super.c 37.3 KB
Newer Older
D
David Teigland 已提交
1 2
/*
 * Copyright (C) Sistina Software, Inc.  1997-2003 All rights reserved.
B
Bob Peterson 已提交
3
 * Copyright (C) 2004-2007 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
#include <linux/bio.h>
D
David Teigland 已提交
11 12 13 14 15
#include <linux/sched.h>
#include <linux/slab.h>
#include <linux/spinlock.h>
#include <linux/completion.h>
#include <linux/buffer_head.h>
16 17 18 19 20
#include <linux/statfs.h>
#include <linux/seq_file.h>
#include <linux/mount.h>
#include <linux/kthread.h>
#include <linux/delay.h>
21
#include <linux/gfs2_ondisk.h>
22 23
#include <linux/crc32.h>
#include <linux/time.h>
24
#include <linux/wait.h>
25
#include <linux/writeback.h>
26
#include <linux/backing-dev.h>
D
David Teigland 已提交
27 28

#include "gfs2.h"
29
#include "incore.h"
D
David Teigland 已提交
30 31 32 33 34 35 36 37 38 39 40 41
#include "bmap.h"
#include "dir.h"
#include "glock.h"
#include "glops.h"
#include "inode.h"
#include "log.h"
#include "meta_io.h"
#include "quota.h"
#include "recovery.h"
#include "rgrp.h"
#include "super.h"
#include "trans.h"
42
#include "util.h"
43
#include "sys.h"
44
#include "xattr.h"
45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73

#define args_neq(a1, a2, x) ((a1)->ar_##x != (a2)->ar_##x)

enum {
	Opt_lockproto,
	Opt_locktable,
	Opt_hostdata,
	Opt_spectator,
	Opt_ignore_local_fs,
	Opt_localflocks,
	Opt_localcaching,
	Opt_debug,
	Opt_nodebug,
	Opt_upgrade,
	Opt_acl,
	Opt_noacl,
	Opt_quota_off,
	Opt_quota_account,
	Opt_quota_on,
	Opt_quota,
	Opt_noquota,
	Opt_suiddir,
	Opt_nosuiddir,
	Opt_data_writeback,
	Opt_data_ordered,
	Opt_meta,
	Opt_discard,
	Opt_nodiscard,
	Opt_commit,
74 75
	Opt_err_withdraw,
	Opt_err_panic,
76 77 78
	Opt_statfs_quantum,
	Opt_statfs_percent,
	Opt_quota_quantum,
79 80
	Opt_barrier,
	Opt_nobarrier,
81 82
	Opt_rgrplvb,
	Opt_norgrplvb,
83 84 85 86 87 88 89 90
	Opt_error,
};

static const match_table_t tokens = {
	{Opt_lockproto, "lockproto=%s"},
	{Opt_locktable, "locktable=%s"},
	{Opt_hostdata, "hostdata=%s"},
	{Opt_spectator, "spectator"},
91
	{Opt_spectator, "norecovery"},
92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112
	{Opt_ignore_local_fs, "ignore_local_fs"},
	{Opt_localflocks, "localflocks"},
	{Opt_localcaching, "localcaching"},
	{Opt_debug, "debug"},
	{Opt_nodebug, "nodebug"},
	{Opt_upgrade, "upgrade"},
	{Opt_acl, "acl"},
	{Opt_noacl, "noacl"},
	{Opt_quota_off, "quota=off"},
	{Opt_quota_account, "quota=account"},
	{Opt_quota_on, "quota=on"},
	{Opt_quota, "quota"},
	{Opt_noquota, "noquota"},
	{Opt_suiddir, "suiddir"},
	{Opt_nosuiddir, "nosuiddir"},
	{Opt_data_writeback, "data=writeback"},
	{Opt_data_ordered, "data=ordered"},
	{Opt_meta, "meta"},
	{Opt_discard, "discard"},
	{Opt_nodiscard, "nodiscard"},
	{Opt_commit, "commit=%d"},
113 114
	{Opt_err_withdraw, "errors=withdraw"},
	{Opt_err_panic, "errors=panic"},
115 116 117
	{Opt_statfs_quantum, "statfs_quantum=%d"},
	{Opt_statfs_percent, "statfs_percent=%d"},
	{Opt_quota_quantum, "quota_quantum=%d"},
118 119
	{Opt_barrier, "barrier"},
	{Opt_nobarrier, "nobarrier"},
120 121
	{Opt_rgrplvb, "rgrplvb"},
	{Opt_norgrplvb, "norgrplvb"},
122 123 124 125 126
	{Opt_error, NULL}
};

/**
 * gfs2_mount_args - Parse mount options
127 128
 * @args: The structure into which the parsed options will be written
 * @options: The options to parse
129 130 131 132
 *
 * Return: errno
 */

133
int gfs2_mount_args(struct gfs2_args *args, char *options)
134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167
{
	char *o;
	int token;
	substring_t tmp[MAX_OPT_ARGS];
	int rv;

	/* Split the options into tokens with the "," character and
	   process them */

	while (1) {
		o = strsep(&options, ",");
		if (o == NULL)
			break;
		if (*o == '\0')
			continue;

		token = match_token(o, tokens, tmp);
		switch (token) {
		case Opt_lockproto:
			match_strlcpy(args->ar_lockproto, &tmp[0],
				      GFS2_LOCKNAME_LEN);
			break;
		case Opt_locktable:
			match_strlcpy(args->ar_locktable, &tmp[0],
				      GFS2_LOCKNAME_LEN);
			break;
		case Opt_hostdata:
			match_strlcpy(args->ar_hostdata, &tmp[0],
				      GFS2_LOCKNAME_LEN);
			break;
		case Opt_spectator:
			args->ar_spectator = 1;
			break;
		case Opt_ignore_local_fs:
168
			/* Retained for backwards compat only */
169 170 171 172 173
			break;
		case Opt_localflocks:
			args->ar_localflocks = 1;
			break;
		case Opt_localcaching:
174
			/* Retained for backwards compat only */
175 176
			break;
		case Opt_debug:
177
			if (args->ar_errors == GFS2_ERRORS_PANIC) {
178
				printk(KERN_WARNING "GFS2: -o debug and -o errors=panic "
179 180 181
				       "are mutually exclusive.\n");
				return -EINVAL;
			}
182 183 184 185 186 187
			args->ar_debug = 1;
			break;
		case Opt_nodebug:
			args->ar_debug = 0;
			break;
		case Opt_upgrade:
188
			/* Retained for backwards compat only */
189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230
			break;
		case Opt_acl:
			args->ar_posix_acl = 1;
			break;
		case Opt_noacl:
			args->ar_posix_acl = 0;
			break;
		case Opt_quota_off:
		case Opt_noquota:
			args->ar_quota = GFS2_QUOTA_OFF;
			break;
		case Opt_quota_account:
			args->ar_quota = GFS2_QUOTA_ACCOUNT;
			break;
		case Opt_quota_on:
		case Opt_quota:
			args->ar_quota = GFS2_QUOTA_ON;
			break;
		case Opt_suiddir:
			args->ar_suiddir = 1;
			break;
		case Opt_nosuiddir:
			args->ar_suiddir = 0;
			break;
		case Opt_data_writeback:
			args->ar_data = GFS2_DATA_WRITEBACK;
			break;
		case Opt_data_ordered:
			args->ar_data = GFS2_DATA_ORDERED;
			break;
		case Opt_meta:
			args->ar_meta = 1;
			break;
		case Opt_discard:
			args->ar_discard = 1;
			break;
		case Opt_nodiscard:
			args->ar_discard = 0;
			break;
		case Opt_commit:
			rv = match_int(&tmp[0], &args->ar_commit);
			if (rv || args->ar_commit <= 0) {
231
				printk(KERN_WARNING "GFS2: commit mount option requires a positive numeric argument\n");
232 233 234
				return rv ? rv : -EINVAL;
			}
			break;
235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256
		case Opt_statfs_quantum:
			rv = match_int(&tmp[0], &args->ar_statfs_quantum);
			if (rv || args->ar_statfs_quantum < 0) {
				printk(KERN_WARNING "GFS2: statfs_quantum mount option requires a non-negative numeric argument\n");
				return rv ? rv : -EINVAL;
			}
			break;
		case Opt_quota_quantum:
			rv = match_int(&tmp[0], &args->ar_quota_quantum);
			if (rv || args->ar_quota_quantum <= 0) {
				printk(KERN_WARNING "GFS2: quota_quantum mount option requires a positive numeric argument\n");
				return rv ? rv : -EINVAL;
			}
			break;
		case Opt_statfs_percent:
			rv = match_int(&tmp[0], &args->ar_statfs_percent);
			if (rv || args->ar_statfs_percent < 0 ||
			    args->ar_statfs_percent > 100) {
				printk(KERN_WARNING "statfs_percent mount option requires a numeric argument between 0 and 100\n");
				return rv ? rv : -EINVAL;
			}
			break;
257 258 259 260 261
		case Opt_err_withdraw:
			args->ar_errors = GFS2_ERRORS_WITHDRAW;
			break;
		case Opt_err_panic:
			if (args->ar_debug) {
262
				printk(KERN_WARNING "GFS2: -o debug and -o errors=panic "
263 264 265 266 267
					"are mutually exclusive.\n");
				return -EINVAL;
			}
			args->ar_errors = GFS2_ERRORS_PANIC;
			break;
268 269 270 271 272 273
		case Opt_barrier:
			args->ar_nobarrier = 0;
			break;
		case Opt_nobarrier:
			args->ar_nobarrier = 1;
			break;
274 275 276 277 278 279
		case Opt_rgrplvb:
			args->ar_rgrplvb = 1;
			break;
		case Opt_norgrplvb:
			args->ar_rgrplvb = 0;
			break;
280 281
		case Opt_error:
		default:
282
			printk(KERN_WARNING "GFS2: invalid mount option: %s\n", o);
283 284 285 286 287 288
			return -EINVAL;
		}
	}

	return 0;
}
D
David Teigland 已提交
289

290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323
/**
 * gfs2_jindex_free - Clear all the journal index information
 * @sdp: The GFS2 superblock
 *
 */

void gfs2_jindex_free(struct gfs2_sbd *sdp)
{
	struct list_head list, *head;
	struct gfs2_jdesc *jd;
	struct gfs2_journal_extent *jext;

	spin_lock(&sdp->sd_jindex_spin);
	list_add(&list, &sdp->sd_jindex_list);
	list_del_init(&sdp->sd_jindex_list);
	sdp->sd_journals = 0;
	spin_unlock(&sdp->sd_jindex_spin);

	while (!list_empty(&list)) {
		jd = list_entry(list.next, struct gfs2_jdesc, jd_list);
		head = &jd->extent_list;
		while (!list_empty(head)) {
			jext = list_entry(head->next,
					  struct gfs2_journal_extent,
					  extent_list);
			list_del(&jext->extent_list);
			kfree(jext);
		}
		list_del(&jd->jd_list);
		iput(jd->jd_inode);
		kfree(jd);
	}
}

D
David Teigland 已提交
324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354
static struct gfs2_jdesc *jdesc_find_i(struct list_head *head, unsigned int jid)
{
	struct gfs2_jdesc *jd;
	int found = 0;

	list_for_each_entry(jd, head, jd_list) {
		if (jd->jd_jid == jid) {
			found = 1;
			break;
		}
	}

	if (!found)
		jd = NULL;

	return jd;
}

struct gfs2_jdesc *gfs2_jdesc_find(struct gfs2_sbd *sdp, unsigned int jid)
{
	struct gfs2_jdesc *jd;

	spin_lock(&sdp->sd_jindex_spin);
	jd = jdesc_find_i(&sdp->sd_jindex_list, jid);
	spin_unlock(&sdp->sd_jindex_spin);

	return jd;
}

int gfs2_jdesc_check(struct gfs2_jdesc *jd)
{
355 356
	struct gfs2_inode *ip = GFS2_I(jd->jd_inode);
	struct gfs2_sbd *sdp = GFS2_SB(jd->jd_inode);
S
Steven Whitehouse 已提交
357
	u64 size = i_size_read(jd->jd_inode);
D
David Teigland 已提交
358

S
Steven Whitehouse 已提交
359
	if (gfs2_check_internal_file_size(jd->jd_inode, 8 << 20, 1 << 30))
D
David Teigland 已提交
360 361
		return -EIO;

S
Steven Whitehouse 已提交
362 363 364
	jd->jd_blocks = size >> sdp->sd_sb.sb_bsize_shift;

	if (gfs2_write_alloc_required(ip, 0, size)) {
D
David Teigland 已提交
365
		gfs2_consist_inode(ip);
366
		return -EIO;
D
David Teigland 已提交
367 368
	}

369
	return 0;
D
David Teigland 已提交
370 371 372 373 374 375 376 377 378 379 380
}

/**
 * gfs2_make_fs_rw - Turn a Read-Only FS into a Read-Write one
 * @sdp: the filesystem
 *
 * Returns: errno
 */

int gfs2_make_fs_rw(struct gfs2_sbd *sdp)
{
381
	struct gfs2_inode *ip = GFS2_I(sdp->sd_jdesc->jd_inode);
382
	struct gfs2_glock *j_gl = ip->i_gl;
D
David Teigland 已提交
383
	struct gfs2_holder t_gh;
A
Al Viro 已提交
384
	struct gfs2_log_header_host head;
D
David Teigland 已提交
385 386
	int error;

387
	error = gfs2_glock_nq_init(sdp->sd_trans_gl, LM_ST_SHARED, 0, &t_gh);
D
David Teigland 已提交
388 389 390
	if (error)
		return error;

391
	j_gl->gl_ops->go_inval(j_gl, DIO_METADATA);
D
David Teigland 已提交
392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408

	error = gfs2_find_jhead(sdp->sd_jdesc, &head);
	if (error)
		goto fail;

	if (!(head.lh_flags & GFS2_LOG_HEAD_UNMOUNT)) {
		gfs2_consist(sdp);
		error = -EIO;
		goto fail;
	}

	/*  Initialize some head of the log stuff  */
	sdp->sd_log_sequence = head.lh_sequence + 1;
	gfs2_log_pointers_init(sdp, head.lh_blkno);

	error = gfs2_quota_init(sdp);
	if (error)
409
		goto fail;
D
David Teigland 已提交
410 411 412 413 414 415 416

	set_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags);

	gfs2_glock_dq_uninit(&t_gh);

	return 0;

417
fail:
D
David Teigland 已提交
418 419 420 421 422 423
	t_gh.gh_flags |= GL_NOCACHE;
	gfs2_glock_dq_uninit(&t_gh);

	return error;
}

424
void gfs2_statfs_change_in(struct gfs2_statfs_change_host *sc, const void *buf)
425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441
{
	const struct gfs2_statfs_change *str = buf;

	sc->sc_total = be64_to_cpu(str->sc_total);
	sc->sc_free = be64_to_cpu(str->sc_free);
	sc->sc_dinodes = be64_to_cpu(str->sc_dinodes);
}

static void gfs2_statfs_change_out(const struct gfs2_statfs_change_host *sc, void *buf)
{
	struct gfs2_statfs_change *str = buf;

	str->sc_total = cpu_to_be64(sc->sc_total);
	str->sc_free = cpu_to_be64(sc->sc_free);
	str->sc_dinodes = cpu_to_be64(sc->sc_dinodes);
}

D
David Teigland 已提交
442 443
int gfs2_statfs_init(struct gfs2_sbd *sdp)
{
444
	struct gfs2_inode *m_ip = GFS2_I(sdp->sd_statfs_inode);
445
	struct gfs2_statfs_change_host *m_sc = &sdp->sd_statfs_master;
446
	struct gfs2_inode *l_ip = GFS2_I(sdp->sd_sc_inode);
447
	struct gfs2_statfs_change_host *l_sc = &sdp->sd_statfs_local;
D
David Teigland 已提交
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
	struct buffer_head *m_bh, *l_bh;
	struct gfs2_holder gh;
	int error;

	error = gfs2_glock_nq_init(m_ip->i_gl, LM_ST_EXCLUSIVE, GL_NOCACHE,
				   &gh);
	if (error)
		return error;

	error = gfs2_meta_inode_buffer(m_ip, &m_bh);
	if (error)
		goto out;

	if (sdp->sd_args.ar_spectator) {
		spin_lock(&sdp->sd_statfs_spin);
		gfs2_statfs_change_in(m_sc, m_bh->b_data +
				      sizeof(struct gfs2_dinode));
		spin_unlock(&sdp->sd_statfs_spin);
	} else {
		error = gfs2_meta_inode_buffer(l_ip, &l_bh);
		if (error)
			goto out_m_bh;

		spin_lock(&sdp->sd_statfs_spin);
		gfs2_statfs_change_in(m_sc, m_bh->b_data +
				      sizeof(struct gfs2_dinode));
		gfs2_statfs_change_in(l_sc, l_bh->b_data +
				      sizeof(struct gfs2_dinode));
		spin_unlock(&sdp->sd_statfs_spin);

		brelse(l_bh);
	}

481
out_m_bh:
D
David Teigland 已提交
482
	brelse(m_bh);
483
out:
D
David Teigland 已提交
484 485 486 487
	gfs2_glock_dq_uninit(&gh);
	return 0;
}

488 489
void gfs2_statfs_change(struct gfs2_sbd *sdp, s64 total, s64 free,
			s64 dinodes)
D
David Teigland 已提交
490
{
491
	struct gfs2_inode *l_ip = GFS2_I(sdp->sd_sc_inode);
492
	struct gfs2_statfs_change_host *l_sc = &sdp->sd_statfs_local;
493
	struct gfs2_statfs_change_host *m_sc = &sdp->sd_statfs_master;
D
David Teigland 已提交
494
	struct buffer_head *l_bh;
495 496
	s64 x, y;
	int need_sync = 0;
D
David Teigland 已提交
497 498 499 500 501 502
	int error;

	error = gfs2_meta_inode_buffer(l_ip, &l_bh);
	if (error)
		return;

503
	gfs2_trans_add_meta(l_ip->i_gl, l_bh);
D
David Teigland 已提交
504 505 506 507 508

	spin_lock(&sdp->sd_statfs_spin);
	l_sc->sc_total += total;
	l_sc->sc_free += free;
	l_sc->sc_dinodes += dinodes;
509
	gfs2_statfs_change_out(l_sc, l_bh->b_data + sizeof(struct gfs2_dinode));
510 511 512 513 514 515
	if (sdp->sd_args.ar_statfs_percent) {
		x = 100 * l_sc->sc_free;
		y = m_sc->sc_free * sdp->sd_args.ar_statfs_percent;
		if (x >= y || x <= -y)
			need_sync = 1;
	}
D
David Teigland 已提交
516 517 518
	spin_unlock(&sdp->sd_statfs_spin);

	brelse(l_bh);
519
	if (need_sync)
520
		gfs2_wake_up_statfs(sdp);
D
David Teigland 已提交
521 522
}

523 524 525 526 527 528 529 530
void update_statfs(struct gfs2_sbd *sdp, struct buffer_head *m_bh,
		   struct buffer_head *l_bh)
{
	struct gfs2_inode *m_ip = GFS2_I(sdp->sd_statfs_inode);
	struct gfs2_inode *l_ip = GFS2_I(sdp->sd_sc_inode);
	struct gfs2_statfs_change_host *m_sc = &sdp->sd_statfs_master;
	struct gfs2_statfs_change_host *l_sc = &sdp->sd_statfs_local;

531
	gfs2_trans_add_meta(l_ip->i_gl, l_bh);
532 533 534 535 536 537 538 539 540 541

	spin_lock(&sdp->sd_statfs_spin);
	m_sc->sc_total += l_sc->sc_total;
	m_sc->sc_free += l_sc->sc_free;
	m_sc->sc_dinodes += l_sc->sc_dinodes;
	memset(l_sc, 0, sizeof(struct gfs2_statfs_change));
	memset(l_bh->b_data + sizeof(struct gfs2_dinode),
	       0, sizeof(struct gfs2_statfs_change));
	spin_unlock(&sdp->sd_statfs_spin);

542
	gfs2_trans_add_meta(m_ip->i_gl, m_bh);
543 544 545
	gfs2_statfs_change_out(m_sc, m_bh->b_data + sizeof(struct gfs2_dinode));
}

546
int gfs2_statfs_sync(struct super_block *sb, int type)
D
David Teigland 已提交
547
{
548
	struct gfs2_sbd *sdp = sb->s_fs_info;
549 550
	struct gfs2_inode *m_ip = GFS2_I(sdp->sd_statfs_inode);
	struct gfs2_inode *l_ip = GFS2_I(sdp->sd_sc_inode);
551 552
	struct gfs2_statfs_change_host *m_sc = &sdp->sd_statfs_master;
	struct gfs2_statfs_change_host *l_sc = &sdp->sd_statfs_local;
D
David Teigland 已提交
553 554 555 556 557 558 559 560 561 562 563 564 565 566 567
	struct gfs2_holder gh;
	struct buffer_head *m_bh, *l_bh;
	int error;

	error = gfs2_glock_nq_init(m_ip->i_gl, LM_ST_EXCLUSIVE, GL_NOCACHE,
				   &gh);
	if (error)
		return error;

	error = gfs2_meta_inode_buffer(m_ip, &m_bh);
	if (error)
		goto out;

	spin_lock(&sdp->sd_statfs_spin);
	gfs2_statfs_change_in(m_sc, m_bh->b_data +
568
			      sizeof(struct gfs2_dinode));
D
David Teigland 已提交
569 570 571 572 573 574 575 576 577 578 579 580 581 582
	if (!l_sc->sc_total && !l_sc->sc_free && !l_sc->sc_dinodes) {
		spin_unlock(&sdp->sd_statfs_spin);
		goto out_bh;
	}
	spin_unlock(&sdp->sd_statfs_spin);

	error = gfs2_meta_inode_buffer(l_ip, &l_bh);
	if (error)
		goto out_bh;

	error = gfs2_trans_begin(sdp, 2 * RES_DINODE, 0);
	if (error)
		goto out_bh2;

583
	update_statfs(sdp, m_bh, l_bh);
584
	sdp->sd_statfs_force_sync = 0;
D
David Teigland 已提交
585 586 587

	gfs2_trans_end(sdp);

588
out_bh2:
D
David Teigland 已提交
589
	brelse(l_bh);
590
out_bh:
D
David Teigland 已提交
591
	brelse(m_bh);
592
out:
D
David Teigland 已提交
593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611
	gfs2_glock_dq_uninit(&gh);
	return error;
}

struct lfcc {
	struct list_head list;
	struct gfs2_holder gh;
};

/**
 * gfs2_lock_fs_check_clean - Stop all writes to the FS and check that all
 *                            journals are clean
 * @sdp: the file system
 * @state: the state to put the transaction lock into
 * @t_gh: the hold on the transaction lock
 *
 * Returns: errno
 */

612 613
static int gfs2_lock_fs_check_clean(struct gfs2_sbd *sdp,
				    struct gfs2_holder *t_gh)
D
David Teigland 已提交
614
{
615
	struct gfs2_inode *ip;
D
David Teigland 已提交
616 617 618
	struct gfs2_jdesc *jd;
	struct lfcc *lfcc;
	LIST_HEAD(list);
A
Al Viro 已提交
619
	struct gfs2_log_header_host lh;
D
David Teigland 已提交
620 621 622 623 624 625 626 627
	int error;

	list_for_each_entry(jd, &sdp->sd_jindex_list, jd_list) {
		lfcc = kmalloc(sizeof(struct lfcc), GFP_KERNEL);
		if (!lfcc) {
			error = -ENOMEM;
			goto out;
		}
628 629
		ip = GFS2_I(jd->jd_inode);
		error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, 0, &lfcc->gh);
D
David Teigland 已提交
630 631 632 633 634 635 636 637
		if (error) {
			kfree(lfcc);
			goto out;
		}
		list_add(&lfcc->list, &list);
	}

	error = gfs2_glock_nq_init(sdp->sd_trans_gl, LM_ST_DEFERRED,
638
				   GL_NOCACHE, t_gh);
D
David Teigland 已提交
639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655

	list_for_each_entry(jd, &sdp->sd_jindex_list, jd_list) {
		error = gfs2_jdesc_check(jd);
		if (error)
			break;
		error = gfs2_find_jhead(jd, &lh);
		if (error)
			break;
		if (!(lh.lh_flags & GFS2_LOG_HEAD_UNMOUNT)) {
			error = -EBUSY;
			break;
		}
	}

	if (error)
		gfs2_glock_dq_uninit(t_gh);

656
out:
D
David Teigland 已提交
657 658 659 660 661 662 663 664 665
	while (!list_empty(&list)) {
		lfcc = list_entry(list.next, struct lfcc, list);
		list_del(&lfcc->list);
		gfs2_glock_dq_uninit(&lfcc->gh);
		kfree(lfcc);
	}
	return error;
}

666 667 668 669 670 671 672 673 674 675
void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf)
{
	struct gfs2_dinode *str = buf;

	str->di_header.mh_magic = cpu_to_be32(GFS2_MAGIC);
	str->di_header.mh_type = cpu_to_be32(GFS2_METATYPE_DI);
	str->di_header.mh_format = cpu_to_be32(GFS2_FORMAT_DI);
	str->di_num.no_addr = cpu_to_be64(ip->i_no_addr);
	str->di_num.no_formal_ino = cpu_to_be64(ip->i_no_formal_ino);
	str->di_mode = cpu_to_be32(ip->i_inode.i_mode);
676 677
	str->di_uid = cpu_to_be32(i_uid_read(&ip->i_inode));
	str->di_gid = cpu_to_be32(i_gid_read(&ip->i_inode));
678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701
	str->di_nlink = cpu_to_be32(ip->i_inode.i_nlink);
	str->di_size = cpu_to_be64(i_size_read(&ip->i_inode));
	str->di_blocks = cpu_to_be64(gfs2_get_inode_blocks(&ip->i_inode));
	str->di_atime = cpu_to_be64(ip->i_inode.i_atime.tv_sec);
	str->di_mtime = cpu_to_be64(ip->i_inode.i_mtime.tv_sec);
	str->di_ctime = cpu_to_be64(ip->i_inode.i_ctime.tv_sec);

	str->di_goal_meta = cpu_to_be64(ip->i_goal);
	str->di_goal_data = cpu_to_be64(ip->i_goal);
	str->di_generation = cpu_to_be64(ip->i_generation);

	str->di_flags = cpu_to_be32(ip->i_diskflags);
	str->di_height = cpu_to_be16(ip->i_height);
	str->di_payload_format = cpu_to_be32(S_ISDIR(ip->i_inode.i_mode) &&
					     !(ip->i_diskflags & GFS2_DIF_EXHASH) ?
					     GFS2_FORMAT_DE : 0);
	str->di_depth = cpu_to_be16(ip->i_depth);
	str->di_entries = cpu_to_be32(ip->i_entries);

	str->di_eattr = cpu_to_be64(ip->i_eattr);
	str->di_atime_nsec = cpu_to_be32(ip->i_inode.i_atime.tv_nsec);
	str->di_mtime_nsec = cpu_to_be32(ip->i_inode.i_mtime.tv_nsec);
	str->di_ctime_nsec = cpu_to_be32(ip->i_inode.i_ctime.tv_nsec);
}
702 703 704 705

/**
 * gfs2_write_inode - Make sure the inode is stable on the disk
 * @inode: The inode
706
 * @wbc: The writeback control structure
707 708 709 710
 *
 * Returns: errno
 */

711
static int gfs2_write_inode(struct inode *inode, struct writeback_control *wbc)
712 713 714
{
	struct gfs2_inode *ip = GFS2_I(inode);
	struct gfs2_sbd *sdp = GFS2_SB(inode);
715
	struct address_space *metamapping = gfs2_glock2aspace(ip->i_gl);
716
	struct backing_dev_info *bdi = metamapping->backing_dev_info;
S
Steven Whitehouse 已提交
717 718
	int ret = 0;

719
	if (wbc->sync_mode == WB_SYNC_ALL)
720
		gfs2_log_flush(GFS2_SB(inode), ip->i_gl);
721 722
	if (bdi->dirty_exceeded)
		gfs2_ail1_flush(sdp, wbc);
723 724
	else
		filemap_fdatawrite(metamapping);
S
Steven Whitehouse 已提交
725
	if (wbc->sync_mode == WB_SYNC_ALL)
726 727 728
		ret = filemap_fdatawait(metamapping);
	if (ret)
		mark_inode_dirty_sync(inode);
729 730 731
	return ret;
}

S
Steven Whitehouse 已提交
732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764
/**
 * gfs2_dirty_inode - check for atime updates
 * @inode: The inode in question
 * @flags: The type of dirty
 *
 * Unfortunately it can be called under any combination of inode
 * glock and transaction lock, so we have to check carefully.
 *
 * At the moment this deals only with atime - it should be possible
 * to expand that role in future, once a review of the locking has
 * been carried out.
 */

static void gfs2_dirty_inode(struct inode *inode, int flags)
{
	struct gfs2_inode *ip = GFS2_I(inode);
	struct gfs2_sbd *sdp = GFS2_SB(inode);
	struct buffer_head *bh;
	struct gfs2_holder gh;
	int need_unlock = 0;
	int need_endtrans = 0;
	int ret;

	if (!(flags & (I_DIRTY_DATASYNC|I_DIRTY_SYNC)))
		return;

	if (!gfs2_glock_is_locked_by_me(ip->i_gl)) {
		ret = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh);
		if (ret) {
			fs_err(sdp, "dirty_inode: glock %d\n", ret);
			return;
		}
		need_unlock = 1;
765 766
	} else if (WARN_ON_ONCE(ip->i_gl->gl_state != LM_ST_EXCLUSIVE))
		return;
S
Steven Whitehouse 已提交
767 768 769 770 771 772 773 774 775 776 777 778

	if (current->journal_info == NULL) {
		ret = gfs2_trans_begin(sdp, RES_DINODE, 0);
		if (ret) {
			fs_err(sdp, "dirty_inode: gfs2_trans_begin %d\n", ret);
			goto out;
		}
		need_endtrans = 1;
	}

	ret = gfs2_meta_inode_buffer(ip, &bh);
	if (ret == 0) {
779
		gfs2_trans_add_meta(ip->i_gl, bh);
S
Steven Whitehouse 已提交
780 781 782 783 784 785 786 787 788 789 790
		gfs2_dinode_out(ip, bh->b_data);
		brelse(bh);
	}

	if (need_endtrans)
		gfs2_trans_end(sdp);
out:
	if (need_unlock)
		gfs2_glock_dq_uninit(&gh);
}

791 792 793 794 795 796 797 798 799 800 801 802
/**
 * gfs2_make_fs_ro - Turn a Read-Write FS into a Read-Only one
 * @sdp: the filesystem
 *
 * Returns: errno
 */

static int gfs2_make_fs_ro(struct gfs2_sbd *sdp)
{
	struct gfs2_holder t_gh;
	int error;

803
	flush_workqueue(gfs2_delete_workqueue);
804
	gfs2_quota_sync(sdp->sd_vfs, 0);
805
	gfs2_statfs_sync(sdp->sd_vfs, 0);
806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909

	error = gfs2_glock_nq_init(sdp->sd_trans_gl, LM_ST_SHARED, GL_NOCACHE,
				   &t_gh);
	if (error && !test_bit(SDF_SHUTDOWN, &sdp->sd_flags))
		return error;

	gfs2_meta_syncfs(sdp);
	gfs2_log_shutdown(sdp);

	clear_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags);

	if (t_gh.gh_gl)
		gfs2_glock_dq_uninit(&t_gh);

	gfs2_quota_cleanup(sdp);

	return error;
}

static int gfs2_umount_recovery_wait(void *word)
{
	schedule();
	return 0;
}

/**
 * gfs2_put_super - Unmount the filesystem
 * @sb: The VFS superblock
 *
 */

static void gfs2_put_super(struct super_block *sb)
{
	struct gfs2_sbd *sdp = sb->s_fs_info;
	int error;
	struct gfs2_jdesc *jd;

	/* No more recovery requests */
	set_bit(SDF_NORECOVERY, &sdp->sd_flags);
	smp_mb();

	/* Wait on outstanding recovery */
restart:
	spin_lock(&sdp->sd_jindex_spin);
	list_for_each_entry(jd, &sdp->sd_jindex_list, jd_list) {
		if (!test_bit(JDF_RECOVERY, &jd->jd_flags))
			continue;
		spin_unlock(&sdp->sd_jindex_spin);
		wait_on_bit(&jd->jd_flags, JDF_RECOVERY,
			    gfs2_umount_recovery_wait, TASK_UNINTERRUPTIBLE);
		goto restart;
	}
	spin_unlock(&sdp->sd_jindex_spin);

	kthread_stop(sdp->sd_quotad_process);
	kthread_stop(sdp->sd_logd_process);

	if (!(sb->s_flags & MS_RDONLY)) {
		error = gfs2_make_fs_ro(sdp);
		if (error)
			gfs2_io_error(sdp);
	}
	/*  At this point, we're through modifying the disk  */

	/*  Release stuff  */

	iput(sdp->sd_jindex);
	iput(sdp->sd_statfs_inode);
	iput(sdp->sd_rindex);
	iput(sdp->sd_quota_inode);

	gfs2_glock_put(sdp->sd_rename_gl);
	gfs2_glock_put(sdp->sd_trans_gl);

	if (!sdp->sd_args.ar_spectator) {
		gfs2_glock_dq_uninit(&sdp->sd_journal_gh);
		gfs2_glock_dq_uninit(&sdp->sd_jinode_gh);
		gfs2_glock_dq_uninit(&sdp->sd_sc_gh);
		gfs2_glock_dq_uninit(&sdp->sd_qc_gh);
		iput(sdp->sd_sc_inode);
		iput(sdp->sd_qc_inode);
	}

	gfs2_glock_dq_uninit(&sdp->sd_live_gh);
	gfs2_clear_rgrpd(sdp);
	gfs2_jindex_free(sdp);
	/*  Take apart glock structures and buffer lists  */
	gfs2_gl_hash_clear(sdp);
	/*  Unmount the locking protocol  */
	gfs2_lm_unmount(sdp);

	/*  At this point, we're through participating in the lockspace  */
	gfs2_sys_fs_del(sdp);
}

/**
 * gfs2_sync_fs - sync the filesystem
 * @sb: the superblock
 *
 * Flushes the log to disk.
 */

static int gfs2_sync_fs(struct super_block *sb, int wait)
{
910
	struct gfs2_sbd *sdp = sb->s_fs_info;
911 912

	gfs2_quota_sync(sb, -1);
913 914
	if (wait && sdp)
		gfs2_log_flush(sdp, NULL);
915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932
	return 0;
}

/**
 * gfs2_freeze - prevent further writes to the filesystem
 * @sb: the VFS structure for the filesystem
 *
 */

static int gfs2_freeze(struct super_block *sb)
{
	struct gfs2_sbd *sdp = sb->s_fs_info;
	int error;

	if (test_bit(SDF_SHUTDOWN, &sdp->sd_flags))
		return -EINVAL;

	for (;;) {
S
Steven Whitehouse 已提交
933
		error = gfs2_lock_fs_check_clean(sdp, &sdp->sd_freeze_gh);
934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960
		if (!error)
			break;

		switch (error) {
		case -EBUSY:
			fs_err(sdp, "waiting for recovery before freeze\n");
			break;

		default:
			fs_err(sdp, "error freezing FS: %d\n", error);
			break;
		}

		fs_err(sdp, "retrying...\n");
		msleep(1000);
	}
	return 0;
}

/**
 * gfs2_unfreeze - reallow writes to the filesystem
 * @sb: the VFS structure for the filesystem
 *
 */

static int gfs2_unfreeze(struct super_block *sb)
{
S
Steven Whitehouse 已提交
961 962 963
	struct gfs2_sbd *sdp = sb->s_fs_info;

	gfs2_glock_dq_uninit(&sdp->sd_freeze_gh);
964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 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 1036 1037 1038 1039 1040 1041 1042 1043 1044 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 1076 1077 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 1103 1104
	return 0;
}

/**
 * statfs_fill - fill in the sg for a given RG
 * @rgd: the RG
 * @sc: the sc structure
 *
 * Returns: 0 on success, -ESTALE if the LVB is invalid
 */

static int statfs_slow_fill(struct gfs2_rgrpd *rgd,
			    struct gfs2_statfs_change_host *sc)
{
	gfs2_rgrp_verify(rgd);
	sc->sc_total += rgd->rd_data;
	sc->sc_free += rgd->rd_free;
	sc->sc_dinodes += rgd->rd_dinodes;
	return 0;
}

/**
 * gfs2_statfs_slow - Stat a filesystem using asynchronous locking
 * @sdp: the filesystem
 * @sc: the sc info that will be returned
 *
 * Any error (other than a signal) will cause this routine to fall back
 * to the synchronous version.
 *
 * FIXME: This really shouldn't busy wait like this.
 *
 * Returns: errno
 */

static int gfs2_statfs_slow(struct gfs2_sbd *sdp, struct gfs2_statfs_change_host *sc)
{
	struct gfs2_rgrpd *rgd_next;
	struct gfs2_holder *gha, *gh;
	unsigned int slots = 64;
	unsigned int x;
	int done;
	int error = 0, err;

	memset(sc, 0, sizeof(struct gfs2_statfs_change_host));
	gha = kcalloc(slots, sizeof(struct gfs2_holder), GFP_KERNEL);
	if (!gha)
		return -ENOMEM;

	rgd_next = gfs2_rgrpd_get_first(sdp);

	for (;;) {
		done = 1;

		for (x = 0; x < slots; x++) {
			gh = gha + x;

			if (gh->gh_gl && gfs2_glock_poll(gh)) {
				err = gfs2_glock_wait(gh);
				if (err) {
					gfs2_holder_uninit(gh);
					error = err;
				} else {
					if (!error)
						error = statfs_slow_fill(
							gh->gh_gl->gl_object, sc);
					gfs2_glock_dq_uninit(gh);
				}
			}

			if (gh->gh_gl)
				done = 0;
			else if (rgd_next && !error) {
				error = gfs2_glock_nq_init(rgd_next->rd_gl,
							   LM_ST_SHARED,
							   GL_ASYNC,
							   gh);
				rgd_next = gfs2_rgrpd_get_next(rgd_next);
				done = 0;
			}

			if (signal_pending(current))
				error = -ERESTARTSYS;
		}

		if (done)
			break;

		yield();
	}

	kfree(gha);
	return error;
}

/**
 * gfs2_statfs_i - Do a statfs
 * @sdp: the filesystem
 * @sg: the sg structure
 *
 * Returns: errno
 */

static int gfs2_statfs_i(struct gfs2_sbd *sdp, struct gfs2_statfs_change_host *sc)
{
	struct gfs2_statfs_change_host *m_sc = &sdp->sd_statfs_master;
	struct gfs2_statfs_change_host *l_sc = &sdp->sd_statfs_local;

	spin_lock(&sdp->sd_statfs_spin);

	*sc = *m_sc;
	sc->sc_total += l_sc->sc_total;
	sc->sc_free += l_sc->sc_free;
	sc->sc_dinodes += l_sc->sc_dinodes;

	spin_unlock(&sdp->sd_statfs_spin);

	if (sc->sc_free < 0)
		sc->sc_free = 0;
	if (sc->sc_free > sc->sc_total)
		sc->sc_free = sc->sc_total;
	if (sc->sc_dinodes < 0)
		sc->sc_dinodes = 0;

	return 0;
}

/**
 * gfs2_statfs - Gather and return stats about the filesystem
 * @sb: The superblock
 * @statfsbuf: The buffer
 *
 * Returns: 0 on success or error code
 */

static int gfs2_statfs(struct dentry *dentry, struct kstatfs *buf)
{
	struct super_block *sb = dentry->d_inode->i_sb;
	struct gfs2_sbd *sdp = sb->s_fs_info;
	struct gfs2_statfs_change_host sc;
	int error;

1105 1106 1107 1108
	error = gfs2_rindex_update(sdp);
	if (error)
		return error;

1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145
	if (gfs2_tune_get(sdp, gt_statfs_slow))
		error = gfs2_statfs_slow(sdp, &sc);
	else
		error = gfs2_statfs_i(sdp, &sc);

	if (error)
		return error;

	buf->f_type = GFS2_MAGIC;
	buf->f_bsize = sdp->sd_sb.sb_bsize;
	buf->f_blocks = sc.sc_total;
	buf->f_bfree = sc.sc_free;
	buf->f_bavail = sc.sc_free;
	buf->f_files = sc.sc_dinodes + sc.sc_free;
	buf->f_ffree = sc.sc_free;
	buf->f_namelen = GFS2_FNAMESIZE;

	return 0;
}

/**
 * gfs2_remount_fs - called when the FS is remounted
 * @sb:  the filesystem
 * @flags:  the remount flags
 * @data:  extra data passed in (not used right now)
 *
 * Returns: errno
 */

static int gfs2_remount_fs(struct super_block *sb, int *flags, char *data)
{
	struct gfs2_sbd *sdp = sb->s_fs_info;
	struct gfs2_args args = sdp->sd_args; /* Default to current settings */
	struct gfs2_tune *gt = &sdp->sd_tune;
	int error;

	spin_lock(&gt->gt_spin);
1146
	args.ar_commit = gt->gt_logd_secs;
1147 1148 1149 1150 1151
	args.ar_quota_quantum = gt->gt_quota_quantum;
	if (gt->gt_statfs_slow)
		args.ar_statfs_quantum = 0;
	else
		args.ar_statfs_quantum = gt->gt_statfs_quantum;
1152
	spin_unlock(&gt->gt_spin);
1153
	error = gfs2_mount_args(&args, data);
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 1181 1182 1183 1184 1185
	if (error)
		return error;

	/* Not allowed to change locking details */
	if (strcmp(args.ar_lockproto, sdp->sd_args.ar_lockproto) ||
	    strcmp(args.ar_locktable, sdp->sd_args.ar_locktable) ||
	    strcmp(args.ar_hostdata, sdp->sd_args.ar_hostdata))
		return -EINVAL;

	/* Some flags must not be changed */
	if (args_neq(&args, &sdp->sd_args, spectator) ||
	    args_neq(&args, &sdp->sd_args, localflocks) ||
	    args_neq(&args, &sdp->sd_args, meta))
		return -EINVAL;

	if (sdp->sd_args.ar_spectator)
		*flags |= MS_RDONLY;

	if ((sb->s_flags ^ *flags) & MS_RDONLY) {
		if (*flags & MS_RDONLY)
			error = gfs2_make_fs_ro(sdp);
		else
			error = gfs2_make_fs_rw(sdp);
		if (error)
			return error;
	}

	sdp->sd_args = args;
	if (sdp->sd_args.ar_posix_acl)
		sb->s_flags |= MS_POSIXACL;
	else
		sb->s_flags &= ~MS_POSIXACL;
1186 1187 1188 1189
	if (sdp->sd_args.ar_nobarrier)
		set_bit(SDF_NOBARRIERS, &sdp->sd_flags);
	else
		clear_bit(SDF_NOBARRIERS, &sdp->sd_flags);
1190
	spin_lock(&gt->gt_spin);
1191
	gt->gt_logd_secs = args.ar_commit;
1192 1193 1194 1195 1196 1197 1198 1199 1200
	gt->gt_quota_quantum = args.ar_quota_quantum;
	if (args.ar_statfs_quantum) {
		gt->gt_statfs_slow = 0;
		gt->gt_statfs_quantum = args.ar_statfs_quantum;
	}
	else {
		gt->gt_statfs_slow = 1;
		gt->gt_statfs_quantum = 30;
	}
1201 1202
	spin_unlock(&gt->gt_spin);

1203
	gfs2_online_uevent(sdp);
1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221
	return 0;
}

/**
 * gfs2_drop_inode - Drop an inode (test for remote unlink)
 * @inode: The inode to drop
 *
 * If we've received a callback on an iopen lock then its because a
 * remote node tried to deallocate the inode but failed due to this node
 * still having the inode open. Here we mark the link count zero
 * since we know that it must have reached zero if the GLF_DEMOTE flag
 * is set on the iopen glock. If we didn't do a disk read since the
 * remote node removed the final link then we might otherwise miss
 * this event. This check ensures that this node will deallocate the
 * inode's blocks, or alternatively pass the baton on to another
 * node for later deallocation.
 */

1222
static int gfs2_drop_inode(struct inode *inode)
1223 1224 1225
{
	struct gfs2_inode *ip = GFS2_I(inode);

1226
	if (inode->i_nlink) {
1227 1228 1229 1230
		struct gfs2_glock *gl = ip->i_iopen_gh.gh_gl;
		if (gl && test_bit(GLF_DEMOTE, &gl->gl_flags))
			clear_nlink(inode);
	}
1231
	return generic_drop_inode(inode);
1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246
}

static int is_ancestor(const struct dentry *d1, const struct dentry *d2)
{
	do {
		if (d1 == d2)
			return 1;
		d1 = d1->d_parent;
	} while (!IS_ROOT(d1));
	return 0;
}

/**
 * gfs2_show_options - Show mount options for /proc/mounts
 * @s: seq_file structure
1247
 * @root: root of this (sub)tree
1248 1249 1250 1251
 *
 * Returns: 0 on success or error code
 */

1252
static int gfs2_show_options(struct seq_file *s, struct dentry *root)
1253
{
1254
	struct gfs2_sbd *sdp = root->d_sb->s_fs_info;
1255
	struct gfs2_args *args = &sdp->sd_args;
1256
	int val;
1257

1258
	if (is_ancestor(root, sdp->sd_master_dir))
1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310
		seq_printf(s, ",meta");
	if (args->ar_lockproto[0])
		seq_printf(s, ",lockproto=%s", args->ar_lockproto);
	if (args->ar_locktable[0])
		seq_printf(s, ",locktable=%s", args->ar_locktable);
	if (args->ar_hostdata[0])
		seq_printf(s, ",hostdata=%s", args->ar_hostdata);
	if (args->ar_spectator)
		seq_printf(s, ",spectator");
	if (args->ar_localflocks)
		seq_printf(s, ",localflocks");
	if (args->ar_debug)
		seq_printf(s, ",debug");
	if (args->ar_posix_acl)
		seq_printf(s, ",acl");
	if (args->ar_quota != GFS2_QUOTA_DEFAULT) {
		char *state;
		switch (args->ar_quota) {
		case GFS2_QUOTA_OFF:
			state = "off";
			break;
		case GFS2_QUOTA_ACCOUNT:
			state = "account";
			break;
		case GFS2_QUOTA_ON:
			state = "on";
			break;
		default:
			state = "unknown";
			break;
		}
		seq_printf(s, ",quota=%s", state);
	}
	if (args->ar_suiddir)
		seq_printf(s, ",suiddir");
	if (args->ar_data != GFS2_DATA_DEFAULT) {
		char *state;
		switch (args->ar_data) {
		case GFS2_DATA_WRITEBACK:
			state = "writeback";
			break;
		case GFS2_DATA_ORDERED:
			state = "ordered";
			break;
		default:
			state = "unknown";
			break;
		}
		seq_printf(s, ",data=%s", state);
	}
	if (args->ar_discard)
		seq_printf(s, ",discard");
1311 1312
	val = sdp->sd_tune.gt_logd_secs;
	if (val != 30)
1313 1314 1315 1316
		seq_printf(s, ",commit=%d", val);
	val = sdp->sd_tune.gt_statfs_quantum;
	if (val != 30)
		seq_printf(s, ",statfs_quantum=%d", val);
1317 1318
	else if (sdp->sd_tune.gt_statfs_slow)
		seq_puts(s, ",statfs_quantum=0");
1319 1320 1321 1322 1323
	val = sdp->sd_tune.gt_quota_quantum;
	if (val != 60)
		seq_printf(s, ",quota_quantum=%d", val);
	if (args->ar_statfs_percent)
		seq_printf(s, ",statfs_percent=%d", args->ar_statfs_percent);
1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339
	if (args->ar_errors != GFS2_ERRORS_DEFAULT) {
		const char *state;

		switch (args->ar_errors) {
		case GFS2_ERRORS_WITHDRAW:
			state = "withdraw";
			break;
		case GFS2_ERRORS_PANIC:
			state = "panic";
			break;
		default:
			state = "unknown";
			break;
		}
		seq_printf(s, ",errors=%s", state);
	}
1340 1341
	if (test_bit(SDF_NOBARRIERS, &sdp->sd_flags))
		seq_printf(s, ",nobarrier");
1342 1343
	if (test_bit(SDF_DEMOTE, &sdp->sd_flags))
		seq_printf(s, ",demote_interface_used");
1344 1345
	if (args->ar_rgrplvb)
		seq_printf(s, ",rgrplvb");
1346 1347 1348
	return 0;
}

1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366
static void gfs2_final_release_pages(struct gfs2_inode *ip)
{
	struct inode *inode = &ip->i_inode;
	struct gfs2_glock *gl = ip->i_gl;

	truncate_inode_pages(gfs2_glock2aspace(ip->i_gl), 0);
	truncate_inode_pages(&inode->i_data, 0);

	if (atomic_read(&gl->gl_revokes) == 0) {
		clear_bit(GLF_LFLUSH, &gl->gl_flags);
		clear_bit(GLF_DIRTY, &gl->gl_flags);
	}
}

static int gfs2_dinode_dealloc(struct gfs2_inode *ip)
{
	struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
	struct gfs2_rgrpd *rgd;
1367
	struct gfs2_holder gh;
1368 1369 1370
	int error;

	if (gfs2_get_inode_blocks(&ip->i_inode) != 1) {
1371
		gfs2_consist_inode(ip);
1372 1373 1374
		return -EIO;
	}

B
Bob Peterson 已提交
1375 1376 1377
	error = gfs2_rindex_update(sdp);
	if (error)
		return error;
1378

1379
	error = gfs2_quota_hold(ip, NO_UID_QUOTA_CHANGE, NO_GID_QUOTA_CHANGE);
1380
	if (error)
1381
		return error;
1382

S
Steven Whitehouse 已提交
1383
	rgd = gfs2_blk2rgrpd(sdp, ip->i_no_addr, 1);
1384 1385 1386
	if (!rgd) {
		gfs2_consist_inode(ip);
		error = -EIO;
1387
		goto out_qs;
1388 1389
	}

1390
	error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, &gh);
1391
	if (error)
1392
		goto out_qs;
1393

1394 1395
	error = gfs2_trans_begin(sdp, RES_RG_BIT + RES_STATFS + RES_QUOTA,
				 sdp->sd_jdesc->jd_blocks);
1396 1397 1398 1399 1400 1401 1402 1403 1404 1405
	if (error)
		goto out_rg_gunlock;

	gfs2_free_di(rgd, ip);

	gfs2_final_release_pages(ip);

	gfs2_trans_end(sdp);

out_rg_gunlock:
1406
	gfs2_glock_dq_uninit(&gh);
1407 1408 1409 1410 1411
out_qs:
	gfs2_quota_unhold(ip);
	return error;
}

1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425
/**
 * gfs2_evict_inode - Remove an inode from cache
 * @inode: The inode to evict
 *
 * There are three cases to consider:
 * 1. i_nlink == 0, we are final opener (and must deallocate)
 * 2. i_nlink == 0, we are not the final opener (and cannot deallocate)
 * 3. i_nlink > 0
 *
 * If the fs is read only, then we have to treat all cases as per #3
 * since we are unable to do any deallocation. The inode will be
 * deallocated by the next read/write node to attempt an allocation
 * in the same resource group
 *
1426 1427 1428 1429 1430 1431 1432
 * We have to (at the moment) hold the inodes main lock to cover
 * the gap between unlocking the shared lock on the iopen lock and
 * taking the exclusive lock. I'd rather do a shared -> exclusive
 * conversion on the iopen lock, but we can change that later. This
 * is safe, just less efficient.
 */

A
Al Viro 已提交
1433
static void gfs2_evict_inode(struct inode *inode)
1434
{
1435 1436
	struct super_block *sb = inode->i_sb;
	struct gfs2_sbd *sdp = sb->s_fs_info;
1437 1438 1439 1440
	struct gfs2_inode *ip = GFS2_I(inode);
	struct gfs2_holder gh;
	int error;

1441
	if (inode->i_nlink || (sb->s_flags & MS_RDONLY))
A
Al Viro 已提交
1442 1443
		goto out;

1444 1445
	/* Must not read inode block until block type has been verified */
	error = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, GL_SKIP, &gh);
1446 1447 1448 1449 1450
	if (unlikely(error)) {
		gfs2_glock_dq_uninit(&ip->i_iopen_gh);
		goto out;
	}

1451 1452 1453 1454 1455
	if (!test_bit(GIF_ALLOC_FAILED, &ip->i_flags)) {
		error = gfs2_check_blk_type(sdp, ip->i_no_addr, GFS2_BLKST_UNLINKED);
		if (error)
			goto out_truncate;
	}
1456

1457 1458 1459 1460 1461 1462
	if (test_bit(GIF_INVALID, &ip->i_flags)) {
		error = gfs2_inode_refresh(ip);
		if (error)
			goto out_truncate;
	}

1463
	ip->i_iopen_gh.gh_flags |= GL_NOCACHE;
1464 1465 1466 1467 1468 1469
	gfs2_glock_dq_wait(&ip->i_iopen_gh);
	gfs2_holder_reinit(LM_ST_EXCLUSIVE, LM_FLAG_TRY_1CB | GL_NOCACHE, &ip->i_iopen_gh);
	error = gfs2_glock_nq(&ip->i_iopen_gh);
	if (error)
		goto out_truncate;

1470 1471
	/* Case 1 starts here */

1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491
	if (S_ISDIR(inode->i_mode) &&
	    (ip->i_diskflags & GFS2_DIF_EXHASH)) {
		error = gfs2_dir_exhash_dealloc(ip);
		if (error)
			goto out_unlock;
	}

	if (ip->i_eattr) {
		error = gfs2_ea_dealloc(ip);
		if (error)
			goto out_unlock;
	}

	if (!gfs2_is_stuffed(ip)) {
		error = gfs2_file_dealloc(ip);
		if (error)
			goto out_unlock;
	}

	error = gfs2_dinode_dealloc(ip);
1492
	goto out_unlock;
1493 1494

out_truncate:
1495
	gfs2_log_flush(sdp, ip->i_gl);
1496 1497 1498 1499 1500
	if (test_bit(GLF_DIRTY, &ip->i_gl->gl_flags)) {
		struct address_space *metamapping = gfs2_glock2aspace(ip->i_gl);
		filemap_fdatawrite(metamapping);
		filemap_fdatawait(metamapping);
	}
1501
	write_inode_now(inode, 1);
1502
	gfs2_ail_flush(ip->i_gl, 0);
1503

1504
	/* Case 2 starts here */
1505 1506 1507
	error = gfs2_trans_begin(sdp, 0, sdp->sd_jdesc->jd_blocks);
	if (error)
		goto out_unlock;
1508 1509
	/* Needs to be done before glock release & also in a transaction */
	truncate_inode_pages(&inode->i_data, 0);
1510 1511 1512
	gfs2_trans_end(sdp);

out_unlock:
1513
	/* Error path for case 1 */
B
Bob Peterson 已提交
1514
	if (gfs2_rs_active(ip->i_res))
1515
		gfs2_rs_deltree(ip, ip->i_res);
B
Bob Peterson 已提交
1516

1517 1518 1519 1520 1521
	if (test_bit(HIF_HOLDER, &ip->i_iopen_gh.gh_iflags))
		gfs2_glock_dq(&ip->i_iopen_gh);
	gfs2_holder_uninit(&ip->i_iopen_gh);
	gfs2_glock_dq_uninit(&gh);
	if (error && error != GLR_TRYFAILED && error != -EROFS)
A
Al Viro 已提交
1522
		fs_warn(sdp, "gfs2_evict_inode: %d\n", error);
1523
out:
1524
	/* Case 3 starts here */
1525
	truncate_inode_pages(&inode->i_data, 0);
1526
	gfs2_rs_delete(ip);
1527
	gfs2_ordered_del_inode(ip);
1528
	clear_inode(inode);
1529
	gfs2_dir_hash_inval(ip);
A
Al Viro 已提交
1530
	ip->i_gl->gl_object = NULL;
1531
	flush_delayed_work(&ip->i_gl->gl_work);
1532
	gfs2_glock_add_to_lru(ip->i_gl);
A
Al Viro 已提交
1533 1534 1535 1536 1537 1538
	gfs2_glock_put(ip->i_gl);
	ip->i_gl = NULL;
	if (ip->i_iopen_gh.gh_gl) {
		ip->i_iopen_gh.gh_gl->gl_object = NULL;
		gfs2_glock_dq_uninit(&ip->i_iopen_gh);
	}
1539 1540 1541 1542 1543 1544 1545 1546 1547 1548
}

static struct inode *gfs2_alloc_inode(struct super_block *sb)
{
	struct gfs2_inode *ip;

	ip = kmem_cache_alloc(gfs2_inode_cachep, GFP_KERNEL);
	if (ip) {
		ip->i_flags = 0;
		ip->i_gl = NULL;
1549
		ip->i_rgd = NULL;
1550
		ip->i_res = NULL;
1551 1552 1553 1554
	}
	return &ip->i_inode;
}

N
Nick Piggin 已提交
1555
static void gfs2_i_callback(struct rcu_head *head)
1556
{
N
Nick Piggin 已提交
1557
	struct inode *inode = container_of(head, struct inode, i_rcu);
1558 1559 1560
	kmem_cache_free(gfs2_inode_cachep, inode);
}

N
Nick Piggin 已提交
1561 1562 1563 1564 1565
static void gfs2_destroy_inode(struct inode *inode)
{
	call_rcu(&inode->i_rcu, gfs2_i_callback);
}

1566 1567 1568 1569
const struct super_operations gfs2_super_ops = {
	.alloc_inode		= gfs2_alloc_inode,
	.destroy_inode		= gfs2_destroy_inode,
	.write_inode		= gfs2_write_inode,
S
Steven Whitehouse 已提交
1570
	.dirty_inode		= gfs2_dirty_inode,
A
Al Viro 已提交
1571
	.evict_inode		= gfs2_evict_inode,
1572 1573 1574 1575 1576 1577 1578 1579 1580 1581
	.put_super		= gfs2_put_super,
	.sync_fs		= gfs2_sync_fs,
	.freeze_fs 		= gfs2_freeze,
	.unfreeze_fs		= gfs2_unfreeze,
	.statfs			= gfs2_statfs,
	.remount_fs		= gfs2_remount_fs,
	.drop_inode		= gfs2_drop_inode,
	.show_options		= gfs2_show_options,
};