glock.c 52.5 KB
Newer Older
D
David Teigland 已提交
1 2
/*
 * Copyright (C) Sistina Software, Inc.  1997-2003 All rights reserved.
3
 * Copyright (C) 2004-2008 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
 */

#include <linux/sched.h>
#include <linux/slab.h>
#include <linux/spinlock.h>
#include <linux/buffer_head.h>
#include <linux/delay.h>
#include <linux/sort.h>
#include <linux/jhash.h>
S
Steven Whitehouse 已提交
17
#include <linux/kallsyms.h>
18
#include <linux/gfs2_ondisk.h>
19
#include <linux/list.h>
20
#include <linux/wait.h>
A
akpm@linux-foundation.org 已提交
21
#include <linux/module.h>
D
David Teigland 已提交
22
#include <asm/uaccess.h>
23 24
#include <linux/seq_file.h>
#include <linux/debugfs.h>
25 26
#include <linux/kthread.h>
#include <linux/freezer.h>
27 28
#include <linux/workqueue.h>
#include <linux/jiffies.h>
29 30 31
#include <linux/rcupdate.h>
#include <linux/rculist_bl.h>
#include <linux/bit_spinlock.h>
32
#include <linux/percpu.h>
33
#include <linux/list_sort.h>
D
David Teigland 已提交
34 35

#include "gfs2.h"
36
#include "incore.h"
D
David Teigland 已提交
37 38 39 40 41 42 43
#include "glock.h"
#include "glops.h"
#include "inode.h"
#include "lops.h"
#include "meta_io.h"
#include "quota.h"
#include "super.h"
44
#include "util.h"
45
#include "bmap.h"
S
Steven Whitehouse 已提交
46 47
#define CREATE_TRACE_POINTS
#include "trace_gfs2.h"
D
David Teigland 已提交
48

49
struct gfs2_glock_iter {
50 51 52 53 54
	int hash;			/* hash bucket index           */
	unsigned nhash;			/* Index within current bucket */
	struct gfs2_sbd *sdp;		/* incore superblock           */
	struct gfs2_glock *gl;		/* current glock struct        */
	loff_t last_pos;		/* last position               */
55 56
};

D
David Teigland 已提交
57 58
typedef void (*glock_examiner) (struct gfs2_glock * gl);

59
static void do_xmote(struct gfs2_glock *gl, struct gfs2_holder *gh, unsigned int target);
60

61
static struct dentry *gfs2_root;
62
static struct workqueue_struct *glock_workqueue;
63
struct workqueue_struct *gfs2_delete_workqueue;
64 65
static LIST_HEAD(lru_list);
static atomic_t lru_count = ATOMIC_INIT(0);
J
Julia Lawall 已提交
66
static DEFINE_SPINLOCK(lru_lock);
67

68
#define GFS2_GL_HASH_SHIFT      15
69 70 71
#define GFS2_GL_HASH_SIZE       (1 << GFS2_GL_HASH_SHIFT)
#define GFS2_GL_HASH_MASK       (GFS2_GL_HASH_SIZE - 1)

72
static struct hlist_bl_head gl_hash_table[GFS2_GL_HASH_SIZE];
73
static struct dentry *gfs2_root;
74

D
David Teigland 已提交
75 76 77 78 79 80 81
/**
 * gl_hash() - Turn glock number into hash bucket number
 * @lock: The glock number
 *
 * Returns: The number of the corresponding hash bucket
 */

82 83
static unsigned int gl_hash(const struct gfs2_sbd *sdp,
			    const struct lm_lockname *name)
D
David Teigland 已提交
84 85 86
{
	unsigned int h;

87
	h = jhash(&name->ln_number, sizeof(u64), 0);
D
David Teigland 已提交
88
	h = jhash(&name->ln_type, sizeof(unsigned int), h);
89
	h = jhash(&sdp, sizeof(struct gfs2_sbd *), h);
D
David Teigland 已提交
90 91 92 93 94
	h &= GFS2_GL_HASH_MASK;

	return h;
}

95 96
static inline void spin_lock_bucket(unsigned int hash)
{
97
	hlist_bl_lock(&gl_hash_table[hash]);
98
}
D
David Teigland 已提交
99

100 101
static inline void spin_unlock_bucket(unsigned int hash)
{
102
	hlist_bl_unlock(&gl_hash_table[hash]);
103
}
D
David Teigland 已提交
104

105
static void gfs2_glock_dealloc(struct rcu_head *rcu)
D
David Teigland 已提交
106
{
107
	struct gfs2_glock *gl = container_of(rcu, struct gfs2_glock, gl_rcu);
D
David Teigland 已提交
108

109
	if (gl->gl_ops->go_flags & GLOF_ASPACE) {
110
		kmem_cache_free(gfs2_glock_aspace_cachep, gl);
111
	} else {
112
		kfree(gl->gl_lksb.sb_lvbptr);
113
		kmem_cache_free(gfs2_glock_cachep, gl);
114
	}
115 116 117
}

void gfs2_glock_free(struct gfs2_glock *gl)
D
David Teigland 已提交
118 119 120
{
	struct gfs2_sbd *sdp = gl->gl_sbd;

121
	call_rcu(&gl->gl_rcu, gfs2_glock_dealloc);
122 123
	if (atomic_dec_and_test(&sdp->sd_glock_disposal))
		wake_up(&sdp->sd_glock_wait);
D
David Teigland 已提交
124 125 126 127 128 129 130 131
}

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

132
void gfs2_glock_hold(struct gfs2_glock *gl)
D
David Teigland 已提交
133
{
134
	GLOCK_BUG_ON(gl, atomic_read(&gl->gl_ref) == 0);
135
	atomic_inc(&gl->gl_ref);
D
David Teigland 已提交
136 137
}

138 139 140 141 142 143 144 145 146 147 148 149 150
/**
 * 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(const struct gfs2_glock *gl)
{
	const struct gfs2_glock_operations *glops = gl->gl_ops;

	if (gl->gl_state == LM_ST_UNLOCKED)
		return 0;
151
	if (!list_empty(&gl->gl_holders))
152 153 154 155 156 157
		return 0;
	if (glops->go_demote_ok)
		return glops->go_demote_ok(gl);
	return 1;
}

158

159 160 161 162 163 164 165 166 167 168
void gfs2_glock_add_to_lru(struct gfs2_glock *gl)
{
	spin_lock(&lru_lock);

	if (!list_empty(&gl->gl_lru))
		list_del_init(&gl->gl_lru);
	else
		atomic_inc(&lru_count);

	list_add_tail(&gl->gl_lru, &lru_list);
169
	set_bit(GLF_LRU, &gl->gl_flags);
170 171 172
	spin_unlock(&lru_lock);
}

173
static void __gfs2_glock_remove_from_lru(struct gfs2_glock *gl)
174 175 176 177 178 179
{
	if (!list_empty(&gl->gl_lru)) {
		list_del_init(&gl->gl_lru);
		atomic_dec(&lru_count);
		clear_bit(GLF_LRU, &gl->gl_flags);
	}
180 181 182 183 184 185
}

static void gfs2_glock_remove_from_lru(struct gfs2_glock *gl)
{
	spin_lock(&lru_lock);
	__gfs2_glock_remove_from_lru(gl);
186 187 188
	spin_unlock(&lru_lock);
}

189 190 191 192 193 194 195 196
/**
 * gfs2_glock_put_nolock() - Decrement reference count on glock
 * @gl: The glock to put
 *
 * This function should only be used if the caller has its own reference
 * to the glock, in addition to the one it is dropping.
 */

197
void gfs2_glock_put_nolock(struct gfs2_glock *gl)
198 199 200 201 202
{
	if (atomic_dec_and_test(&gl->gl_ref))
		GLOCK_BUG_ON(gl, 1);
}

D
David Teigland 已提交
203 204 205 206 207 208
/**
 * gfs2_glock_put() - Decrement reference count on glock
 * @gl: The glock to put
 *
 */

209
void gfs2_glock_put(struct gfs2_glock *gl)
D
David Teigland 已提交
210
{
211 212
	struct gfs2_sbd *sdp = gl->gl_sbd;
	struct address_space *mapping = gfs2_glock2aspace(gl);
D
David Teigland 已提交
213

214 215 216
	if (atomic_dec_and_lock(&gl->gl_ref, &lru_lock)) {
		__gfs2_glock_remove_from_lru(gl);
		spin_unlock(&lru_lock);
217 218 219
		spin_lock_bucket(gl->gl_hash);
		hlist_bl_del_rcu(&gl->gl_list);
		spin_unlock_bucket(gl->gl_hash);
220
		GLOCK_BUG_ON(gl, !list_empty(&gl->gl_holders));
221 222 223
		GLOCK_BUG_ON(gl, mapping && mapping->nrpages);
		trace_gfs2_glock_put(gl);
		sdp->sd_lockstruct.ls_ops->lm_put_lock(gl);
D
David Teigland 已提交
224 225 226 227 228 229 230 231 232 233 234
	}
}

/**
 * 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
 */

235
static struct gfs2_glock *search_bucket(unsigned int hash,
236
					const struct gfs2_sbd *sdp,
237
					const struct lm_lockname *name)
D
David Teigland 已提交
238 239
{
	struct gfs2_glock *gl;
240
	struct hlist_bl_node *h;
D
David Teigland 已提交
241

242
	hlist_bl_for_each_entry_rcu(gl, h, &gl_hash_table[hash], gl_list) {
D
David Teigland 已提交
243 244
		if (!lm_name_equal(&gl->gl_name, name))
			continue;
245 246
		if (gl->gl_sbd != sdp)
			continue;
247 248
		if (atomic_inc_not_zero(&gl->gl_ref))
			return gl;
D
David Teigland 已提交
249 250 251 252 253
	}

	return NULL;
}

254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271
/**
 * may_grant - check if its ok to grant a new lock
 * @gl: The glock
 * @gh: The lock request which we wish to grant
 *
 * Returns: true if its ok to grant the lock
 */

static inline int may_grant(const struct gfs2_glock *gl, const struct gfs2_holder *gh)
{
	const struct gfs2_holder *gh_head = list_entry(gl->gl_holders.next, const struct gfs2_holder, gh_list);
	if ((gh->gh_state == LM_ST_EXCLUSIVE ||
	     gh_head->gh_state == LM_ST_EXCLUSIVE) && gh != gh_head)
		return 0;
	if (gl->gl_state == gh->gh_state)
		return 1;
	if (gh->gh_flags & GL_EXACT)
		return 0;
272 273 274 275 276 277
	if (gl->gl_state == LM_ST_EXCLUSIVE) {
		if (gh->gh_state == LM_ST_SHARED && gh_head->gh_state == LM_ST_SHARED)
			return 1;
		if (gh->gh_state == LM_ST_DEFERRED && gh_head->gh_state == LM_ST_DEFERRED)
			return 1;
	}
278 279 280 281 282 283 284 285 286 287 288 289
	if (gl->gl_state != LM_ST_UNLOCKED && (gh->gh_flags & LM_FLAG_ANY))
		return 1;
	return 0;
}

static void gfs2_holder_wake(struct gfs2_holder *gh)
{
	clear_bit(HIF_WAIT, &gh->gh_iflags);
	smp_mb__after_clear_bit();
	wake_up_bit(&gh->gh_iflags, HIF_WAIT);
}

290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313
/**
 * do_error - Something unexpected has happened during a lock request
 *
 */

static inline void do_error(struct gfs2_glock *gl, const int ret)
{
	struct gfs2_holder *gh, *tmp;

	list_for_each_entry_safe(gh, tmp, &gl->gl_holders, gh_list) {
		if (test_bit(HIF_HOLDER, &gh->gh_iflags))
			continue;
		if (ret & LM_OUT_ERROR)
			gh->gh_error = -EIO;
		else if (gh->gh_flags & (LM_FLAG_TRY | LM_FLAG_TRY_1CB))
			gh->gh_error = GLR_TRYFAILED;
		else
			continue;
		list_del_init(&gh->gh_list);
		trace_gfs2_glock_queue(gh, 0);
		gfs2_holder_wake(gh);
	}
}

314 315 316 317
/**
 * do_promote - promote as many requests as possible on the current queue
 * @gl: The glock
 * 
318 319
 * Returns: 1 if there is a blocked holder at the head of the list, or 2
 *          if a type specific operation is underway.
320 321 322
 */

static int do_promote(struct gfs2_glock *gl)
323 324
__releases(&gl->gl_spin)
__acquires(&gl->gl_spin)
325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341
{
	const struct gfs2_glock_operations *glops = gl->gl_ops;
	struct gfs2_holder *gh, *tmp;
	int ret;

restart:
	list_for_each_entry_safe(gh, tmp, &gl->gl_holders, gh_list) {
		if (test_bit(HIF_HOLDER, &gh->gh_iflags))
			continue;
		if (may_grant(gl, gh)) {
			if (gh->gh_list.prev == &gl->gl_holders &&
			    glops->go_lock) {
				spin_unlock(&gl->gl_spin);
				/* FIXME: eliminate this eventually */
				ret = glops->go_lock(gh);
				spin_lock(&gl->gl_spin);
				if (ret) {
342 343
					if (ret == 1)
						return 2;
344 345
					gh->gh_error = ret;
					list_del_init(&gh->gh_list);
S
Steven Whitehouse 已提交
346
					trace_gfs2_glock_queue(gh, 0);
347 348 349 350
					gfs2_holder_wake(gh);
					goto restart;
				}
				set_bit(HIF_HOLDER, &gh->gh_iflags);
S
Steven Whitehouse 已提交
351
				trace_gfs2_promote(gh, 1);
352 353 354 355
				gfs2_holder_wake(gh);
				goto restart;
			}
			set_bit(HIF_HOLDER, &gh->gh_iflags);
S
Steven Whitehouse 已提交
356
			trace_gfs2_promote(gh, 0);
357 358 359 360 361
			gfs2_holder_wake(gh);
			continue;
		}
		if (gh->gh_list.prev == &gl->gl_holders)
			return 1;
362
		do_error(gl, 0);
363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401
		break;
	}
	return 0;
}

/**
 * find_first_waiter - find the first gh that's waiting for the glock
 * @gl: the glock
 */

static inline struct gfs2_holder *find_first_waiter(const struct gfs2_glock *gl)
{
	struct gfs2_holder *gh;

	list_for_each_entry(gh, &gl->gl_holders, gh_list) {
		if (!test_bit(HIF_HOLDER, &gh->gh_iflags))
			return gh;
	}
	return NULL;
}

/**
 * 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) {
		if (held2)
			gfs2_glock_hold(gl);
		else
402
			gfs2_glock_put_nolock(gl);
403
	}
404 405
	if (held1 && held2 && list_empty(&gl->gl_holders))
		clear_bit(GLF_QUEUED, &gl->gl_flags);
406

407 408 409 410
	if (new_state != gl->gl_target)
		/* shorten our minimum hold time */
		gl->gl_hold_time = max(gl->gl_hold_time - GL_GLOCK_HOLD_DECR,
				       GL_GLOCK_MIN_HOLD);
411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434
	gl->gl_state = new_state;
	gl->gl_tchange = jiffies;
}

static void gfs2_demote_wake(struct gfs2_glock *gl)
{
	gl->gl_demote_state = LM_ST_EXCLUSIVE;
	clear_bit(GLF_DEMOTE, &gl->gl_flags);
	smp_mb__after_clear_bit();
	wake_up_bit(&gl->gl_flags, GLF_DEMOTE);
}

/**
 * finish_xmote - The DLM has replied to one of our lock requests
 * @gl: The glock
 * @ret: The status from the DLM
 *
 */

static void finish_xmote(struct gfs2_glock *gl, unsigned int ret)
{
	const struct gfs2_glock_operations *glops = gl->gl_ops;
	struct gfs2_holder *gh;
	unsigned state = ret & LM_OUT_ST_MASK;
435
	int rv;
436 437

	spin_lock(&gl->gl_spin);
S
Steven Whitehouse 已提交
438
	trace_gfs2_glock_state_change(gl, state);
439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497
	state_change(gl, state);
	gh = find_first_waiter(gl);

	/* Demote to UN request arrived during demote to SH or DF */
	if (test_bit(GLF_DEMOTE_IN_PROGRESS, &gl->gl_flags) &&
	    state != LM_ST_UNLOCKED && gl->gl_demote_state == LM_ST_UNLOCKED)
		gl->gl_target = LM_ST_UNLOCKED;

	/* Check for state != intended state */
	if (unlikely(state != gl->gl_target)) {
		if (gh && !test_bit(GLF_DEMOTE_IN_PROGRESS, &gl->gl_flags)) {
			/* move to back of queue and try next entry */
			if (ret & LM_OUT_CANCELED) {
				if ((gh->gh_flags & LM_FLAG_PRIORITY) == 0)
					list_move_tail(&gh->gh_list, &gl->gl_holders);
				gh = find_first_waiter(gl);
				gl->gl_target = gh->gh_state;
				goto retry;
			}
			/* Some error or failed "try lock" - report it */
			if ((ret & LM_OUT_ERROR) ||
			    (gh->gh_flags & (LM_FLAG_TRY | LM_FLAG_TRY_1CB))) {
				gl->gl_target = gl->gl_state;
				do_error(gl, ret);
				goto out;
			}
		}
		switch(state) {
		/* Unlocked due to conversion deadlock, try again */
		case LM_ST_UNLOCKED:
retry:
			do_xmote(gl, gh, gl->gl_target);
			break;
		/* Conversion fails, unlock and try again */
		case LM_ST_SHARED:
		case LM_ST_DEFERRED:
			do_xmote(gl, gh, LM_ST_UNLOCKED);
			break;
		default: /* Everything else */
			printk(KERN_ERR "GFS2: wanted %u got %u\n", gl->gl_target, state);
			GLOCK_BUG_ON(gl, 1);
		}
		spin_unlock(&gl->gl_spin);
		return;
	}

	/* Fast path - we got what we asked for */
	if (test_and_clear_bit(GLF_DEMOTE_IN_PROGRESS, &gl->gl_flags))
		gfs2_demote_wake(gl);
	if (state != LM_ST_UNLOCKED) {
		if (glops->go_xmote_bh) {
			spin_unlock(&gl->gl_spin);
			rv = glops->go_xmote_bh(gl, gh);
			spin_lock(&gl->gl_spin);
			if (rv) {
				do_error(gl, rv);
				goto out;
			}
		}
498 499 500
		rv = do_promote(gl);
		if (rv == 2)
			goto out_locked;
501 502 503
	}
out:
	clear_bit(GLF_LOCK, &gl->gl_flags);
504
out_locked:
505 506 507 508 509 510 511 512 513 514 515 516
	spin_unlock(&gl->gl_spin);
}

/**
 * do_xmote - Calls the DLM to change the state of a lock
 * @gl: The lock state
 * @gh: The holder (only for promotes)
 * @target: The target lock state
 *
 */

static void do_xmote(struct gfs2_glock *gl, struct gfs2_holder *gh, unsigned int target)
517 518
__releases(&gl->gl_spin)
__acquires(&gl->gl_spin)
519 520 521 522 523 524 525 526
{
	const struct gfs2_glock_operations *glops = gl->gl_ops;
	struct gfs2_sbd *sdp = gl->gl_sbd;
	unsigned int lck_flags = gh ? gh->gh_flags : 0;
	int ret;

	lck_flags &= (LM_FLAG_TRY | LM_FLAG_TRY_1CB | LM_FLAG_NOEXP |
		      LM_FLAG_PRIORITY);
527 528
	GLOCK_BUG_ON(gl, gl->gl_state == target);
	GLOCK_BUG_ON(gl, gl->gl_state == gl->gl_target);
529 530 531 532 533
	if ((target == LM_ST_UNLOCKED || target == LM_ST_DEFERRED) &&
	    glops->go_inval) {
		set_bit(GLF_INVALIDATE_IN_PROGRESS, &gl->gl_flags);
		do_error(gl, 0); /* Fail queued try locks */
	}
534
	gl->gl_req = target;
535 536 537 538 539
	set_bit(GLF_BLOCKING, &gl->gl_flags);
	if ((gl->gl_req == LM_ST_UNLOCKED) ||
	    (gl->gl_state == LM_ST_EXCLUSIVE) ||
	    (lck_flags & (LM_FLAG_TRY|LM_FLAG_TRY_1CB)))
		clear_bit(GLF_BLOCKING, &gl->gl_flags);
540
	spin_unlock(&gl->gl_spin);
541 542
	if (glops->go_sync)
		glops->go_sync(gl);
543 544 545 546 547
	if (test_bit(GLF_INVALIDATE_IN_PROGRESS, &gl->gl_flags))
		glops->go_inval(gl, target == LM_ST_DEFERRED ? 0 : DIO_METADATA);
	clear_bit(GLF_INVALIDATE_IN_PROGRESS, &gl->gl_flags);

	gfs2_glock_hold(gl);
548 549 550
	if (sdp->sd_lockstruct.ls_ops->lm_lock)	{
		/* lock_dlm */
		ret = sdp->sd_lockstruct.ls_ops->lm_lock(gl, target, lck_flags);
551 552 553 554
		if (ret) {
			printk(KERN_ERR "GFS2: lm_lock ret %d\n", ret);
			GLOCK_BUG_ON(gl, 1);
		}
555 556
	} else { /* lock_nolock */
		finish_xmote(gl, target);
557 558 559
		if (queue_delayed_work(glock_workqueue, &gl->gl_work, 0) == 0)
			gfs2_glock_put(gl);
	}
560

561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588
	spin_lock(&gl->gl_spin);
}

/**
 * find_first_holder - find the first "holder" gh
 * @gl: the glock
 */

static inline struct gfs2_holder *find_first_holder(const struct gfs2_glock *gl)
{
	struct gfs2_holder *gh;

	if (!list_empty(&gl->gl_holders)) {
		gh = list_entry(gl->gl_holders.next, struct gfs2_holder, gh_list);
		if (test_bit(HIF_HOLDER, &gh->gh_iflags))
			return gh;
	}
	return NULL;
}

/**
 * run_queue - do all outstanding tasks related to a glock
 * @gl: The glock in question
 * @nonblock: True if we must not block in run_queue
 *
 */

static void run_queue(struct gfs2_glock *gl, const int nonblock)
589 590
__releases(&gl->gl_spin)
__acquires(&gl->gl_spin)
591 592
{
	struct gfs2_holder *gh = NULL;
593
	int ret;
594 595 596 597 598 599 600 601 602

	if (test_and_set_bit(GLF_LOCK, &gl->gl_flags))
		return;

	GLOCK_BUG_ON(gl, test_bit(GLF_DEMOTE_IN_PROGRESS, &gl->gl_flags));

	if (test_bit(GLF_DEMOTE, &gl->gl_flags) &&
	    gl->gl_demote_state != gl->gl_state) {
		if (find_first_holder(gl))
603
			goto out_unlock;
604 605 606
		if (nonblock)
			goto out_sched;
		set_bit(GLF_DEMOTE_IN_PROGRESS, &gl->gl_flags);
607
		GLOCK_BUG_ON(gl, gl->gl_demote_state == LM_ST_EXCLUSIVE);
608 609 610 611
		gl->gl_target = gl->gl_demote_state;
	} else {
		if (test_bit(GLF_DEMOTE, &gl->gl_flags))
			gfs2_demote_wake(gl);
612 613
		ret = do_promote(gl);
		if (ret == 0)
614
			goto out_unlock;
615
		if (ret == 2)
616
			goto out;
617 618 619 620 621 622
		gh = find_first_waiter(gl);
		gl->gl_target = gh->gh_state;
		if (!(gh->gh_flags & (LM_FLAG_TRY | LM_FLAG_TRY_1CB)))
			do_error(gl, 0); /* Fail queued try locks */
	}
	do_xmote(gl, gh, gl->gl_target);
623
out:
624 625 626
	return;

out_sched:
627 628
	clear_bit(GLF_LOCK, &gl->gl_flags);
	smp_mb__after_clear_bit();
629 630
	gfs2_glock_hold(gl);
	if (queue_delayed_work(glock_workqueue, &gl->gl_work, 0) == 0)
631
		gfs2_glock_put_nolock(gl);
632 633
	return;

634
out_unlock:
635
	clear_bit(GLF_LOCK, &gl->gl_flags);
636 637
	smp_mb__after_clear_bit();
	return;
638 639
}

640 641 642 643
static void delete_work_func(struct work_struct *work)
{
	struct gfs2_glock *gl = container_of(work, struct gfs2_glock, gl_delete);
	struct gfs2_sbd *sdp = gl->gl_sbd;
644
	struct gfs2_inode *ip;
645
	struct inode *inode;
646 647 648 649
	u64 no_addr = gl->gl_name.ln_number;

	ip = gl->gl_object;
	/* Note: Unsafe to dereference ip as we don't hold right refs/locks */
650 651

	if (ip)
652
		inode = gfs2_ilookup(sdp->sd_vfs, no_addr, 1);
653 654 655 656 657
	else
		inode = gfs2_lookup_by_inum(sdp, no_addr, NULL, GFS2_BLKST_UNLINKED);
	if (inode && !IS_ERR(inode)) {
		d_prune_aliases(inode);
		iput(inode);
658 659 660 661
	}
	gfs2_glock_put(gl);
}

662 663
static void glock_work_func(struct work_struct *work)
{
664
	unsigned long delay = 0;
665
	struct gfs2_glock *gl = container_of(work, struct gfs2_glock, gl_work.work);
666
	int drop_ref = 0;
667

668
	if (test_and_clear_bit(GLF_REPLY_PENDING, &gl->gl_flags)) {
669
		finish_xmote(gl, gl->gl_reply);
670 671
		drop_ref = 1;
	}
672
	spin_lock(&gl->gl_spin);
673
	if (test_bit(GLF_PENDING_DEMOTE, &gl->gl_flags) &&
674 675
	    gl->gl_state != LM_ST_UNLOCKED &&
	    gl->gl_demote_state != LM_ST_EXCLUSIVE) {
676
		unsigned long holdtime, now = jiffies;
677

678
		holdtime = gl->gl_tchange + gl->gl_hold_time;
679 680
		if (time_before(now, holdtime))
			delay = holdtime - now;
681 682 683 684 685

		if (!delay) {
			clear_bit(GLF_PENDING_DEMOTE, &gl->gl_flags);
			set_bit(GLF_DEMOTE, &gl->gl_flags);
		}
686 687
	}
	run_queue(gl, 0);
688
	spin_unlock(&gl->gl_spin);
689
	if (!delay)
690
		gfs2_glock_put(gl);
691 692 693 694 695 696
	else {
		if (gl->gl_name.ln_type != LM_TYPE_INODE)
			delay = 0;
		if (queue_delayed_work(glock_workqueue, &gl->gl_work, delay) == 0)
			gfs2_glock_put(gl);
	}
697 698
	if (drop_ref)
		gfs2_glock_put(gl);
699 700
}

D
David Teigland 已提交
701 702 703 704 705 706 707 708 709 710 711 712 713
/**
 * 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
 */

714
int gfs2_glock_get(struct gfs2_sbd *sdp, u64 number,
715
		   const struct gfs2_glock_operations *glops, int create,
D
David Teigland 已提交
716 717
		   struct gfs2_glock **glp)
{
718
	struct super_block *s = sdp->sd_vfs;
719
	struct lm_lockname name = { .ln_number = number, .ln_type = glops->go_type };
D
David Teigland 已提交
720
	struct gfs2_glock *gl, *tmp;
721
	unsigned int hash = gl_hash(sdp, &name);
722
	struct address_space *mapping;
723
	struct kmem_cache *cachep;
D
David Teigland 已提交
724

725
	rcu_read_lock();
726
	gl = search_bucket(hash, sdp, &name);
727
	rcu_read_unlock();
D
David Teigland 已提交
728

729 730
	*glp = gl;
	if (gl)
D
David Teigland 已提交
731
		return 0;
732 733
	if (!create)
		return -ENOENT;
D
David Teigland 已提交
734

735
	if (glops->go_flags & GLOF_ASPACE)
736
		cachep = gfs2_glock_aspace_cachep;
737
	else
738 739
		cachep = gfs2_glock_cachep;
	gl = kmem_cache_alloc(cachep, GFP_KERNEL);
D
David Teigland 已提交
740 741 742
	if (!gl)
		return -ENOMEM;

743 744 745
	memset(&gl->gl_lksb, 0, sizeof(struct dlm_lksb));

	if (glops->go_flags & GLOF_LVB) {
746 747
		gl->gl_lksb.sb_lvbptr = kzalloc(GFS2_MIN_LVB_SIZE, GFP_KERNEL);
		if (!gl->gl_lksb.sb_lvbptr) {
748 749 750 751 752
			kmem_cache_free(cachep, gl);
			return -ENOMEM;
		}
	}

753
	atomic_inc(&sdp->sd_glock_disposal);
754
	gl->gl_sbd = sdp;
755
	gl->gl_flags = 0;
D
David Teigland 已提交
756
	gl->gl_name = name;
757
	atomic_set(&gl->gl_ref, 1);
D
David Teigland 已提交
758
	gl->gl_state = LM_ST_UNLOCKED;
759
	gl->gl_target = LM_ST_UNLOCKED;
760
	gl->gl_demote_state = LM_ST_EXCLUSIVE;
761
	gl->gl_hash = hash;
D
David Teigland 已提交
762
	gl->gl_ops = glops;
763 764 765 766 767 768 769
	gl->gl_dstamp = ktime_set(0, 0);
	preempt_disable();
	/* We use the global stats to estimate the initial per-glock stats */
	gl->gl_stats = this_cpu_ptr(sdp->sd_lkstats)->lkstats[glops->go_type];
	preempt_enable();
	gl->gl_stats.stats[GFS2_LKS_DCOUNT] = 0;
	gl->gl_stats.stats[GFS2_LKS_QCOUNT] = 0;
770
	gl->gl_tchange = jiffies;
771
	gl->gl_object = NULL;
772
	gl->gl_hold_time = GL_GLOCK_DFT_HOLD;
773
	INIT_DELAYED_WORK(&gl->gl_work, glock_work_func);
774
	INIT_WORK(&gl->gl_delete, delete_work_func);
D
David Teigland 已提交
775

776 777 778 779 780 781
	mapping = gfs2_glock2aspace(gl);
	if (mapping) {
                mapping->a_ops = &gfs2_meta_aops;
		mapping->host = s->s_bdev->bd_inode;
		mapping->flags = 0;
		mapping_set_gfp_mask(mapping, GFP_NOFS);
782
		mapping->private_data = NULL;
783 784
		mapping->backing_dev_info = s->s_bdi;
		mapping->writeback_index = 0;
D
David Teigland 已提交
785 786
	}

787
	spin_lock_bucket(hash);
788
	tmp = search_bucket(hash, sdp, &name);
D
David Teigland 已提交
789
	if (tmp) {
790
		spin_unlock_bucket(hash);
791
		kfree(gl->gl_lksb.sb_lvbptr);
792
		kmem_cache_free(cachep, gl);
793
		atomic_dec(&sdp->sd_glock_disposal);
D
David Teigland 已提交
794 795
		gl = tmp;
	} else {
796 797
		hlist_bl_add_head_rcu(&gl->gl_list, &gl_hash_table[hash]);
		spin_unlock_bucket(hash);
D
David Teigland 已提交
798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813
	}

	*glp = gl;

	return 0;
}

/**
 * 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
 *
 */

814
void gfs2_holder_init(struct gfs2_glock *gl, unsigned int state, unsigned flags,
D
David Teigland 已提交
815 816 817 818
		      struct gfs2_holder *gh)
{
	INIT_LIST_HEAD(&gh->gh_list);
	gh->gh_gl = gl;
S
Steven Whitehouse 已提交
819
	gh->gh_ip = (unsigned long)__builtin_return_address(0);
820
	gh->gh_owner_pid = get_pid(task_pid(current));
D
David Teigland 已提交
821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837
	gh->gh_state = state;
	gh->gh_flags = flags;
	gh->gh_error = 0;
	gh->gh_iflags = 0;
	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.
 *
 */

838
void gfs2_holder_reinit(unsigned int state, unsigned flags, struct gfs2_holder *gh)
D
David Teigland 已提交
839 840
{
	gh->gh_state = state;
841
	gh->gh_flags = flags;
842
	gh->gh_iflags = 0;
S
Steven Whitehouse 已提交
843
	gh->gh_ip = (unsigned long)__builtin_return_address(0);
B
Bob Peterson 已提交
844 845 846
	if (gh->gh_owner_pid)
		put_pid(gh->gh_owner_pid);
	gh->gh_owner_pid = get_pid(task_pid(current));
D
David Teigland 已提交
847 848 849 850 851 852 853 854 855 856
}

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

void gfs2_holder_uninit(struct gfs2_holder *gh)
{
857
	put_pid(gh->gh_owner_pid);
D
David Teigland 已提交
858 859
	gfs2_glock_put(gh->gh_gl);
	gh->gh_gl = NULL;
S
Steven Whitehouse 已提交
860
	gh->gh_ip = 0;
D
David Teigland 已提交
861 862
}

863 864 865 866 867 868 869 870 871 872
/**
 * gfs2_glock_holder_wait
 * @word: unused
 *
 * This function and gfs2_glock_demote_wait both show up in the WCHAN
 * field. Thus I've separated these otherwise identical functions in
 * order to be more informative to the user.
 */

static int gfs2_glock_holder_wait(void *word)
873 874 875 876 877
{
        schedule();
        return 0;
}

878 879 880 881 882 883
static int gfs2_glock_demote_wait(void *word)
{
	schedule();
	return 0;
}

884 885 886 887 888 889 890 891
/**
 * gfs2_glock_wait - wait on a glock acquisition
 * @gh: the glock holder
 *
 * Returns: 0 on success
 */

int gfs2_glock_wait(struct gfs2_holder *gh)
892
{
893 894
	unsigned long time1 = jiffies;

895
	might_sleep();
896
	wait_on_bit(&gh->gh_iflags, HIF_WAIT, gfs2_glock_holder_wait, TASK_UNINTERRUPTIBLE);
897 898 899 900 901
	if (time_after(jiffies, time1 + HZ)) /* have we waited > a second? */
		/* Lengthen the minimum hold time. */
		gh->gh_gl->gl_hold_time = min(gh->gh_gl->gl_hold_time +
					      GL_GLOCK_HOLD_INCR,
					      GL_GLOCK_MAX_HOLD);
902
	return gh->gh_error;
903 904
}

D
David Teigland 已提交
905
/**
906 907 908
 * handle_callback - process a demote request
 * @gl: the glock
 * @state: the state the caller wants us to change to
D
David Teigland 已提交
909
 *
910 911
 * There are only two requests that we are going to see in actual
 * practise: LM_ST_SHARED and LM_ST_UNLOCKED
D
David Teigland 已提交
912 913
 */

914
static void handle_callback(struct gfs2_glock *gl, unsigned int state,
915
			    unsigned long delay)
D
David Teigland 已提交
916
{
917
	int bit = delay ? GLF_PENDING_DEMOTE : GLF_DEMOTE;
D
David Teigland 已提交
918

919 920 921 922 923 924 925
	set_bit(bit, &gl->gl_flags);
	if (gl->gl_demote_state == LM_ST_EXCLUSIVE) {
		gl->gl_demote_state = state;
		gl->gl_demote_time = jiffies;
	} else if (gl->gl_demote_state != LM_ST_UNLOCKED &&
			gl->gl_demote_state != state) {
		gl->gl_demote_state = LM_ST_UNLOCKED;
D
David Teigland 已提交
926
	}
927 928
	if (gl->gl_ops->go_callback)
		gl->gl_ops->go_callback(gl);
S
Steven Whitehouse 已提交
929
	trace_gfs2_demote_rq(gl);
D
David Teigland 已提交
930 931
}

932
void gfs2_print_dbg(struct seq_file *seq, const char *fmt, ...)
933
{
934
	struct va_format vaf;
935 936 937
	va_list args;

	va_start(args, fmt);
938

939
	if (seq) {
940
		seq_vprintf(seq, fmt, args);
941
	} else {
942 943 944 945
		vaf.fmt = fmt;
		vaf.va = &args;

		printk(KERN_ERR " %pV", &vaf);
946
	}
947

948 949 950
	va_end(args);
}

D
David Teigland 已提交
951 952 953 954
/**
 * add_to_queue - Add a holder to the wait queue (but look for recursion)
 * @gh: the holder structure to add
 *
955 956 957 958
 * Eventually we should move the recursive locking trap to a
 * debugging option or something like that. This is the fast
 * path and needs to have the minimum number of distractions.
 * 
D
David Teigland 已提交
959 960
 */

961
static inline void add_to_queue(struct gfs2_holder *gh)
962 963
__releases(&gl->gl_spin)
__acquires(&gl->gl_spin)
D
David Teigland 已提交
964 965
{
	struct gfs2_glock *gl = gh->gh_gl;
966 967 968
	struct gfs2_sbd *sdp = gl->gl_sbd;
	struct list_head *insert_pt = NULL;
	struct gfs2_holder *gh2;
969
	int try_futile = 0;
D
David Teigland 已提交
970

971
	BUG_ON(gh->gh_owner_pid == NULL);
972 973
	if (test_and_set_bit(HIF_WAIT, &gh->gh_iflags))
		BUG();
974

975 976
	if (gh->gh_flags & (LM_FLAG_TRY | LM_FLAG_TRY_1CB)) {
		if (test_bit(GLF_LOCK, &gl->gl_flags))
977
			try_futile = !may_grant(gl, gh);
978 979 980 981 982 983 984 985
		if (test_bit(GLF_INVALIDATE_IN_PROGRESS, &gl->gl_flags))
			goto fail;
	}

	list_for_each_entry(gh2, &gl->gl_holders, gh_list) {
		if (unlikely(gh2->gh_owner_pid == gh->gh_owner_pid &&
		    (gh->gh_gl->gl_ops->go_type != LM_TYPE_FLOCK)))
			goto trap_recursive;
986 987
		if (try_futile &&
		    !(gh2->gh_flags & (LM_FLAG_TRY | LM_FLAG_TRY_1CB))) {
988 989 990 991
fail:
			gh->gh_error = GLR_TRYFAILED;
			gfs2_holder_wake(gh);
			return;
992
		}
993 994 995 996 997
		if (test_bit(HIF_HOLDER, &gh2->gh_iflags))
			continue;
		if (unlikely((gh->gh_flags & LM_FLAG_PRIORITY) && !insert_pt))
			insert_pt = &gh2->gh_list;
	}
998
	set_bit(GLF_QUEUED, &gl->gl_flags);
999
	trace_gfs2_glock_queue(gh, 1);
1000 1001
	gfs2_glstats_inc(gl, GFS2_LKS_QCOUNT);
	gfs2_sbstats_inc(gl, GFS2_LKS_QCOUNT);
1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012
	if (likely(insert_pt == NULL)) {
		list_add_tail(&gh->gh_list, &gl->gl_holders);
		if (unlikely(gh->gh_flags & LM_FLAG_PRIORITY))
			goto do_cancel;
		return;
	}
	list_add_tail(&gh->gh_list, insert_pt);
do_cancel:
	gh = list_entry(gl->gl_holders.next, struct gfs2_holder, gh_list);
	if (!(gh->gh_flags & LM_FLAG_PRIORITY)) {
		spin_unlock(&gl->gl_spin);
S
Steven Whitehouse 已提交
1013
		if (sdp->sd_lockstruct.ls_ops->lm_cancel)
1014
			sdp->sd_lockstruct.ls_ops->lm_cancel(gl);
1015
		spin_lock(&gl->gl_spin);
D
David Teigland 已提交
1016
	}
1017
	return;
D
David Teigland 已提交
1018

1019 1020 1021 1022 1023 1024 1025 1026 1027
trap_recursive:
	print_symbol(KERN_ERR "original: %s\n", gh2->gh_ip);
	printk(KERN_ERR "pid: %d\n", pid_nr(gh2->gh_owner_pid));
	printk(KERN_ERR "lock type: %d req lock state : %d\n",
	       gh2->gh_gl->gl_name.ln_type, gh2->gh_state);
	print_symbol(KERN_ERR "new: %s\n", gh->gh_ip);
	printk(KERN_ERR "pid: %d\n", pid_nr(gh->gh_owner_pid));
	printk(KERN_ERR "lock type: %d req lock state : %d\n",
	       gh->gh_gl->gl_name.ln_type, gh->gh_state);
1028
	gfs2_dump_glock(NULL, gl);
1029
	BUG();
D
David Teigland 已提交
1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046
}

/**
 * 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;

1047
	if (unlikely(test_bit(SDF_SHUTDOWN, &sdp->sd_flags)))
D
David Teigland 已提交
1048 1049
		return -EIO;

1050 1051 1052
	if (test_bit(GLF_LRU, &gl->gl_flags))
		gfs2_glock_remove_from_lru(gl);

D
David Teigland 已提交
1053 1054
	spin_lock(&gl->gl_spin);
	add_to_queue(gh);
1055 1056 1057
	if ((LM_FLAG_NOEXP & gh->gh_flags) &&
	    test_and_clear_bit(GLF_FROZEN, &gl->gl_flags))
		set_bit(GLF_REPLY_PENDING, &gl->gl_flags);
1058
	run_queue(gl, 1);
D
David Teigland 已提交
1059 1060
	spin_unlock(&gl->gl_spin);

1061 1062
	if (!(gh->gh_flags & GL_ASYNC))
		error = gfs2_glock_wait(gh);
D
David Teigland 已提交
1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075

	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)
{
1076
	return test_bit(HIF_WAIT, &gh->gh_iflags) ? 0 : 1;
D
David Teigland 已提交
1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087
}

/**
 * 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;
1088
	const struct gfs2_glock_operations *glops = gl->gl_ops;
1089
	unsigned delay = 0;
1090
	int fast_path = 0;
D
David Teigland 已提交
1091

1092
	spin_lock(&gl->gl_spin);
D
David Teigland 已提交
1093
	if (gh->gh_flags & GL_NOCACHE)
1094
		handle_callback(gl, LM_ST_UNLOCKED, 0);
D
David Teigland 已提交
1095 1096

	list_del_init(&gh->gh_list);
1097
	if (find_first_holder(gl) == NULL) {
1098
		if (glops->go_unlock) {
1099
			GLOCK_BUG_ON(gl, test_and_set_bit(GLF_LOCK, &gl->gl_flags));
1100
			spin_unlock(&gl->gl_spin);
D
David Teigland 已提交
1101
			glops->go_unlock(gh);
1102
			spin_lock(&gl->gl_spin);
1103
			clear_bit(GLF_LOCK, &gl->gl_flags);
1104
		}
1105 1106 1107 1108
		if (list_empty(&gl->gl_holders) &&
		    !test_bit(GLF_PENDING_DEMOTE, &gl->gl_flags) &&
		    !test_bit(GLF_DEMOTE, &gl->gl_flags))
			fast_path = 1;
D
David Teigland 已提交
1109
	}
1110 1111 1112
	if (!test_bit(GLF_LFLUSH, &gl->gl_flags) && demote_ok(gl))
		gfs2_glock_add_to_lru(gl);

S
Steven Whitehouse 已提交
1113
	trace_gfs2_glock_queue(gh, 0);
D
David Teigland 已提交
1114
	spin_unlock(&gl->gl_spin);
1115 1116
	if (likely(fast_path))
		return;
1117 1118 1119

	gfs2_glock_hold(gl);
	if (test_bit(GLF_PENDING_DEMOTE, &gl->gl_flags) &&
1120 1121 1122
	    !test_bit(GLF_DEMOTE, &gl->gl_flags) &&
	    gl->gl_name.ln_type == LM_TYPE_INODE)
		delay = gl->gl_hold_time;
1123 1124
	if (queue_delayed_work(glock_workqueue, &gl->gl_work, delay) == 0)
		gfs2_glock_put(gl);
D
David Teigland 已提交
1125 1126
}

A
Abhijith Das 已提交
1127 1128 1129 1130
void gfs2_glock_dq_wait(struct gfs2_holder *gh)
{
	struct gfs2_glock *gl = gh->gh_gl;
	gfs2_glock_dq(gh);
1131 1132
	might_sleep();
	wait_on_bit(&gl->gl_flags, GLF_DEMOTE, gfs2_glock_demote_wait, TASK_UNINTERRUPTIBLE);
A
Abhijith Das 已提交
1133 1134
}

D
David Teigland 已提交
1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152
/**
 * 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
L
Lucas De Marchi 已提交
1153
 * @flags: modifier flags for the acquisition
D
David Teigland 已提交
1154 1155 1156 1157 1158
 * @gh: the struct gfs2_holder
 *
 * Returns: errno
 */

1159
int gfs2_glock_nq_num(struct gfs2_sbd *sdp, u64 number,
1160 1161
		      const struct gfs2_glock_operations *glops,
		      unsigned int state, int flags, struct gfs2_holder *gh)
D
David Teigland 已提交
1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183
{
	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)
{
1184 1185 1186 1187
	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
1188 1189

	if (a->ln_number > b->ln_number)
1190 1191 1192
		return 1;
	if (a->ln_number < b->ln_number)
		return -1;
1193
	BUG_ON(gh_a->gh_gl->gl_ops->go_type == gh_b->gh_gl->gl_ops->go_type);
1194
	return 0;
D
David Teigland 已提交
1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242
}

/**
 * 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
 *
 *
 * 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)
{
1243 1244
	struct gfs2_holder *tmp[4];
	struct gfs2_holder **pph = tmp;
D
David Teigland 已提交
1245 1246
	int error = 0;

1247 1248
	switch(num_gh) {
	case 0:
D
David Teigland 已提交
1249
		return 0;
1250
	case 1:
D
David Teigland 已提交
1251 1252
		ghs->gh_flags &= ~(LM_FLAG_TRY | GL_ASYNC);
		return gfs2_glock_nq(ghs);
1253 1254
	default:
		if (num_gh <= 4)
D
David Teigland 已提交
1255
			break;
1256 1257 1258
		pph = kmalloc(num_gh * sizeof(struct gfs2_holder *), GFP_NOFS);
		if (!pph)
			return -ENOMEM;
D
David Teigland 已提交
1259 1260
	}

1261
	error = nq_m_sync(num_gh, ghs, pph);
D
David Teigland 已提交
1262

1263 1264
	if (pph != tmp)
		kfree(pph);
D
David Teigland 已提交
1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277

	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)
{
1278 1279
	while (num_gh--)
		gfs2_glock_dq(&ghs[num_gh]);
D
David Teigland 已提交
1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290
}

/**
 * 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)
{
1291 1292
	while (num_gh--)
		gfs2_glock_dq_uninit(&ghs[num_gh]);
D
David Teigland 已提交
1293 1294
}

1295
void gfs2_glock_cb(struct gfs2_glock *gl, unsigned int state)
1296
{
1297 1298 1299
	unsigned long delay = 0;
	unsigned long holdtime;
	unsigned long now = jiffies;
D
David Teigland 已提交
1300

1301
	gfs2_glock_hold(gl);
1302 1303 1304
	holdtime = gl->gl_tchange + gl->gl_hold_time;
	if (test_bit(GLF_QUEUED, &gl->gl_flags) &&
	    gl->gl_name.ln_type == LM_TYPE_INODE) {
1305 1306 1307
		if (time_before(now, holdtime))
			delay = holdtime - now;
		if (test_bit(GLF_REPLY_PENDING, &gl->gl_flags))
1308
			delay = gl->gl_hold_time;
1309
	}
D
David Teigland 已提交
1310

1311
	spin_lock(&gl->gl_spin);
1312
	handle_callback(gl, state, delay);
1313
	spin_unlock(&gl->gl_spin);
1314 1315
	if (queue_delayed_work(glock_workqueue, &gl->gl_work, delay) == 0)
		gfs2_glock_put(gl);
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
/**
 * gfs2_should_freeze - Figure out if glock should be frozen
 * @gl: The glock in question
 *
 * Glocks are not frozen if (a) the result of the dlm operation is
 * an error, (b) the locking operation was an unlock operation or
 * (c) if there is a "noexp" flagged request anywhere in the queue
 *
 * Returns: 1 if freezing should occur, 0 otherwise
 */

static int gfs2_should_freeze(const struct gfs2_glock *gl)
{
	const struct gfs2_holder *gh;

	if (gl->gl_reply & ~LM_OUT_ST_MASK)
		return 0;
	if (gl->gl_target == LM_ST_UNLOCKED)
		return 0;

	list_for_each_entry(gh, &gl->gl_holders, gh_list) {
		if (test_bit(HIF_HOLDER, &gh->gh_iflags))
			continue;
		if (LM_FLAG_NOEXP & gh->gh_flags)
			return 0;
	}

	return 1;
}

D
David Teigland 已提交
1348
/**
1349 1350 1351
 * gfs2_glock_complete - Callback used by locking
 * @gl: Pointer to the glock
 * @ret: The return value from the dlm
D
David Teigland 已提交
1352
 *
1353 1354
 * The gl_reply field is under the gl_spin lock so that it is ok
 * to use a bitfield shared with other glock state fields.
D
David Teigland 已提交
1355 1356
 */

1357
void gfs2_glock_complete(struct gfs2_glock *gl, int ret)
D
David Teigland 已提交
1358
{
1359
	struct lm_lockstruct *ls = &gl->gl_sbd->sd_lockstruct;
1360

1361
	spin_lock(&gl->gl_spin);
1362
	gl->gl_reply = ret;
1363

1364
	if (unlikely(test_bit(DFL_BLOCK_LOCKS, &ls->ls_recover_flags))) {
1365
		if (gfs2_should_freeze(gl)) {
1366
			set_bit(GLF_FROZEN, &gl->gl_flags);
1367
			spin_unlock(&gl->gl_spin);
D
David Teigland 已提交
1368
			return;
1369
		}
D
David Teigland 已提交
1370
	}
1371 1372

	spin_unlock(&gl->gl_spin);
1373
	set_bit(GLF_REPLY_PENDING, &gl->gl_flags);
1374
	smp_wmb();
1375 1376 1377
	gfs2_glock_hold(gl);
	if (queue_delayed_work(glock_workqueue, &gl->gl_work, 0) == 0)
		gfs2_glock_put(gl);
D
David Teigland 已提交
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 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434
static int glock_cmp(void *priv, struct list_head *a, struct list_head *b)
{
	struct gfs2_glock *gla, *glb;

	gla = list_entry(a, struct gfs2_glock, gl_lru);
	glb = list_entry(b, struct gfs2_glock, gl_lru);

	if (gla->gl_name.ln_number > glb->gl_name.ln_number)
		return 1;
	if (gla->gl_name.ln_number < glb->gl_name.ln_number)
		return -1;

	return 0;
}

/**
 * gfs2_dispose_glock_lru - Demote a list of glocks
 * @list: The list to dispose of
 *
 * Disposing of glocks may involve disk accesses, so that here we sort
 * the glocks by number (i.e. disk location of the inodes) so that if
 * there are any such accesses, they'll be sent in order (mostly).
 *
 * Must be called under the lru_lock, but may drop and retake this
 * lock. While the lru_lock is dropped, entries may vanish from the
 * list, but no new entries will appear on the list (since it is
 * private)
 */

static void gfs2_dispose_glock_lru(struct list_head *list)
__releases(&lru_lock)
__acquires(&lru_lock)
{
	struct gfs2_glock *gl;

	list_sort(NULL, list, glock_cmp);

	while(!list_empty(list)) {
		gl = list_entry(list->next, struct gfs2_glock, gl_lru);
		list_del_init(&gl->gl_lru);
		clear_bit(GLF_LRU, &gl->gl_flags);
		gfs2_glock_hold(gl);
		spin_unlock(&lru_lock);
		spin_lock(&gl->gl_spin);
		if (demote_ok(gl))
			handle_callback(gl, LM_ST_UNLOCKED, 0);
		WARN_ON(!test_and_clear_bit(GLF_LOCK, &gl->gl_flags));
		smp_mb__after_clear_bit();
		if (queue_delayed_work(glock_workqueue, &gl->gl_work, 0) == 0)
			gfs2_glock_put_nolock(gl);
		spin_unlock(&gl->gl_spin);
		spin_lock(&lru_lock);
	}
}

1435 1436 1437 1438
/**
 * gfs2_scan_glock_lru - Scan the LRU looking for locks to demote
 * @nr: The number of entries to scan
 *
1439 1440 1441
 * This function selects the entries on the LRU which are able to
 * be demoted, and then kicks off the process by calling
 * gfs2_dispose_glock_lru() above.
1442
 */
D
David Teigland 已提交
1443

1444
static void gfs2_scan_glock_lru(int nr)
D
David Teigland 已提交
1445 1446
{
	struct gfs2_glock *gl;
1447
	LIST_HEAD(skipped);
1448
	LIST_HEAD(dispose);
D
David Teigland 已提交
1449

1450 1451 1452 1453 1454 1455
	spin_lock(&lru_lock);
	while(nr && !list_empty(&lru_list)) {
		gl = list_entry(lru_list.next, struct gfs2_glock, gl_lru);

		/* Test for being demotable */
		if (!test_and_set_bit(GLF_LOCK, &gl->gl_flags)) {
1456 1457 1458
			list_move(&gl->gl_lru, &dispose);
			atomic_dec(&lru_count);
			nr--;
S
Steven Whitehouse 已提交
1459
			continue;
1460
		}
1461 1462

		list_move(&gl->gl_lru, &skipped);
D
David Teigland 已提交
1463
	}
1464
	list_splice(&skipped, &lru_list);
1465 1466
	if (!list_empty(&dispose))
		gfs2_dispose_glock_lru(&dispose);
1467
	spin_unlock(&lru_lock);
1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478
}

static int gfs2_shrink_glock_memory(struct shrinker *shrink,
				    struct shrink_control *sc)
{
	if (sc->nr_to_scan) {
		if (!(sc->gfp_mask & __GFP_FS))
			return -1;
		gfs2_scan_glock_lru(sc->nr_to_scan);
	}

1479
	return (atomic_read(&lru_count) / 100) * sysctl_vfs_cache_pressure;
D
David Teigland 已提交
1480 1481
}

1482 1483 1484 1485 1486
static struct shrinker glock_shrinker = {
	.shrink = gfs2_shrink_glock_memory,
	.seeks = DEFAULT_SEEKS,
};

D
David Teigland 已提交
1487 1488 1489 1490 1491 1492 1493 1494
/**
 * examine_bucket - Call a function for glock in a hash bucket
 * @examiner: the function
 * @sdp: the filesystem
 * @bucket: the bucket
 *
 */

1495
static void examine_bucket(glock_examiner examiner, const struct gfs2_sbd *sdp,
1496
			  unsigned int hash)
D
David Teigland 已提交
1497
{
1498 1499 1500
	struct gfs2_glock *gl;
	struct hlist_bl_head *head = &gl_hash_table[hash];
	struct hlist_bl_node *pos;
D
David Teigland 已提交
1501

1502 1503 1504
	rcu_read_lock();
	hlist_bl_for_each_entry_rcu(gl, pos, head, gl_list) {
		if ((gl->gl_sbd == sdp) && atomic_read(&gl->gl_ref))
1505
			examiner(gl);
D
David Teigland 已提交
1506
	}
1507
	rcu_read_unlock();
1508
	cond_resched();
1509 1510 1511 1512 1513 1514 1515 1516
}

static void glock_hash_walk(glock_examiner examiner, const struct gfs2_sbd *sdp)
{
	unsigned x;

	for (x = 0; x < GFS2_GL_HASH_SIZE; x++)
		examine_bucket(examiner, sdp, x);
D
David Teigland 已提交
1517 1518
}

1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537

/**
 * thaw_glock - thaw out a glock which has an unprocessed reply waiting
 * @gl: The glock to thaw
 *
 * N.B. When we freeze a glock, we leave a ref to the glock outstanding,
 * so this has to result in the ref count being dropped by one.
 */

static void thaw_glock(struct gfs2_glock *gl)
{
	if (!test_and_clear_bit(GLF_FROZEN, &gl->gl_flags))
		return;
	set_bit(GLF_REPLY_PENDING, &gl->gl_flags);
	gfs2_glock_hold(gl);
	if (queue_delayed_work(glock_workqueue, &gl->gl_work, 0) == 0)
		gfs2_glock_put(gl);
}

D
David Teigland 已提交
1538 1539 1540 1541 1542 1543 1544 1545
/**
 * 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)
{
1546
	gfs2_glock_remove_from_lru(gl);
D
David Teigland 已提交
1547

1548
	spin_lock(&gl->gl_spin);
1549
	if (gl->gl_state != LM_ST_UNLOCKED)
1550
		handle_callback(gl, LM_ST_UNLOCKED, 0);
1551 1552 1553 1554
	spin_unlock(&gl->gl_spin);
	gfs2_glock_hold(gl);
	if (queue_delayed_work(glock_workqueue, &gl->gl_work, 0) == 0)
		gfs2_glock_put(gl);
D
David Teigland 已提交
1555 1556
}

1557 1558 1559 1560 1561 1562 1563 1564
/**
 * gfs2_glock_thaw - Thaw any frozen glocks
 * @sdp: The super block
 *
 */

void gfs2_glock_thaw(struct gfs2_sbd *sdp)
{
1565 1566
	glock_hash_walk(thaw_glock, sdp);
}
1567

1568 1569 1570 1571
static int dump_glock(struct seq_file *seq, struct gfs2_glock *gl)
{
	int ret;
	spin_lock(&gl->gl_spin);
1572
	ret = gfs2_dump_glock(seq, gl);
1573 1574 1575 1576 1577 1578 1579
	spin_unlock(&gl->gl_spin);
	return ret;
}

static void dump_glock_func(struct gfs2_glock *gl)
{
	dump_glock(NULL, gl);
1580 1581
}

D
David Teigland 已提交
1582 1583 1584 1585 1586
/**
 * gfs2_gl_hash_clear - Empty out the glock hash table
 * @sdp: the filesystem
 * @wait: wait until it's all gone
 *
1587
 * Called when unmounting the filesystem.
D
David Teigland 已提交
1588 1589
 */

1590
void gfs2_gl_hash_clear(struct gfs2_sbd *sdp)
D
David Teigland 已提交
1591
{
1592
	set_bit(SDF_SKIP_DLM_UNLOCK, &sdp->sd_flags);
1593
	glock_hash_walk(clear_glock, sdp);
1594 1595
	flush_workqueue(glock_workqueue);
	wait_event(sdp->sd_glock_wait, atomic_read(&sdp->sd_glock_disposal) == 0);
1596
	glock_hash_walk(dump_glock_func, sdp);
D
David Teigland 已提交
1597 1598
}

1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612
void gfs2_glock_finish_truncate(struct gfs2_inode *ip)
{
	struct gfs2_glock *gl = ip->i_gl;
	int ret;

	ret = gfs2_truncatei_resume(ip);
	gfs2_assert_withdraw(gl->gl_sbd, ret == 0);

	spin_lock(&gl->gl_spin);
	clear_bit(GLF_LOCK, &gl->gl_flags);
	run_queue(gl, 1);
	spin_unlock(&gl->gl_spin);
}

1613
static const char *state2str(unsigned state)
1614
{
1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637
	switch(state) {
	case LM_ST_UNLOCKED:
		return "UN";
	case LM_ST_SHARED:
		return "SH";
	case LM_ST_DEFERRED:
		return "DF";
	case LM_ST_EXCLUSIVE:
		return "EX";
	}
	return "??";
}

static const char *hflags2str(char *buf, unsigned flags, unsigned long iflags)
{
	char *p = buf;
	if (flags & LM_FLAG_TRY)
		*p++ = 't';
	if (flags & LM_FLAG_TRY_1CB)
		*p++ = 'T';
	if (flags & LM_FLAG_NOEXP)
		*p++ = 'e';
	if (flags & LM_FLAG_ANY)
1638
		*p++ = 'A';
1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654
	if (flags & LM_FLAG_PRIORITY)
		*p++ = 'p';
	if (flags & GL_ASYNC)
		*p++ = 'a';
	if (flags & GL_EXACT)
		*p++ = 'E';
	if (flags & GL_NOCACHE)
		*p++ = 'c';
	if (test_bit(HIF_HOLDER, &iflags))
		*p++ = 'H';
	if (test_bit(HIF_WAIT, &iflags))
		*p++ = 'W';
	if (test_bit(HIF_FIRST, &iflags))
		*p++ = 'F';
	*p = 0;
	return buf;
1655 1656
}

D
David Teigland 已提交
1657 1658
/**
 * dump_holder - print information about a glock holder
1659
 * @seq: the seq_file struct
D
David Teigland 已提交
1660 1661 1662 1663 1664
 * @gh: the glock holder
 *
 * Returns: 0 on success, -ENOBUFS when we run out of space
 */

1665
static int dump_holder(struct seq_file *seq, const struct gfs2_holder *gh)
D
David Teigland 已提交
1666
{
1667 1668
	struct task_struct *gh_owner = NULL;
	char flags_buf[32];
D
David Teigland 已提交
1669

1670
	if (gh->gh_owner_pid)
1671
		gh_owner = pid_task(gh->gh_owner_pid, PIDTYPE_PID);
1672 1673 1674 1675 1676 1677 1678
	gfs2_print_dbg(seq, " H: s:%s f:%s e:%d p:%ld [%s] %pS\n",
		       state2str(gh->gh_state),
		       hflags2str(flags_buf, gh->gh_flags, gh->gh_iflags),
		       gh->gh_error,
		       gh->gh_owner_pid ? (long)pid_nr(gh->gh_owner_pid) : -1,
		       gh_owner ? gh_owner->comm : "(ended)",
		       (void *)gh->gh_ip);
1679
	return 0;
D
David Teigland 已提交
1680 1681
}

1682
static const char *gflags2str(char *buf, const struct gfs2_glock *gl)
1683
{
1684
	const unsigned long *gflags = &gl->gl_flags;
1685
	char *p = buf;
1686

1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702
	if (test_bit(GLF_LOCK, gflags))
		*p++ = 'l';
	if (test_bit(GLF_DEMOTE, gflags))
		*p++ = 'D';
	if (test_bit(GLF_PENDING_DEMOTE, gflags))
		*p++ = 'd';
	if (test_bit(GLF_DEMOTE_IN_PROGRESS, gflags))
		*p++ = 'p';
	if (test_bit(GLF_DIRTY, gflags))
		*p++ = 'y';
	if (test_bit(GLF_LFLUSH, gflags))
		*p++ = 'f';
	if (test_bit(GLF_INVALIDATE_IN_PROGRESS, gflags))
		*p++ = 'i';
	if (test_bit(GLF_REPLY_PENDING, gflags))
		*p++ = 'r';
1703
	if (test_bit(GLF_INITIAL, gflags))
1704
		*p++ = 'I';
1705 1706
	if (test_bit(GLF_FROZEN, gflags))
		*p++ = 'F';
1707 1708
	if (test_bit(GLF_QUEUED, gflags))
		*p++ = 'q';
1709 1710 1711 1712
	if (test_bit(GLF_LRU, gflags))
		*p++ = 'L';
	if (gl->gl_object)
		*p++ = 'o';
1713 1714
	if (test_bit(GLF_BLOCKING, gflags))
		*p++ = 'b';
1715 1716
	*p = 0;
	return buf;
D
David Teigland 已提交
1717 1718 1719
}

/**
1720
 * gfs2_dump_glock - print information about a glock
1721
 * @seq: The seq_file struct
D
David Teigland 已提交
1722
 * @gl: the glock
1723 1724 1725 1726 1727 1728 1729 1730 1731 1732
 *
 * The file format is as follows:
 * One line per object, capital letters are used to indicate objects
 * G = glock, I = Inode, R = rgrp, H = holder. Glocks are not indented,
 * other objects are indented by a single space and follow the glock to
 * which they are related. Fields are indicated by lower case letters
 * followed by a colon and the field value, except for strings which are in
 * [] so that its possible to see if they are composed of spaces for
 * example. The field's are n = number (id of the object), f = flags,
 * t = type, s = state, r = refcount, e = error, p = pid.
D
David Teigland 已提交
1733 1734 1735 1736
 *
 * Returns: 0 on success, -ENOBUFS when we run out of space
 */

1737
int gfs2_dump_glock(struct seq_file *seq, const struct gfs2_glock *gl)
D
David Teigland 已提交
1738
{
1739 1740 1741 1742 1743
	const struct gfs2_glock_operations *glops = gl->gl_ops;
	unsigned long long dtime;
	const struct gfs2_holder *gh;
	char gflags_buf[32];
	int error = 0;
D
David Teigland 已提交
1744

1745 1746 1747 1748
	dtime = jiffies - gl->gl_demote_time;
	dtime *= 1000000/HZ; /* demote time in uSec */
	if (!test_bit(GLF_DEMOTE, &gl->gl_flags))
		dtime = 0;
1749
	gfs2_print_dbg(seq, "G:  s:%s n:%u/%llx f:%s t:%s d:%s/%llu a:%d v:%d r:%d m:%ld\n",
1750 1751 1752
		  state2str(gl->gl_state),
		  gl->gl_name.ln_type,
		  (unsigned long long)gl->gl_name.ln_number,
1753
		  gflags2str(gflags_buf, gl),
1754 1755 1756
		  state2str(gl->gl_target),
		  state2str(gl->gl_demote_state), dtime,
		  atomic_read(&gl->gl_ail_count),
1757
		  atomic_read(&gl->gl_revokes),
1758
		  atomic_read(&gl->gl_ref), gl->gl_hold_time);
D
David Teigland 已提交
1759 1760

	list_for_each_entry(gh, &gl->gl_holders, gh_list) {
1761
		error = dump_holder(seq, gh);
D
David Teigland 已提交
1762 1763 1764
		if (error)
			goto out;
	}
1765 1766
	if (gl->gl_state != LM_ST_UNLOCKED && glops->go_dump)
		error = glops->go_dump(seq, gl);
1767
out:
D
David Teigland 已提交
1768 1769 1770
	return error;
}

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 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826
static int gfs2_glstats_seq_show(struct seq_file *seq, void *iter_ptr)
{
	struct gfs2_glock *gl = iter_ptr;

	seq_printf(seq, "G: n:%u/%llx rtt:%lld/%lld rttb:%lld/%lld irt:%lld/%lld dcnt: %lld qcnt: %lld\n",
		   gl->gl_name.ln_type,
		   (unsigned long long)gl->gl_name.ln_number,
		   (long long)gl->gl_stats.stats[GFS2_LKS_SRTT],
		   (long long)gl->gl_stats.stats[GFS2_LKS_SRTTVAR],
		   (long long)gl->gl_stats.stats[GFS2_LKS_SRTTB],
		   (long long)gl->gl_stats.stats[GFS2_LKS_SRTTVARB],
		   (long long)gl->gl_stats.stats[GFS2_LKS_SIRT],
		   (long long)gl->gl_stats.stats[GFS2_LKS_SIRTVAR],
		   (long long)gl->gl_stats.stats[GFS2_LKS_DCOUNT],
		   (long long)gl->gl_stats.stats[GFS2_LKS_QCOUNT]);
	return 0;
}

static const char *gfs2_gltype[] = {
	"type",
	"reserved",
	"nondisk",
	"inode",
	"rgrp",
	"meta",
	"iopen",
	"flock",
	"plock",
	"quota",
	"journal",
};

static const char *gfs2_stype[] = {
	[GFS2_LKS_SRTT]		= "srtt",
	[GFS2_LKS_SRTTVAR]	= "srttvar",
	[GFS2_LKS_SRTTB]	= "srttb",
	[GFS2_LKS_SRTTVARB]	= "srttvarb",
	[GFS2_LKS_SIRT]		= "sirt",
	[GFS2_LKS_SIRTVAR]	= "sirtvar",
	[GFS2_LKS_DCOUNT]	= "dlm",
	[GFS2_LKS_QCOUNT]	= "queue",
};

#define GFS2_NR_SBSTATS (ARRAY_SIZE(gfs2_gltype) * ARRAY_SIZE(gfs2_stype))

static int gfs2_sbstats_seq_show(struct seq_file *seq, void *iter_ptr)
{
	struct gfs2_glock_iter *gi = seq->private;
	struct gfs2_sbd *sdp = gi->sdp;
	unsigned index = gi->hash >> 3;
	unsigned subindex = gi->hash & 0x07;
	s64 value;
	int i;

	if (index == 0 && subindex != 0)
		return 0;
1827

1828 1829
	seq_printf(seq, "%-10s %8s:", gfs2_gltype[index],
		   (index == 0) ? "cpu": gfs2_stype[subindex]);
D
David Teigland 已提交
1830

1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842
	for_each_possible_cpu(i) {
                const struct gfs2_pcpu_lkstats *lkstats = per_cpu_ptr(sdp->sd_lkstats, i);
		if (index == 0) {
			value = i;
		} else {
			value = lkstats->lkstats[index - 1].stats[subindex];
		}
		seq_printf(seq, " %15lld", (long long)value);
	}
	seq_putc(seq, '\n');
	return 0;
}
1843

1844 1845 1846 1847
int __init gfs2_glock_init(void)
{
	unsigned i;
	for(i = 0; i < GFS2_GL_HASH_SIZE; i++) {
1848
		INIT_HLIST_BL_HEAD(&gl_hash_table[i]);
1849
	}
1850

1851
	glock_workqueue = alloc_workqueue("glock_workqueue", WQ_MEM_RECLAIM |
1852
					  WQ_HIGHPRI | WQ_FREEZABLE, 0);
1853
	if (IS_ERR(glock_workqueue))
1854
		return PTR_ERR(glock_workqueue);
1855
	gfs2_delete_workqueue = alloc_workqueue("delete_workqueue",
1856
						WQ_MEM_RECLAIM | WQ_FREEZABLE,
1857
						0);
1858 1859 1860 1861
	if (IS_ERR(gfs2_delete_workqueue)) {
		destroy_workqueue(glock_workqueue);
		return PTR_ERR(gfs2_delete_workqueue);
	}
1862 1863

	register_shrinker(&glock_shrinker);
1864

1865 1866 1867
	return 0;
}

1868 1869
void gfs2_glock_exit(void)
{
1870
	unregister_shrinker(&glock_shrinker);
1871
	destroy_workqueue(glock_workqueue);
1872
	destroy_workqueue(gfs2_delete_workqueue);
1873 1874
}

1875 1876 1877 1878 1879 1880 1881 1882
static inline struct gfs2_glock *glock_hash_chain(unsigned hash)
{
	return hlist_bl_entry(hlist_bl_first_rcu(&gl_hash_table[hash]),
			      struct gfs2_glock, gl_list);
}

static inline struct gfs2_glock *glock_hash_next(struct gfs2_glock *gl)
{
1883
	return hlist_bl_entry(rcu_dereference(gl->gl_list.next),
1884 1885 1886
			      struct gfs2_glock, gl_list);
}

1887
static int gfs2_glock_iter_next(struct gfs2_glock_iter *gi)
1888
{
1889 1890
	struct gfs2_glock *gl;

1891 1892 1893 1894
	do {
		gl = gi->gl;
		if (gl) {
			gi->gl = glock_hash_next(gl);
1895
			gi->nhash++;
1896
		} else {
1897 1898 1899 1900
			if (gi->hash >= GFS2_GL_HASH_SIZE) {
				rcu_read_unlock();
				return 1;
			}
1901
			gi->gl = glock_hash_chain(gi->hash);
1902
			gi->nhash = 0;
1903 1904 1905 1906 1907 1908 1909 1910
		}
		while (gi->gl == NULL) {
			gi->hash++;
			if (gi->hash >= GFS2_GL_HASH_SIZE) {
				rcu_read_unlock();
				return 1;
			}
			gi->gl = glock_hash_chain(gi->hash);
1911
			gi->nhash = 0;
1912 1913 1914
		}
	/* Skip entries for other sb and dead entries */
	} while (gi->sdp != gi->gl->gl_sbd || atomic_read(&gi->gl->gl_ref) == 0);
1915

1916 1917 1918
	return 0;
}

1919
static void *gfs2_glock_seq_start(struct seq_file *seq, loff_t *pos)
1920
{
1921
	struct gfs2_glock_iter *gi = seq->private;
1922 1923
	loff_t n = *pos;

1924 1925 1926 1927 1928 1929
	if (gi->last_pos <= *pos)
		n = gi->nhash + (*pos - gi->last_pos);
	else
		gi->hash = 0;

	gi->nhash = 0;
1930
	rcu_read_lock();
1931

1932
	do {
1933
		if (gfs2_glock_iter_next(gi))
1934
			return NULL;
1935
	} while (n--);
1936

1937
	gi->last_pos = *pos;
1938
	return gi->gl;
1939 1940
}

1941
static void *gfs2_glock_seq_next(struct seq_file *seq, void *iter_ptr,
1942 1943
				 loff_t *pos)
{
1944
	struct gfs2_glock_iter *gi = seq->private;
1945 1946

	(*pos)++;
1947
	gi->last_pos = *pos;
1948
	if (gfs2_glock_iter_next(gi))
1949 1950
		return NULL;

1951
	return gi->gl;
1952 1953
}

1954
static void gfs2_glock_seq_stop(struct seq_file *seq, void *iter_ptr)
1955
{
1956
	struct gfs2_glock_iter *gi = seq->private;
1957 1958 1959 1960

	if (gi->gl)
		rcu_read_unlock();
	gi->gl = NULL;
1961 1962
}

1963
static int gfs2_glock_seq_show(struct seq_file *seq, void *iter_ptr)
1964
{
1965
	return dump_glock(seq, iter_ptr);
1966 1967
}

1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996
static void *gfs2_sbstats_seq_start(struct seq_file *seq, loff_t *pos)
{
	struct gfs2_glock_iter *gi = seq->private;

	gi->hash = *pos;
	if (*pos >= GFS2_NR_SBSTATS)
		return NULL;
	preempt_disable();
	return SEQ_START_TOKEN;
}

static void *gfs2_sbstats_seq_next(struct seq_file *seq, void *iter_ptr,
				   loff_t *pos)
{
	struct gfs2_glock_iter *gi = seq->private;
	(*pos)++;
	gi->hash++;
	if (gi->hash >= GFS2_NR_SBSTATS) {
		preempt_enable();
		return NULL;
	}
	return SEQ_START_TOKEN;
}

static void gfs2_sbstats_seq_stop(struct seq_file *seq, void *iter_ptr)
{
	preempt_enable();
}

1997
static const struct seq_operations gfs2_glock_seq_ops = {
1998 1999 2000 2001 2002 2003
	.start = gfs2_glock_seq_start,
	.next  = gfs2_glock_seq_next,
	.stop  = gfs2_glock_seq_stop,
	.show  = gfs2_glock_seq_show,
};

2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017
static const struct seq_operations gfs2_glstats_seq_ops = {
	.start = gfs2_glock_seq_start,
	.next  = gfs2_glock_seq_next,
	.stop  = gfs2_glock_seq_stop,
	.show  = gfs2_glstats_seq_show,
};

static const struct seq_operations gfs2_sbstats_seq_ops = {
	.start = gfs2_sbstats_seq_start,
	.next  = gfs2_sbstats_seq_next,
	.stop  = gfs2_sbstats_seq_stop,
	.show  = gfs2_sbstats_seq_show,
};

2018 2019
#define GFS2_SEQ_GOODSIZE min(PAGE_SIZE << PAGE_ALLOC_COSTLY_ORDER, 65536UL)

2020
static int gfs2_glocks_open(struct inode *inode, struct file *file)
2021
{
2022 2023 2024 2025 2026 2027
	int ret = seq_open_private(file, &gfs2_glock_seq_ops,
				   sizeof(struct gfs2_glock_iter));
	if (ret == 0) {
		struct seq_file *seq = file->private_data;
		struct gfs2_glock_iter *gi = seq->private;
		gi->sdp = inode->i_private;
2028
		seq->buf = kmalloc(GFS2_SEQ_GOODSIZE, GFP_KERNEL | __GFP_NOWARN);
2029
		if (seq->buf)
2030
			seq->size = GFS2_SEQ_GOODSIZE;
2031 2032
	}
	return ret;
2033 2034
}

2035 2036 2037 2038 2039 2040 2041 2042
static int gfs2_glstats_open(struct inode *inode, struct file *file)
{
	int ret = seq_open_private(file, &gfs2_glstats_seq_ops,
				   sizeof(struct gfs2_glock_iter));
	if (ret == 0) {
		struct seq_file *seq = file->private_data;
		struct gfs2_glock_iter *gi = seq->private;
		gi->sdp = inode->i_private;
2043
		seq->buf = kmalloc(GFS2_SEQ_GOODSIZE, GFP_KERNEL | __GFP_NOWARN);
2044
		if (seq->buf)
2045
			seq->size = GFS2_SEQ_GOODSIZE;
2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070
	}
	return ret;
}

static int gfs2_sbstats_open(struct inode *inode, struct file *file)
{
	int ret = seq_open_private(file, &gfs2_sbstats_seq_ops,
				   sizeof(struct gfs2_glock_iter));
	if (ret == 0) {
		struct seq_file *seq = file->private_data;
		struct gfs2_glock_iter *gi = seq->private;
		gi->sdp = inode->i_private;
	}
	return ret;
}

static const struct file_operations gfs2_glocks_fops = {
	.owner   = THIS_MODULE,
	.open    = gfs2_glocks_open,
	.read    = seq_read,
	.llseek  = seq_lseek,
	.release = seq_release_private,
};

static const struct file_operations gfs2_glstats_fops = {
2071
	.owner   = THIS_MODULE,
2072 2073 2074 2075 2076 2077 2078 2079 2080
	.open    = gfs2_glstats_open,
	.read    = seq_read,
	.llseek  = seq_lseek,
	.release = seq_release_private,
};

static const struct file_operations gfs2_sbstats_fops = {
	.owner   = THIS_MODULE,
	.open	 = gfs2_sbstats_open,
2081 2082
	.read    = seq_read,
	.llseek  = seq_lseek,
2083
	.release = seq_release_private,
2084 2085 2086 2087
};

int gfs2_create_debugfs_file(struct gfs2_sbd *sdp)
{
R
Robert Peterson 已提交
2088 2089 2090 2091 2092 2093
	sdp->debugfs_dir = debugfs_create_dir(sdp->sd_table_name, gfs2_root);
	if (!sdp->debugfs_dir)
		return -ENOMEM;
	sdp->debugfs_dentry_glocks = debugfs_create_file("glocks",
							 S_IFREG | S_IRUGO,
							 sdp->debugfs_dir, sdp,
2094
							 &gfs2_glocks_fops);
R
Robert Peterson 已提交
2095
	if (!sdp->debugfs_dentry_glocks)
2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110
		goto fail;

	sdp->debugfs_dentry_glstats = debugfs_create_file("glstats",
							S_IFREG | S_IRUGO,
							sdp->debugfs_dir, sdp,
							&gfs2_glstats_fops);
	if (!sdp->debugfs_dentry_glstats)
		goto fail;

	sdp->debugfs_dentry_sbstats = debugfs_create_file("sbstats",
							S_IFREG | S_IRUGO,
							sdp->debugfs_dir, sdp,
							&gfs2_sbstats_fops);
	if (!sdp->debugfs_dentry_sbstats)
		goto fail;
2111 2112

	return 0;
2113 2114 2115
fail:
	gfs2_delete_debugfs_file(sdp);
	return -ENOMEM;
2116 2117 2118 2119
}

void gfs2_delete_debugfs_file(struct gfs2_sbd *sdp)
{
2120
	if (sdp->debugfs_dir) {
R
Robert Peterson 已提交
2121 2122 2123 2124
		if (sdp->debugfs_dentry_glocks) {
			debugfs_remove(sdp->debugfs_dentry_glocks);
			sdp->debugfs_dentry_glocks = NULL;
		}
2125 2126 2127 2128 2129 2130 2131 2132
		if (sdp->debugfs_dentry_glstats) {
			debugfs_remove(sdp->debugfs_dentry_glstats);
			sdp->debugfs_dentry_glstats = NULL;
		}
		if (sdp->debugfs_dentry_sbstats) {
			debugfs_remove(sdp->debugfs_dentry_sbstats);
			sdp->debugfs_dentry_sbstats = NULL;
		}
R
Robert Peterson 已提交
2133 2134 2135
		debugfs_remove(sdp->debugfs_dir);
		sdp->debugfs_dir = NULL;
	}
2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146
}

int gfs2_register_debugfs(void)
{
	gfs2_root = debugfs_create_dir("gfs2", NULL);
	return gfs2_root ? 0 : -ENOMEM;
}

void gfs2_unregister_debugfs(void)
{
	debugfs_remove(gfs2_root);
R
Robert Peterson 已提交
2147
	gfs2_root = NULL;
2148
}