylock(struct gfs2_glock *gl)
D
David Teigland 已提交
710 711 712 713
{
	int acquired = 1;

	spin_lock(&gl->gl_spin);
714
	if (test_and_set_bit(GLF_LOCK, &gl->gl_flags)) {
D
David Teigland 已提交
715
		acquired = 0;
716
	} else {
717 718 719
		gl->gl_owner = current;
		gl->gl_ip = (unsigned long)__builtin_return_address(0);
	}
D
David Teigland 已提交
720 721 722 723 724 725 726 727 728 729 730
	spin_unlock(&gl->gl_spin);

	return acquired;
}

/**
 * gfs2_glmutex_unlock - release a local lock on a glock
 * @gl: the glock
 *
 */

731
static void gfs2_glmutex_unlock(struct gfs2_glock *gl)
D
David Teigland 已提交
732 733 734
{
	spin_lock(&gl->gl_spin);
	clear_bit(GLF_LOCK, &gl->gl_flags);
735 736
	gl->gl_owner = NULL;
	gl->gl_ip = 0;
D
David Teigland 已提交
737
	run_queue(gl);
738
	BUG_ON(!spin_is_locked(&gl->gl_spin));
D
David Teigland 已提交
739 740 741 742 743 744 745 746
	spin_unlock(&gl->gl_spin);
}

/**
 * handle_callback - add a demote request to a lock's queue
 * @gl: the glock
 * @state: the state the caller wants us to change to
 *
747
 * Note: This may fail sliently if we are out of memory.
D
David Teigland 已提交
748 749 750 751 752 753
 */

static void handle_callback(struct gfs2_glock *gl, unsigned int state)
{
	struct gfs2_holder *gh, *new_gh = NULL;

754
restart:
D
David Teigland 已提交
755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771
	spin_lock(&gl->gl_spin);

	list_for_each_entry(gh, &gl->gl_waiters2, gh_list) {
		if (test_bit(HIF_DEMOTE, &gh->gh_iflags) &&
		    gl->gl_req_gh != gh) {
			if (gh->gh_state != state)
				gh->gh_state = LM_ST_UNLOCKED;
			goto out;
		}
	}

	if (new_gh) {
		list_add_tail(&new_gh->gh_list, &gl->gl_waiters2);
		new_gh = NULL;
	} else {
		spin_unlock(&gl->gl_spin);

772 773 774
		new_gh = gfs2_holder_get(gl, state, LM_FLAG_TRY, GFP_KERNEL);
		if (!new_gh)
			return;
D
David Teigland 已提交
775 776 777 778 779 780
		set_bit(HIF_DEMOTE, &new_gh->gh_iflags);
		set_bit(HIF_DEALLOC, &new_gh->gh_iflags);

		goto restart;
	}

781
out:
D
David Teigland 已提交
782 783 784 785 786 787
	spin_unlock(&gl->gl_spin);

	if (new_gh)
		gfs2_holder_put(new_gh);
}

788 789 790 791 792 793 794 795 796 797 798
void gfs2_glock_inode_squish(struct inode *inode)
{
	struct gfs2_holder gh;
	struct gfs2_glock *gl = GFS2_I(inode)->i_gl;
	gfs2_holder_init(gl, LM_ST_UNLOCKED, 0, &gh);
	set_bit(HIF_DEMOTE, &gh.gh_iflags);
	spin_lock(&gl->gl_spin);
	gfs2_assert(inode->i_sb->s_fs_info, list_empty(&gl->gl_holders));
	list_add_tail(&gh.gh_list, &gl->gl_waiters2);
	run_queue(gl);
	spin_unlock(&gl->gl_spin);
799
	wait_for_completion(&gh.gh_wait);
800 801 802
	gfs2_holder_uninit(&gh);
}

D
David Teigland 已提交
803 804 805 806 807 808 809 810 811 812 813 814 815 816 817
/**
 * state_change - record that the glock is now in a different state
 * @gl: the glock
 * @new_state the new state
 *
 */

static void state_change(struct gfs2_glock *gl, unsigned int new_state)
{
	int held1, held2;

	held1 = (gl->gl_state != LM_ST_UNLOCKED);
	held2 = (new_state != LM_ST_UNLOCKED);

	if (held1 != held2) {
818
		if (held2)
D
David Teigland 已提交
819
			gfs2_glock_hold(gl);
820
		else
D
David Teigland 已提交
821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836
			gfs2_glock_put(gl);
	}

	gl->gl_state = new_state;
}

/**
 * xmote_bh - Called after the lock module is done acquiring a lock
 * @gl: The glock in question
 * @ret: the int returned from the lock module
 *
 */

static void xmote_bh(struct gfs2_glock *gl, unsigned int ret)
{
	struct gfs2_sbd *sdp = gl->gl_sbd;
837
	const struct gfs2_glock_operations *glops = gl->gl_ops;
D
David Teigland 已提交
838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870
	struct gfs2_holder *gh = gl->gl_req_gh;
	int prev_state = gl->gl_state;
	int op_done = 1;

	gfs2_assert_warn(sdp, test_bit(GLF_LOCK, &gl->gl_flags));
	gfs2_assert_warn(sdp, queue_empty(gl, &gl->gl_holders));
	gfs2_assert_warn(sdp, !(ret & LM_OUT_ASYNC));

	state_change(gl, ret & LM_OUT_ST_MASK);

	if (prev_state != LM_ST_UNLOCKED && !(ret & LM_OUT_CACHEABLE)) {
		if (glops->go_inval)
			glops->go_inval(gl, DIO_METADATA | DIO_DATA);
	} else if (gl->gl_state == LM_ST_DEFERRED) {
		/* We might not want to do this here.
		   Look at moving to the inode glops. */
		if (glops->go_inval)
			glops->go_inval(gl, DIO_DATA);
	}

	/*  Deal with each possible exit condition  */

	if (!gh)
		gl->gl_stamp = jiffies;
	else if (unlikely(test_bit(SDF_SHUTDOWN, &sdp->sd_flags))) {
		spin_lock(&gl->gl_spin);
		list_del_init(&gh->gh_list);
		gh->gh_error = -EIO;
		spin_unlock(&gl->gl_spin);
	} else if (test_bit(HIF_DEMOTE, &gh->gh_iflags)) {
		spin_lock(&gl->gl_spin);
		list_del_init(&gh->gh_list);
		if (gl->gl_state == gh->gh_state ||
871
		    gl->gl_state == LM_ST_UNLOCKED) {
D
David Teigland 已提交
872
			gh->gh_error = 0;
873
		} else {
D
David Teigland 已提交
874 875 876 877 878 879 880 881
			if (gfs2_assert_warn(sdp, gh->gh_flags &
					(LM_FLAG_TRY | LM_FLAG_TRY_1CB)) == -1)
				fs_warn(sdp, "ret = 0x%.8X\n", ret);
			gh->gh_error = GLR_TRYFAILED;
		}
		spin_unlock(&gl->gl_spin);

		if (ret & LM_OUT_CANCELED)
S
Steven Whitehouse 已提交
882
			handle_callback(gl, LM_ST_UNLOCKED);
D
David Teigland 已提交
883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944

	} else if (ret & LM_OUT_CANCELED) {
		spin_lock(&gl->gl_spin);
		list_del_init(&gh->gh_list);
		gh->gh_error = GLR_CANCELED;
		spin_unlock(&gl->gl_spin);

	} else if (relaxed_state_ok(gl->gl_state, gh->gh_state, gh->gh_flags)) {
		spin_lock(&gl->gl_spin);
		list_move_tail(&gh->gh_list, &gl->gl_holders);
		gh->gh_error = 0;
		set_bit(HIF_HOLDER, &gh->gh_iflags);
		spin_unlock(&gl->gl_spin);

		set_bit(HIF_FIRST, &gh->gh_iflags);

		op_done = 0;

	} else if (gh->gh_flags & (LM_FLAG_TRY | LM_FLAG_TRY_1CB)) {
		spin_lock(&gl->gl_spin);
		list_del_init(&gh->gh_list);
		gh->gh_error = GLR_TRYFAILED;
		spin_unlock(&gl->gl_spin);

	} else {
		if (gfs2_assert_withdraw(sdp, 0) == -1)
			fs_err(sdp, "ret = 0x%.8X\n", ret);
	}

	if (glops->go_xmote_bh)
		glops->go_xmote_bh(gl);

	if (op_done) {
		spin_lock(&gl->gl_spin);
		gl->gl_req_gh = NULL;
		gl->gl_req_bh = NULL;
		clear_bit(GLF_LOCK, &gl->gl_flags);
		run_queue(gl);
		spin_unlock(&gl->gl_spin);
	}

	gfs2_glock_put(gl);

	if (gh) {
		if (test_bit(HIF_DEALLOC, &gh->gh_iflags))
			gfs2_holder_put(gh);
		else
			complete(&gh->gh_wait);
	}
}

/**
 * gfs2_glock_xmote_th - Call into the lock module to acquire or change a glock
 * @gl: The glock in question
 * @state: the requested state
 * @flags: modifier flags to the lock call
 *
 */

void gfs2_glock_xmote_th(struct gfs2_glock *gl, unsigned int state, int flags)
{
	struct gfs2_sbd *sdp = gl->gl_sbd;
945
	const struct gfs2_glock_operations *glops = gl->gl_ops;
D
David Teigland 已提交
946 947 948 949 950 951 952 953 954 955
	int lck_flags = flags & (LM_FLAG_TRY | LM_FLAG_TRY_1CB |
				 LM_FLAG_NOEXP | LM_FLAG_ANY |
				 LM_FLAG_PRIORITY);
	unsigned int lck_ret;

	gfs2_assert_warn(sdp, test_bit(GLF_LOCK, &gl->gl_flags));
	gfs2_assert_warn(sdp, queue_empty(gl, &gl->gl_holders));
	gfs2_assert_warn(sdp, state != LM_ST_UNLOCKED);
	gfs2_assert_warn(sdp, state != gl->gl_state);

S
Steven Whitehouse 已提交
956 957
	if (gl->gl_state == LM_ST_EXCLUSIVE && glops->go_sync)
		glops->go_sync(gl, DIO_METADATA | DIO_DATA | DIO_RELEASE);
D
David Teigland 已提交
958 959 960 961

	gfs2_glock_hold(gl);
	gl->gl_req_bh = xmote_bh;

962
	lck_ret = gfs2_lm_lock(sdp, gl->gl_lock, gl->gl_state, state, lck_flags);
D
David Teigland 已提交
963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985

	if (gfs2_assert_withdraw(sdp, !(lck_ret & LM_OUT_ERROR)))
		return;

	if (lck_ret & LM_OUT_ASYNC)
		gfs2_assert_warn(sdp, lck_ret == LM_OUT_ASYNC);
	else
		xmote_bh(gl, lck_ret);
}

/**
 * drop_bh - Called after a lock module unlock completes
 * @gl: the glock
 * @ret: the return status
 *
 * Doesn't wake up the process waiting on the struct gfs2_holder (if any)
 * Doesn't drop the reference on the glock the top half took out
 *
 */

static void drop_bh(struct gfs2_glock *gl, unsigned int ret)
{
	struct gfs2_sbd *sdp = gl->gl_sbd;
986
	const struct gfs2_glock_operations *glops = gl->gl_ops;
D
David Teigland 已提交
987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035
	struct gfs2_holder *gh = gl->gl_req_gh;

	clear_bit(GLF_PREFETCH, &gl->gl_flags);

	gfs2_assert_warn(sdp, test_bit(GLF_LOCK, &gl->gl_flags));
	gfs2_assert_warn(sdp, queue_empty(gl, &gl->gl_holders));
	gfs2_assert_warn(sdp, !ret);

	state_change(gl, LM_ST_UNLOCKED);

	if (glops->go_inval)
		glops->go_inval(gl, DIO_METADATA | DIO_DATA);

	if (gh) {
		spin_lock(&gl->gl_spin);
		list_del_init(&gh->gh_list);
		gh->gh_error = 0;
		spin_unlock(&gl->gl_spin);
	}

	if (glops->go_drop_bh)
		glops->go_drop_bh(gl);

	spin_lock(&gl->gl_spin);
	gl->gl_req_gh = NULL;
	gl->gl_req_bh = NULL;
	clear_bit(GLF_LOCK, &gl->gl_flags);
	run_queue(gl);
	spin_unlock(&gl->gl_spin);

	gfs2_glock_put(gl);

	if (gh) {
		if (test_bit(HIF_DEALLOC, &gh->gh_iflags))
			gfs2_holder_put(gh);
		else
			complete(&gh->gh_wait);
	}
}

/**
 * gfs2_glock_drop_th - call into the lock module to unlock a lock
 * @gl: the glock
 *
 */

void gfs2_glock_drop_th(struct gfs2_glock *gl)
{
	struct gfs2_sbd *sdp = gl->gl_sbd;
1036
	const struct gfs2_glock_operations *glops = gl->gl_ops;
D
David Teigland 已提交
1037 1038 1039 1040 1041 1042
	unsigned int ret;

	gfs2_assert_warn(sdp, test_bit(GLF_LOCK, &gl->gl_flags));
	gfs2_assert_warn(sdp, queue_empty(gl, &gl->gl_holders));
	gfs2_assert_warn(sdp, gl->gl_state != LM_ST_UNLOCKED);

S
Steven Whitehouse 已提交
1043 1044
	if (gl->gl_state == LM_ST_EXCLUSIVE && glops->go_sync)
		glops->go_sync(gl, DIO_METADATA | DIO_DATA | DIO_RELEASE);
D
David Teigland 已提交
1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075

	gfs2_glock_hold(gl);
	gl->gl_req_bh = drop_bh;

	ret = gfs2_lm_unlock(sdp, gl->gl_lock, gl->gl_state);

	if (gfs2_assert_withdraw(sdp, !(ret & LM_OUT_ERROR)))
		return;

	if (!ret)
		drop_bh(gl, ret);
	else
		gfs2_assert_warn(sdp, ret == LM_OUT_ASYNC);
}

/**
 * do_cancels - cancel requests for locks stuck waiting on an expire flag
 * @gh: the LM_FLAG_PRIORITY holder waiting to acquire the lock
 *
 * Don't cancel GL_NOCANCEL requests.
 */

static void do_cancels(struct gfs2_holder *gh)
{
	struct gfs2_glock *gl = gh->gh_gl;

	spin_lock(&gl->gl_spin);

	while (gl->gl_req_gh != gh &&
	       !test_bit(HIF_HOLDER, &gh->gh_iflags) &&
	       !list_empty(&gh->gh_list)) {
S
Steven Whitehouse 已提交
1076 1077
		if (gl->gl_req_bh && !(gl->gl_req_gh &&
				     (gl->gl_req_gh->gh_flags & GL_NOCANCEL))) {
D
David Teigland 已提交
1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102
			spin_unlock(&gl->gl_spin);
			gfs2_lm_cancel(gl->gl_sbd, gl->gl_lock);
			msleep(100);
			spin_lock(&gl->gl_spin);
		} else {
			spin_unlock(&gl->gl_spin);
			msleep(100);
			spin_lock(&gl->gl_spin);
		}
	}

	spin_unlock(&gl->gl_spin);
}

/**
 * glock_wait_internal - wait on a glock acquisition
 * @gh: the glock holder
 *
 * Returns: 0 on success
 */

static int glock_wait_internal(struct gfs2_holder *gh)
{
	struct gfs2_glock *gl = gh->gh_gl;
	struct gfs2_sbd *sdp = gl->gl_sbd;
1103
	const struct gfs2_glock_operations *glops = gl->gl_ops;
D
David Teigland 已提交
1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130

	if (test_bit(HIF_ABORTED, &gh->gh_iflags))
		return -EIO;

	if (gh->gh_flags & (LM_FLAG_TRY | LM_FLAG_TRY_1CB)) {
		spin_lock(&gl->gl_spin);
		if (gl->gl_req_gh != gh &&
		    !test_bit(HIF_HOLDER, &gh->gh_iflags) &&
		    !list_empty(&gh->gh_list)) {
			list_del_init(&gh->gh_list);
			gh->gh_error = GLR_TRYFAILED;
			run_queue(gl);
			spin_unlock(&gl->gl_spin);
			return gh->gh_error;
		}
		spin_unlock(&gl->gl_spin);
	}

	if (gh->gh_flags & LM_FLAG_PRIORITY)
		do_cancels(gh);

	wait_for_completion(&gh->gh_wait);

	if (gh->gh_error)
		return gh->gh_error;

	gfs2_assert_withdraw(sdp, test_bit(HIF_HOLDER, &gh->gh_iflags));
1131
	gfs2_assert_withdraw(sdp, relaxed_state_ok(gl->gl_state, gh->gh_state,
D
David Teigland 已提交
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 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180
						   gh->gh_flags));

	if (test_bit(HIF_FIRST, &gh->gh_iflags)) {
		gfs2_assert_warn(sdp, test_bit(GLF_LOCK, &gl->gl_flags));

		if (glops->go_lock) {
			gh->gh_error = glops->go_lock(gh);
			if (gh->gh_error) {
				spin_lock(&gl->gl_spin);
				list_del_init(&gh->gh_list);
				spin_unlock(&gl->gl_spin);
			}
		}

		spin_lock(&gl->gl_spin);
		gl->gl_req_gh = NULL;
		gl->gl_req_bh = NULL;
		clear_bit(GLF_LOCK, &gl->gl_flags);
		run_queue(gl);
		spin_unlock(&gl->gl_spin);
	}

	return gh->gh_error;
}

static inline struct gfs2_holder *
find_holder_by_owner(struct list_head *head, struct task_struct *owner)
{
	struct gfs2_holder *gh;

	list_for_each_entry(gh, head, gh_list) {
		if (gh->gh_owner == owner)
			return gh;
	}

	return NULL;
}

/**
 * add_to_queue - Add a holder to the wait queue (but look for recursion)
 * @gh: the holder structure to add
 *
 */

static void add_to_queue(struct gfs2_holder *gh)
{
	struct gfs2_glock *gl = gh->gh_gl;
	struct gfs2_holder *existing;

1181 1182
	BUG_ON(!gh->gh_owner);

D
David Teigland 已提交
1183 1184
	existing = find_holder_by_owner(&gl->gl_holders, gh->gh_owner);
	if (existing) {
1185
		print_symbol(KERN_WARNING "original: %s\n", existing->gh_ip);
1186
		printk(KERN_INFO "pid : %d\n", existing->gh_owner->pid);
1187
		printk(KERN_INFO "lock type : %d lock state : %d\n",
1188
				existing->gh_gl->gl_name.ln_type, existing->gh_gl->gl_state);
1189
		print_symbol(KERN_WARNING "new: %s\n", gh->gh_ip);
1190
		printk(KERN_INFO "pid : %d\n", gh->gh_owner->pid);
1191
		printk(KERN_INFO "lock type : %d lock state : %d\n",
1192
				gl->gl_name.ln_type, gl->gl_state);
1193
		BUG();
D
David Teigland 已提交
1194 1195 1196 1197
	}

	existing = find_holder_by_owner(&gl->gl_waiters3, gh->gh_owner);
	if (existing) {
1198 1199 1200
		print_symbol(KERN_WARNING "original: %s\n", existing->gh_ip);
		print_symbol(KERN_WARNING "new: %s\n", gh->gh_ip);
		BUG();
D
David Teigland 已提交
1201 1202 1203 1204 1205
	}

	if (gh->gh_flags & LM_FLAG_PRIORITY)
		list_add(&gh->gh_list, &gl->gl_waiters3);
	else
1206
		list_add_tail(&gh->gh_list, &gl->gl_waiters3);
D
David Teigland 已提交
1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223
}

/**
 * gfs2_glock_nq - enqueue a struct gfs2_holder onto a glock (acquire a glock)
 * @gh: the holder structure
 *
 * if (gh->gh_flags & GL_ASYNC), this never returns an error
 *
 * Returns: 0, GLR_TRYFAILED, or errno on failure
 */

int gfs2_glock_nq(struct gfs2_holder *gh)
{
	struct gfs2_glock *gl = gh->gh_gl;
	struct gfs2_sbd *sdp = gl->gl_sbd;
	int error = 0;

1224
restart:
D
David Teigland 已提交
1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239
	if (unlikely(test_bit(SDF_SHUTDOWN, &sdp->sd_flags))) {
		set_bit(HIF_ABORTED, &gh->gh_iflags);
		return -EIO;
	}

	set_bit(HIF_PROMOTE, &gh->gh_iflags);

	spin_lock(&gl->gl_spin);
	add_to_queue(gh);
	run_queue(gl);
	spin_unlock(&gl->gl_spin);

	if (!(gh->gh_flags & GL_ASYNC)) {
		error = glock_wait_internal(gh);
		if (error == GLR_CANCELED) {
1240
			msleep(100);
D
David Teigland 已提交
1241 1242 1243 1244 1245 1246
			goto restart;
		}
	}

	clear_bit(GLF_PREFETCH, &gl->gl_flags);

1247 1248 1249
	if (error == GLR_TRYFAILED && (gh->gh_flags & GL_DUMP))
		dump_glock(gl);

D
David Teigland 已提交
1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271
	return error;
}

/**
 * gfs2_glock_poll - poll to see if an async request has been completed
 * @gh: the holder
 *
 * Returns: 1 if the request is ready to be gfs2_glock_wait()ed on
 */

int gfs2_glock_poll(struct gfs2_holder *gh)
{
	struct gfs2_glock *gl = gh->gh_gl;
	int ready = 0;

	spin_lock(&gl->gl_spin);

	if (test_bit(HIF_HOLDER, &gh->gh_iflags))
		ready = 1;
	else if (list_empty(&gh->gh_list)) {
		if (gh->gh_error == GLR_CANCELED) {
			spin_unlock(&gl->gl_spin);
1272
			msleep(100);
D
David Teigland 已提交
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
			if (gfs2_glock_nq(gh))
				return 1;
			return 0;
		} else
			ready = 1;
	}

	spin_unlock(&gl->gl_spin);

	return ready;
}

/**
 * gfs2_glock_wait - wait for a lock acquisition that ended in a GLR_ASYNC
 * @gh: the holder structure
 *
 * Returns: 0, GLR_TRYFAILED, or errno on failure
 */

int gfs2_glock_wait(struct gfs2_holder *gh)
{
	int error;

	error = glock_wait_internal(gh);
	if (error == GLR_CANCELED) {
1298
		msleep(100);
D
David Teigland 已提交
1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314
		gh->gh_flags &= ~GL_ASYNC;
		error = gfs2_glock_nq(gh);
	}

	return error;
}

/**
 * gfs2_glock_dq - dequeue a struct gfs2_holder from a glock (release a glock)
 * @gh: the glock holder
 *
 */

void gfs2_glock_dq(struct gfs2_holder *gh)
{
	struct gfs2_glock *gl = gh->gh_gl;
1315
	const struct gfs2_glock_operations *glops = gl->gl_ops;
D
David Teigland 已提交
1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348

	if (gh->gh_flags & GL_NOCACHE)
		handle_callback(gl, LM_ST_UNLOCKED);

	gfs2_glmutex_lock(gl);

	spin_lock(&gl->gl_spin);
	list_del_init(&gh->gh_list);

	if (list_empty(&gl->gl_holders)) {
		spin_unlock(&gl->gl_spin);

		if (glops->go_unlock)
			glops->go_unlock(gh);

		gl->gl_stamp = jiffies;

		spin_lock(&gl->gl_spin);
	}

	clear_bit(GLF_LOCK, &gl->gl_flags);
	run_queue(gl);
	spin_unlock(&gl->gl_spin);
}

/**
 * gfs2_glock_prefetch - Try to prefetch a glock
 * @gl: the glock
 * @state: the state to prefetch in
 * @flags: flags passed to go_xmote_th()
 *
 */

1349 1350
static void gfs2_glock_prefetch(struct gfs2_glock *gl, unsigned int state,
				int flags)
D
David Teigland 已提交
1351
{
1352
	const struct gfs2_glock_operations *glops = gl->gl_ops;
D
David Teigland 已提交
1353 1354 1355

	spin_lock(&gl->gl_spin);

S
Steven Whitehouse 已提交
1356 1357
	if (test_bit(GLF_LOCK, &gl->gl_flags) || !list_empty(&gl->gl_holders) ||
	    !list_empty(&gl->gl_waiters1) || !list_empty(&gl->gl_waiters2) ||
D
David Teigland 已提交
1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372
	    !list_empty(&gl->gl_waiters3) ||
	    relaxed_state_ok(gl->gl_state, state, flags)) {
		spin_unlock(&gl->gl_spin);
		return;
	}

	set_bit(GLF_PREFETCH, &gl->gl_flags);
	set_bit(GLF_LOCK, &gl->gl_flags);
	spin_unlock(&gl->gl_spin);

	glops->go_xmote_th(gl, state, flags);
}

static void greedy_work(void *data)
{
1373
	struct greedy *gr = data;
D
David Teigland 已提交
1374 1375
	struct gfs2_holder *gh = &gr->gr_gh;
	struct gfs2_glock *gl = gh->gh_gl;
1376
	const struct gfs2_glock_operations *glops = gl->gl_ops;
D
David Teigland 已提交
1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411

	clear_bit(GLF_SKIP_WAITERS2, &gl->gl_flags);

	if (glops->go_greedy)
		glops->go_greedy(gl);

	spin_lock(&gl->gl_spin);

	if (list_empty(&gl->gl_waiters2)) {
		clear_bit(GLF_GREEDY, &gl->gl_flags);
		spin_unlock(&gl->gl_spin);
		gfs2_holder_uninit(gh);
		kfree(gr);
	} else {
		gfs2_glock_hold(gl);
		list_add_tail(&gh->gh_list, &gl->gl_waiters2);
		run_queue(gl);
		spin_unlock(&gl->gl_spin);
		gfs2_glock_put(gl);
	}
}

/**
 * gfs2_glock_be_greedy -
 * @gl:
 * @time:
 *
 * Returns: 0 if go_greedy will be called, 1 otherwise
 */

int gfs2_glock_be_greedy(struct gfs2_glock *gl, unsigned int time)
{
	struct greedy *gr;
	struct gfs2_holder *gh;

1412
	if (!time || gl->gl_sbd->sd_args.ar_localcaching ||
D
David Teigland 已提交
1413 1414 1415 1416 1417 1418 1419 1420 1421 1422
	    test_and_set_bit(GLF_GREEDY, &gl->gl_flags))
		return 1;

	gr = kmalloc(sizeof(struct greedy), GFP_KERNEL);
	if (!gr) {
		clear_bit(GLF_GREEDY, &gl->gl_flags);
		return 1;
	}
	gh = &gr->gr_gh;

1423
	gfs2_holder_init(gl, 0, 0, gh);
D
David Teigland 已提交
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
	set_bit(HIF_GREEDY, &gh->gh_iflags);
	INIT_WORK(&gr->gr_work, greedy_work, gr);

	set_bit(GLF_SKIP_WAITERS2, &gl->gl_flags);
	schedule_delayed_work(&gr->gr_work, time);

	return 0;
}

/**
 * gfs2_glock_dq_uninit - dequeue a holder from a glock and initialize it
 * @gh: the holder structure
 *
 */

void gfs2_glock_dq_uninit(struct gfs2_holder *gh)
{
	gfs2_glock_dq(gh);
	gfs2_holder_uninit(gh);
}

/**
 * gfs2_glock_nq_num - acquire a glock based on lock number
 * @sdp: the filesystem
 * @number: the lock number
 * @glops: the glock operations for the type of glock
 * @state: the state to acquire the glock in
 * @flags: modifier flags for the aquisition
 * @gh: the struct gfs2_holder
 *
 * Returns: errno
 */

1457
int gfs2_glock_nq_num(struct gfs2_sbd *sdp, u64 number,
1458 1459
		      const struct gfs2_glock_operations *glops,
		      unsigned int state, int flags, struct gfs2_holder *gh)
D
David Teigland 已提交
1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481
{
	struct gfs2_glock *gl;
	int error;

	error = gfs2_glock_get(sdp, number, glops, CREATE, &gl);
	if (!error) {
		error = gfs2_glock_nq_init(gl, state, flags, gh);
		gfs2_glock_put(gl);
	}

	return error;
}

/**
 * glock_compare - Compare two struct gfs2_glock structures for sorting
 * @arg_a: the first structure
 * @arg_b: the second structure
 *
 */

static int glock_compare(const void *arg_a, const void *arg_b)
{
1482 1483 1484 1485
	const struct gfs2_holder *gh_a = *(const struct gfs2_holder **)arg_a;
	const struct gfs2_holder *gh_b = *(const struct gfs2_holder **)arg_b;
	const struct lm_lockname *a = &gh_a->gh_gl->gl_name;
	const struct lm_lockname *b = &gh_b->gh_gl->gl_name;
D
David Teigland 已提交
1486 1487

	if (a->ln_number > b->ln_number)
1488 1489 1490 1491 1492 1493 1494 1495
		return 1;
	if (a->ln_number < b->ln_number)
		return -1;
	if (gh_a->gh_state == LM_ST_SHARED && gh_b->gh_state == LM_ST_EXCLUSIVE)
		return 1;
	if (!(gh_a->gh_flags & GL_LOCAL_EXCL) && (gh_b->gh_flags & GL_LOCAL_EXCL))
		return 1;
	return 0;
D
David Teigland 已提交
1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 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 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648
}

/**
 * nq_m_sync - synchonously acquire more than one glock in deadlock free order
 * @num_gh: the number of structures
 * @ghs: an array of struct gfs2_holder structures
 *
 * Returns: 0 on success (all glocks acquired),
 *          errno on failure (no glocks acquired)
 */

static int nq_m_sync(unsigned int num_gh, struct gfs2_holder *ghs,
		     struct gfs2_holder **p)
{
	unsigned int x;
	int error = 0;

	for (x = 0; x < num_gh; x++)
		p[x] = &ghs[x];

	sort(p, num_gh, sizeof(struct gfs2_holder *), glock_compare, NULL);

	for (x = 0; x < num_gh; x++) {
		p[x]->gh_flags &= ~(LM_FLAG_TRY | GL_ASYNC);

		error = gfs2_glock_nq(p[x]);
		if (error) {
			while (x--)
				gfs2_glock_dq(p[x]);
			break;
		}
	}

	return error;
}

/**
 * gfs2_glock_nq_m - acquire multiple glocks
 * @num_gh: the number of structures
 * @ghs: an array of struct gfs2_holder structures
 *
 * Figure out how big an impact this function has.  Either:
 * 1) Replace this code with code that calls gfs2_glock_prefetch()
 * 2) Forget async stuff and just call nq_m_sync()
 * 3) Leave it like it is
 *
 * Returns: 0 on success (all glocks acquired),
 *          errno on failure (no glocks acquired)
 */

int gfs2_glock_nq_m(unsigned int num_gh, struct gfs2_holder *ghs)
{
	int *e;
	unsigned int x;
	int borked = 0, serious = 0;
	int error = 0;

	if (!num_gh)
		return 0;

	if (num_gh == 1) {
		ghs->gh_flags &= ~(LM_FLAG_TRY | GL_ASYNC);
		return gfs2_glock_nq(ghs);
	}

	e = kcalloc(num_gh, sizeof(struct gfs2_holder *), GFP_KERNEL);
	if (!e)
		return -ENOMEM;

	for (x = 0; x < num_gh; x++) {
		ghs[x].gh_flags |= LM_FLAG_TRY | GL_ASYNC;
		error = gfs2_glock_nq(&ghs[x]);
		if (error) {
			borked = 1;
			serious = error;
			num_gh = x;
			break;
		}
	}

	for (x = 0; x < num_gh; x++) {
		error = e[x] = glock_wait_internal(&ghs[x]);
		if (error) {
			borked = 1;
			if (error != GLR_TRYFAILED && error != GLR_CANCELED)
				serious = error;
		}
	}

	if (!borked) {
		kfree(e);
		return 0;
	}

	for (x = 0; x < num_gh; x++)
		if (!e[x])
			gfs2_glock_dq(&ghs[x]);

	if (serious)
		error = serious;
	else {
		for (x = 0; x < num_gh; x++)
			gfs2_holder_reinit(ghs[x].gh_state, ghs[x].gh_flags,
					  &ghs[x]);
		error = nq_m_sync(num_gh, ghs, (struct gfs2_holder **)e);
	}

	kfree(e);

	return error;
}

/**
 * gfs2_glock_dq_m - release multiple glocks
 * @num_gh: the number of structures
 * @ghs: an array of struct gfs2_holder structures
 *
 */

void gfs2_glock_dq_m(unsigned int num_gh, struct gfs2_holder *ghs)
{
	unsigned int x;

	for (x = 0; x < num_gh; x++)
		gfs2_glock_dq(&ghs[x]);
}

/**
 * gfs2_glock_dq_uninit_m - release multiple glocks
 * @num_gh: the number of structures
 * @ghs: an array of struct gfs2_holder structures
 *
 */

void gfs2_glock_dq_uninit_m(unsigned int num_gh, struct gfs2_holder *ghs)
{
	unsigned int x;

	for (x = 0; x < num_gh; x++)
		gfs2_glock_dq_uninit(&ghs[x]);
}

/**
 * gfs2_glock_prefetch_num - prefetch a glock based on lock number
 * @sdp: the filesystem
 * @number: the lock number
 * @glops: the glock operations for the type of glock
 * @state: the state to acquire the glock in
 * @flags: modifier flags for the aquisition
 *
 * Returns: errno
 */

1649
void gfs2_glock_prefetch_num(struct gfs2_sbd *sdp, u64 number,
1650
			     const struct gfs2_glock_operations *glops,
D
David Teigland 已提交
1651 1652 1653 1654 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 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736
			     unsigned int state, int flags)
{
	struct gfs2_glock *gl;
	int error;

	if (atomic_read(&sdp->sd_reclaim_count) <
	    gfs2_tune_get(sdp, gt_reclaim_limit)) {
		error = gfs2_glock_get(sdp, number, glops, CREATE, &gl);
		if (!error) {
			gfs2_glock_prefetch(gl, state, flags);
			gfs2_glock_put(gl);
		}
	}
}

/**
 * gfs2_lvb_hold - attach a LVB from a glock
 * @gl: The glock in question
 *
 */

int gfs2_lvb_hold(struct gfs2_glock *gl)
{
	int error;

	gfs2_glmutex_lock(gl);

	if (!atomic_read(&gl->gl_lvb_count)) {
		error = gfs2_lm_hold_lvb(gl->gl_sbd, gl->gl_lock, &gl->gl_lvb);
		if (error) {
			gfs2_glmutex_unlock(gl);
			return error;
		}
		gfs2_glock_hold(gl);
	}
	atomic_inc(&gl->gl_lvb_count);

	gfs2_glmutex_unlock(gl);

	return 0;
}

/**
 * gfs2_lvb_unhold - detach a LVB from a glock
 * @gl: The glock in question
 *
 */

void gfs2_lvb_unhold(struct gfs2_glock *gl)
{
	gfs2_glock_hold(gl);
	gfs2_glmutex_lock(gl);

	gfs2_assert(gl->gl_sbd, atomic_read(&gl->gl_lvb_count) > 0);
	if (atomic_dec_and_test(&gl->gl_lvb_count)) {
		gfs2_lm_unhold_lvb(gl->gl_sbd, gl->gl_lock, gl->gl_lvb);
		gl->gl_lvb = NULL;
		gfs2_glock_put(gl);
	}

	gfs2_glmutex_unlock(gl);
	gfs2_glock_put(gl);
}

static void blocking_cb(struct gfs2_sbd *sdp, struct lm_lockname *name,
			unsigned int state)
{
	struct gfs2_glock *gl;

	gl = gfs2_glock_find(sdp, name);
	if (!gl)
		return;

	if (gl->gl_ops->go_callback)
		gl->gl_ops->go_callback(gl, state);
	handle_callback(gl, state);

	spin_lock(&gl->gl_spin);
	run_queue(gl);
	spin_unlock(&gl->gl_spin);

	gfs2_glock_put(gl);
}

/**
 * gfs2_glock_cb - Callback used by locking module
S
Steven Whitehouse 已提交
1737
 * @sdp: Pointer to the superblock
D
David Teigland 已提交
1738 1739 1740 1741 1742 1743 1744 1745
 * @type: Type of callback
 * @data: Type dependent data pointer
 *
 * Called by the locking module when it wants to tell us something.
 * Either we need to drop a lock, one of our ASYNC requests completed, or
 * a journal from another client needs to be recovered.
 */

1746
void gfs2_glock_cb(void *cb_data, unsigned int type, void *data)
D
David Teigland 已提交
1747
{
1748
	struct gfs2_sbd *sdp = cb_data;
D
David Teigland 已提交
1749 1750 1751

	switch (type) {
	case LM_CB_NEED_E:
1752
		blocking_cb(sdp, data, LM_ST_UNLOCKED);
D
David Teigland 已提交
1753 1754 1755
		return;

	case LM_CB_NEED_D:
1756
		blocking_cb(sdp, data, LM_ST_DEFERRED);
D
David Teigland 已提交
1757 1758 1759
		return;

	case LM_CB_NEED_S:
1760
		blocking_cb(sdp, data, LM_ST_SHARED);
D
David Teigland 已提交
1761 1762 1763
		return;

	case LM_CB_ASYNC: {
1764
		struct lm_async_cb *async = data;
D
David Teigland 已提交
1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802
		struct gfs2_glock *gl;

		gl = gfs2_glock_find(sdp, &async->lc_name);
		if (gfs2_assert_warn(sdp, gl))
			return;
		if (!gfs2_assert_warn(sdp, gl->gl_req_bh))
			gl->gl_req_bh(gl, async->lc_ret);
		gfs2_glock_put(gl);
		return;
	}

	case LM_CB_NEED_RECOVERY:
		gfs2_jdesc_make_dirty(sdp, *(unsigned int *)data);
		if (sdp->sd_recoverd_process)
			wake_up_process(sdp->sd_recoverd_process);
		return;

	case LM_CB_DROPLOCKS:
		gfs2_gl_hash_clear(sdp, NO_WAIT);
		gfs2_quota_scan(sdp);
		return;

	default:
		gfs2_assert_warn(sdp, 0);
		return;
	}
}

/**
 * demote_ok - Check to see if it's ok to unlock a glock
 * @gl: the glock
 *
 * Returns: 1 if it's ok
 */

static int demote_ok(struct gfs2_glock *gl)
{
	struct gfs2_sbd *sdp = gl->gl_sbd;
1803
	const struct gfs2_glock_operations *glops = gl->gl_ops;
D
David Teigland 已提交
1804 1805 1806 1807 1808
	int demote = 1;

	if (test_bit(GLF_STICKY, &gl->gl_flags))
		demote = 0;
	else if (test_bit(GLF_PREFETCH, &gl->gl_flags))
S
Steven Whitehouse 已提交
1809
		demote = time_after_eq(jiffies, gl->gl_stamp +
D
David Teigland 已提交
1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 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 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866
				    gfs2_tune_get(sdp, gt_prefetch_secs) * HZ);
	else if (glops->go_demote_ok)
		demote = glops->go_demote_ok(gl);

	return demote;
}

/**
 * gfs2_glock_schedule_for_reclaim - Add a glock to the reclaim list
 * @gl: the glock
 *
 */

void gfs2_glock_schedule_for_reclaim(struct gfs2_glock *gl)
{
	struct gfs2_sbd *sdp = gl->gl_sbd;

	spin_lock(&sdp->sd_reclaim_lock);
	if (list_empty(&gl->gl_reclaim)) {
		gfs2_glock_hold(gl);
		list_add(&gl->gl_reclaim, &sdp->sd_reclaim_list);
		atomic_inc(&sdp->sd_reclaim_count);
	}
	spin_unlock(&sdp->sd_reclaim_lock);

	wake_up(&sdp->sd_reclaim_wq);
}

/**
 * gfs2_reclaim_glock - process the next glock on the filesystem's reclaim list
 * @sdp: the filesystem
 *
 * Called from gfs2_glockd() glock reclaim daemon, or when promoting a
 * different glock and we notice that there are a lot of glocks in the
 * reclaim list.
 *
 */

void gfs2_reclaim_glock(struct gfs2_sbd *sdp)
{
	struct gfs2_glock *gl;

	spin_lock(&sdp->sd_reclaim_lock);
	if (list_empty(&sdp->sd_reclaim_list)) {
		spin_unlock(&sdp->sd_reclaim_lock);
		return;
	}
	gl = list_entry(sdp->sd_reclaim_list.next,
			struct gfs2_glock, gl_reclaim);
	list_del_init(&gl->gl_reclaim);
	spin_unlock(&sdp->sd_reclaim_lock);

	atomic_dec(&sdp->sd_reclaim_count);
	atomic_inc(&sdp->sd_reclaimed);

	if (gfs2_glmutex_trylock(gl)) {
		if (queue_empty(gl, &gl->gl_holders) &&
S
Steven Whitehouse 已提交
1867
		    gl->gl_state != LM_ST_UNLOCKED && demote_ok(gl))
D
David Teigland 已提交
1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884
			handle_callback(gl, LM_ST_UNLOCKED);
		gfs2_glmutex_unlock(gl);
	}

	gfs2_glock_put(gl);
}

/**
 * examine_bucket - Call a function for glock in a hash bucket
 * @examiner: the function
 * @sdp: the filesystem
 * @bucket: the bucket
 *
 * Returns: 1 if the bucket has entries
 */

static int examine_bucket(glock_examiner examiner, struct gfs2_sbd *sdp,
1885
			  unsigned int hash)
D
David Teigland 已提交
1886
{
1887 1888
	struct gfs2_glock *gl, *prev = NULL;
	int has_entries = 0;
1889
	struct hlist_head *head = &gl_hash_table[hash].hb_list;
D
David Teigland 已提交
1890

1891
	read_lock(gl_lock_addr(hash));
1892 1893
	/* Can't use hlist_for_each_entry - don't want prefetch here */
	if (hlist_empty(head))
1894
		goto out;
1895 1896
	gl = list_entry(head->first, struct gfs2_glock, gl_list);
	while(1) {
1897
		if (gl->gl_sbd == sdp) {
D
David Teigland 已提交
1898
			gfs2_glock_hold(gl);
1899 1900 1901 1902 1903
			read_unlock(gl_lock_addr(hash));
			if (prev)
				gfs2_glock_put(prev);
			prev = gl;
			examiner(gl);
1904
			has_entries = 1;
1905
			read_lock(gl_lock_addr(hash));
D
David Teigland 已提交
1906
		}
1907 1908
		if (gl->gl_list.next == NULL)
			break;
1909
		gl = list_entry(gl->gl_list.next, struct gfs2_glock, gl_list);
D
David Teigland 已提交
1910
	}
1911 1912 1913 1914 1915
out:
	read_unlock(gl_lock_addr(hash));
	if (prev)
		gfs2_glock_put(prev);
	return has_entries;
D
David Teigland 已提交
1916 1917 1918 1919 1920 1921 1922 1923 1924 1925
}

/**
 * scan_glock - look at a glock and see if we can reclaim it
 * @gl: the glock to look at
 *
 */

static void scan_glock(struct gfs2_glock *gl)
{
1926
	if (gl->gl_ops == &gfs2_inode_glops)
1927
		return;
1928

D
David Teigland 已提交
1929 1930
	if (gfs2_glmutex_trylock(gl)) {
		if (queue_empty(gl, &gl->gl_holders) &&
1931
		    gl->gl_state != LM_ST_UNLOCKED && demote_ok(gl))
D
David Teigland 已提交
1932 1933 1934 1935 1936
			goto out_schedule;
		gfs2_glmutex_unlock(gl);
	}
	return;

1937
out_schedule:
D
David Teigland 已提交
1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951
	gfs2_glmutex_unlock(gl);
	gfs2_glock_schedule_for_reclaim(gl);
}

/**
 * gfs2_scand_internal - Look for glocks and inodes to toss from memory
 * @sdp: the filesystem
 *
 */

void gfs2_scand_internal(struct gfs2_sbd *sdp)
{
	unsigned int x;

1952
	for (x = 0; x < GFS2_GL_HASH_SIZE; x++)
1953
		examine_bucket(scan_glock, sdp, x);
D
David Teigland 已提交
1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970
}

/**
 * clear_glock - look at a glock and see if we can free it from glock cache
 * @gl: the glock to look at
 *
 */

static void clear_glock(struct gfs2_glock *gl)
{
	struct gfs2_sbd *sdp = gl->gl_sbd;
	int released;

	spin_lock(&sdp->sd_reclaim_lock);
	if (!list_empty(&gl->gl_reclaim)) {
		list_del_init(&gl->gl_reclaim);
		atomic_dec(&sdp->sd_reclaim_count);
1971
		spin_unlock(&sdp->sd_reclaim_lock);
D
David Teigland 已提交
1972 1973
		released = gfs2_glock_put(gl);
		gfs2_assert(sdp, !released);
1974 1975
	} else {
		spin_unlock(&sdp->sd_reclaim_lock);
D
David Teigland 已提交
1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004
	}

	if (gfs2_glmutex_trylock(gl)) {
		if (queue_empty(gl, &gl->gl_holders) &&
		    gl->gl_state != LM_ST_UNLOCKED)
			handle_callback(gl, LM_ST_UNLOCKED);
		gfs2_glmutex_unlock(gl);
	}
}

/**
 * gfs2_gl_hash_clear - Empty out the glock hash table
 * @sdp: the filesystem
 * @wait: wait until it's all gone
 *
 * Called when unmounting the filesystem, or when inter-node lock manager
 * requests DROPLOCKS because it is running out of capacity.
 */

void gfs2_gl_hash_clear(struct gfs2_sbd *sdp, int wait)
{
	unsigned long t;
	unsigned int x;
	int cont;

	t = jiffies;

	for (;;) {
		cont = 0;
2005
		for (x = 0; x < GFS2_GL_HASH_SIZE; x++) {
2006
			if (examine_bucket(clear_glock, sdp, x))
D
David Teigland 已提交
2007
				cont = 1;
2008
		}
D
David Teigland 已提交
2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021

		if (!wait || !cont)
			break;

		if (time_after_eq(jiffies,
				  t + gfs2_tune_get(sdp, gt_stall_secs) * HZ)) {
			fs_warn(sdp, "Unmount seems to be stalled. "
				     "Dumping lock state...\n");
			gfs2_dump_lockstate(sdp);
			t = jiffies;
		}

		invalidate_inodes(sdp->sd_vfs);
S
Steven Whitehouse 已提交
2022
		msleep(10);
D
David Teigland 已提交
2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042
	}
}

/*
 *  Diagnostic routines to help debug distributed deadlock
 */

/**
 * dump_holder - print information about a glock holder
 * @str: a string naming the type of holder
 * @gh: the glock holder
 *
 * Returns: 0 on success, -ENOBUFS when we run out of space
 */

static int dump_holder(char *str, struct gfs2_holder *gh)
{
	unsigned int x;
	int error = -ENOBUFS;

2043 2044
	printk(KERN_INFO "  %s\n", str);
	printk(KERN_INFO "    owner = %ld\n",
D
David Teigland 已提交
2045
		   (gh->gh_owner) ? (long)gh->gh_owner->pid : -1);
2046 2047
	printk(KERN_INFO "    gh_state = %u\n", gh->gh_state);
	printk(KERN_INFO "    gh_flags =");
D
David Teigland 已提交
2048 2049 2050 2051
	for (x = 0; x < 32; x++)
		if (gh->gh_flags & (1 << x))
			printk(" %u", x);
	printk(" \n");
2052 2053
	printk(KERN_INFO "    error = %d\n", gh->gh_error);
	printk(KERN_INFO "    gh_iflags =");
D
David Teigland 已提交
2054 2055 2056 2057
	for (x = 0; x < 32; x++)
		if (test_bit(x, &gh->gh_iflags))
			printk(" %u", x);
	printk(" \n");
S
Steven Whitehouse 已提交
2058
	print_symbol(KERN_INFO "    initialized at: %s\n", gh->gh_ip);
D
David Teigland 已提交
2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076

	error = 0;

	return error;
}

/**
 * dump_inode - print information about an inode
 * @ip: the inode
 *
 * Returns: 0 on success, -ENOBUFS when we run out of space
 */

static int dump_inode(struct gfs2_inode *ip)
{
	unsigned int x;
	int error = -ENOBUFS;

2077 2078
	printk(KERN_INFO "  Inode:\n");
	printk(KERN_INFO "    num = %llu %llu\n",
2079 2080
		    (unsigned long long)ip->i_num.no_formal_ino,
		    (unsigned long long)ip->i_num.no_addr);
2081 2082
	printk(KERN_INFO "    type = %u\n", IF2DT(ip->i_di.di_mode));
	printk(KERN_INFO "    i_flags =");
D
David Teigland 已提交
2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108
	for (x = 0; x < 32; x++)
		if (test_bit(x, &ip->i_flags))
			printk(" %u", x);
	printk(" \n");

	error = 0;

	return error;
}

/**
 * dump_glock - print information about a glock
 * @gl: the glock
 * @count: where we are in the buffer
 *
 * Returns: 0 on success, -ENOBUFS when we run out of space
 */

static int dump_glock(struct gfs2_glock *gl)
{
	struct gfs2_holder *gh;
	unsigned int x;
	int error = -ENOBUFS;

	spin_lock(&gl->gl_spin);

2109
	printk(KERN_INFO "Glock 0x%p (%u, %llu)\n", gl, gl->gl_name.ln_type,
2110
	       (unsigned long long)gl->gl_name.ln_number);
2111
	printk(KERN_INFO "  gl_flags =");
2112
	for (x = 0; x < 32; x++) {
D
David Teigland 已提交
2113 2114
		if (test_bit(x, &gl->gl_flags))
			printk(" %u", x);
2115
	}
D
David Teigland 已提交
2116
	printk(" \n");
2117
	printk(KERN_INFO "  gl_ref = %d\n", atomic_read(&gl->gl_ref));
2118
	printk(KERN_INFO "  gl_state = %u\n", gl->gl_state);
2119 2120
	printk(KERN_INFO "  gl_owner = %s\n", gl->gl_owner->comm);
	print_symbol(KERN_INFO "  gl_ip = %s\n", gl->gl_ip);
2121 2122 2123 2124 2125
	printk(KERN_INFO "  req_gh = %s\n", (gl->gl_req_gh) ? "yes" : "no");
	printk(KERN_INFO "  req_bh = %s\n", (gl->gl_req_bh) ? "yes" : "no");
	printk(KERN_INFO "  lvb_count = %d\n", atomic_read(&gl->gl_lvb_count));
	printk(KERN_INFO "  object = %s\n", (gl->gl_object) ? "yes" : "no");
	printk(KERN_INFO "  le = %s\n",
D
David Teigland 已提交
2126
		   (list_empty(&gl->gl_le.le_list)) ? "no" : "yes");
2127
	printk(KERN_INFO "  reclaim = %s\n",
D
David Teigland 已提交
2128 2129
		    (list_empty(&gl->gl_reclaim)) ? "no" : "yes");
	if (gl->gl_aspace)
2130
		printk(KERN_INFO "  aspace = 0x%p nrpages = %lu\n", gl->gl_aspace,
2131
		       gl->gl_aspace->i_mapping->nrpages);
D
David Teigland 已提交
2132
	else
2133 2134
		printk(KERN_INFO "  aspace = no\n");
	printk(KERN_INFO "  ail = %d\n", atomic_read(&gl->gl_ail_count));
D
David Teigland 已提交
2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159
	if (gl->gl_req_gh) {
		error = dump_holder("Request", gl->gl_req_gh);
		if (error)
			goto out;
	}
	list_for_each_entry(gh, &gl->gl_holders, gh_list) {
		error = dump_holder("Holder", gh);
		if (error)
			goto out;
	}
	list_for_each_entry(gh, &gl->gl_waiters1, gh_list) {
		error = dump_holder("Waiter1", gh);
		if (error)
			goto out;
	}
	list_for_each_entry(gh, &gl->gl_waiters2, gh_list) {
		error = dump_holder("Waiter2", gh);
		if (error)
			goto out;
	}
	list_for_each_entry(gh, &gl->gl_waiters3, gh_list) {
		error = dump_holder("Waiter3", gh);
		if (error)
			goto out;
	}
2160
	if (gl->gl_ops == &gfs2_inode_glops && gl->gl_object) {
D
David Teigland 已提交
2161 2162
		if (!test_bit(GLF_LOCK, &gl->gl_flags) &&
		    list_empty(&gl->gl_holders)) {
2163
			error = dump_inode(gl->gl_object);
D
David Teigland 已提交
2164 2165 2166 2167
			if (error)
				goto out;
		} else {
			error = -ENOBUFS;
2168
			printk(KERN_INFO "  Inode: busy\n");
D
David Teigland 已提交
2169 2170 2171 2172 2173
		}
	}

	error = 0;

2174
out:
D
David Teigland 已提交
2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187
	spin_unlock(&gl->gl_spin);
	return error;
}

/**
 * gfs2_dump_lockstate - print out the current lockstate
 * @sdp: the filesystem
 * @ub: the buffer to copy the information into
 *
 * If @ub is NULL, dump the lockstate to the console.
 *
 */

2188
static int gfs2_dump_lockstate(struct gfs2_sbd *sdp)
D
David Teigland 已提交
2189 2190
{
	struct gfs2_glock *gl;
2191
	struct hlist_node *h;
D
David Teigland 已提交
2192 2193 2194 2195 2196
	unsigned int x;
	int error = 0;

	for (x = 0; x < GFS2_GL_HASH_SIZE; x++) {

2197
		read_lock(gl_lock_addr(x));
D
David Teigland 已提交
2198

2199
		hlist_for_each_entry(gl, h, &gl_hash_table[x].hb_list, gl_list) {
2200 2201
			if (gl->gl_sbd != sdp)
				continue;
D
David Teigland 已提交
2202 2203 2204 2205 2206 2207

			error = dump_glock(gl);
			if (error)
				break;
		}

2208
		read_unlock(gl_lock_addr(x));
D
David Teigland 已提交
2209 2210 2211 2212 2213 2214 2215 2216 2217

		if (error)
			break;
	}


	return error;
}

2218 2219 2220 2221
int __init gfs2_glock_init(void)
{
	unsigned i;
	for(i = 0; i < GFS2_GL_HASH_SIZE; i++) {
2222
		INIT_HLIST_HEAD(&gl_hash_table[i].hb_list);
2223
	}
2224 2225 2226 2227 2228
#ifdef GL_HASH_LOCK_SZ
	for(i = 0; i < GL_HASH_LOCK_SZ; i++) {
		rwlock_init(&gl_hash_locks[i]);
	}
#endif
2229 2230 2231
	return 0;
}