super.c 22.4 KB
Newer Older
D
David Teigland 已提交
1 2
/*
 * Copyright (C) Sistina Software, Inc.  1997-2003 All rights reserved.
3
 * Copyright (C) 2004-2006 Red Hat, Inc.  All rights reserved.
D
David Teigland 已提交
4 5 6
 *
 * This copyrighted material is made available to anyone wishing to use,
 * modify, copy, or redistribute it subject to the terms and conditions
7
 * of the GNU General Public License version 2.
D
David Teigland 已提交
8 9 10 11 12 13 14
 */

#include <linux/sched.h>
#include <linux/slab.h>
#include <linux/spinlock.h>
#include <linux/completion.h>
#include <linux/buffer_head.h>
15
#include <linux/crc32.h>
16
#include <linux/gfs2_ondisk.h>
17
#include <linux/bio.h>
18
#include <linux/lm_interface.h>
D
David Teigland 已提交
19 20

#include "gfs2.h"
21
#include "incore.h"
D
David Teigland 已提交
22 23 24 25 26 27 28 29 30 31 32 33
#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"
34
#include "util.h"
D
David Teigland 已提交
35

36
static const u32 gfs2_old_fs_formats[] = {
37 38 39
        0
};

40
static const u32 gfs2_old_multihost_formats[] = {
41 42 43
        0
};

D
David Teigland 已提交
44 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 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93
/**
 * gfs2_tune_init - Fill a gfs2_tune structure with default values
 * @gt: tune
 *
 */

void gfs2_tune_init(struct gfs2_tune *gt)
{
	spin_lock_init(&gt->gt_spin);

	gt->gt_ilimit = 100;
	gt->gt_ilimit_tries = 3;
	gt->gt_ilimit_min = 1;
	gt->gt_demote_secs = 300;
	gt->gt_incore_log_blocks = 1024;
	gt->gt_log_flush_secs = 60;
	gt->gt_jindex_refresh_secs = 60;
	gt->gt_scand_secs = 15;
	gt->gt_recoverd_secs = 60;
	gt->gt_logd_secs = 1;
	gt->gt_quotad_secs = 5;
	gt->gt_quota_simul_sync = 64;
	gt->gt_quota_warn_period = 10;
	gt->gt_quota_scale_num = 1;
	gt->gt_quota_scale_den = 1;
	gt->gt_quota_cache_secs = 300;
	gt->gt_quota_quantum = 60;
	gt->gt_atime_quantum = 3600;
	gt->gt_new_files_jdata = 0;
	gt->gt_new_files_directio = 0;
	gt->gt_max_readahead = 1 << 18;
	gt->gt_lockdump_size = 131072;
	gt->gt_stall_secs = 600;
	gt->gt_complain_secs = 10;
	gt->gt_reclaim_limit = 5000;
	gt->gt_statfs_quantum = 30;
	gt->gt_statfs_slow = 0;
}

/**
 * gfs2_check_sb - Check superblock
 * @sdp: the filesystem
 * @sb: The superblock
 * @silent: Don't print a message if the check fails
 *
 * Checks the version code of the FS is one that we understand how to
 * read and that the sizes of the various on-disk structures have not
 * changed.
 */

A
Al Viro 已提交
94
int gfs2_check_sb(struct gfs2_sbd *sdp, struct gfs2_sb_host *sb, int silent)
D
David Teigland 已提交
95 96 97 98 99 100
{
	unsigned int x;

	if (sb->sb_header.mh_magic != GFS2_MAGIC ||
	    sb->sb_header.mh_type != GFS2_METATYPE_SB) {
		if (!silent)
101
			printk(KERN_WARNING "GFS2: not a GFS2 filesystem\n");
D
David Teigland 已提交
102 103 104 105 106 107 108 109 110 111 112 113 114 115 116
		return -EINVAL;
	}

	/*  If format numbers match exactly, we're done.  */

	if (sb->sb_fs_format == GFS2_FORMAT_FS &&
	    sb->sb_multihost_format == GFS2_FORMAT_MULTI)
		return 0;

	if (sb->sb_fs_format != GFS2_FORMAT_FS) {
		for (x = 0; gfs2_old_fs_formats[x]; x++)
			if (gfs2_old_fs_formats[x] == sb->sb_fs_format)
				break;

		if (!gfs2_old_fs_formats[x]) {
117 118
			printk(KERN_WARNING
			       "GFS2: code version (%u, %u) is incompatible "
D
David Teigland 已提交
119 120 121
			       "with ondisk format (%u, %u)\n",
			       GFS2_FORMAT_FS, GFS2_FORMAT_MULTI,
			       sb->sb_fs_format, sb->sb_multihost_format);
122 123
			printk(KERN_WARNING
			       "GFS2: I don't know how to upgrade this FS\n");
D
David Teigland 已提交
124 125 126 127 128 129
			return -EINVAL;
		}
	}

	if (sb->sb_multihost_format != GFS2_FORMAT_MULTI) {
		for (x = 0; gfs2_old_multihost_formats[x]; x++)
130 131
			if (gfs2_old_multihost_formats[x] ==
			    sb->sb_multihost_format)
D
David Teigland 已提交
132 133 134
				break;

		if (!gfs2_old_multihost_formats[x]) {
135 136
			printk(KERN_WARNING
			       "GFS2: code version (%u, %u) is incompatible "
D
David Teigland 已提交
137 138 139
			       "with ondisk format (%u, %u)\n",
			       GFS2_FORMAT_FS, GFS2_FORMAT_MULTI,
			       sb->sb_fs_format, sb->sb_multihost_format);
140 141
			printk(KERN_WARNING
			       "GFS2: I don't know how to upgrade this FS\n");
D
David Teigland 已提交
142 143 144 145 146
			return -EINVAL;
		}
	}

	if (!sdp->sd_args.ar_upgrade) {
147 148
		printk(KERN_WARNING
		       "GFS2: code version (%u, %u) is incompatible "
D
David Teigland 已提交
149 150 151
		       "with ondisk format (%u, %u)\n",
		       GFS2_FORMAT_FS, GFS2_FORMAT_MULTI,
		       sb->sb_fs_format, sb->sb_multihost_format);
152 153
		printk(KERN_INFO
		       "GFS2: Use the \"upgrade\" mount option to upgrade "
D
David Teigland 已提交
154
		       "the FS\n");
155
		printk(KERN_INFO "GFS2: See the manual for more details\n");
D
David Teigland 已提交
156 157 158 159 160 161
		return -EINVAL;
	}

	return 0;
}

162 163 164 165 166 167 168 169 170

static int end_bio_io_page(struct bio *bio, unsigned int bytes_done, int error)
{
	struct page *page = bio->bi_private;
	if (bio->bi_size)
		return 1;

	if (!error)
		SetPageUptodate(page);
171 172
	else
		printk(KERN_WARNING "gfs2: error %d reading superblock\n", error);
173 174 175 176
	unlock_page(page);
	return 0;
}

177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194
/**
 * gfs2_read_super - Read the gfs2 super block from disk
 * @sb: The VFS super block
 * @sector: The location of the super block
 *
 * This uses the bio functions to read the super block from disk
 * because we want to be 100% sure that we never read cached data.
 * A super block is read twice only during each GFS2 mount and is
 * never written to by the filesystem. The first time its read no
 * locks are held, and the only details which are looked at are those
 * relating to the locking protocol. Once locking is up and working,
 * the sb is read again under the lock to establish the location of
 * the master directory (contains pointers to journals etc) and the
 * root directory.
 *
 * Returns: A page containing the sb or NULL
 */

195
struct page *gfs2_read_super(struct super_block *sb, sector_t sector)
196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213
{
	struct page *page;
	struct bio *bio;

	page = alloc_page(GFP_KERNEL);
	if (unlikely(!page))
		return NULL;

	ClearPageUptodate(page);
	ClearPageDirty(page);
	lock_page(page);

	bio = bio_alloc(GFP_KERNEL, 1);
	if (unlikely(!bio)) {
		__free_page(page);
		return NULL;
	}

214
	bio->bi_sector = sector * (sb->s_blocksize >> 9);
215 216 217 218 219
	bio->bi_bdev = sb->s_bdev;
	bio_add_page(bio, page, PAGE_SIZE, 0);

	bio->bi_end_io = end_bio_io_page;
	bio->bi_private = page;
220
	submit_bio(READ_SYNC | (1 << BIO_RW_META), bio);
221 222 223 224 225 226 227 228 229
	wait_on_page_locked(page);
	bio_put(bio);
	if (!PageUptodate(page)) {
		__free_page(page);
		return NULL;
	}
	return page;
}

D
David Teigland 已提交
230 231 232 233 234 235 236 237 238 239
/**
 * gfs2_read_sb - Read super block
 * @sdp: The GFS2 superblock
 * @gl: the glock for the superblock (assumed to be held)
 * @silent: Don't print message if mount fails
 *
 */

int gfs2_read_sb(struct gfs2_sbd *sdp, struct gfs2_glock *gl, int silent)
{
240 241
	u32 hash_blocks, ind_blocks, leaf_blocks;
	u32 tmp_blocks;
D
David Teigland 已提交
242 243
	unsigned int x;
	int error;
244 245
	struct page *page;
	char *sb;
D
David Teigland 已提交
246

247 248
	page = gfs2_read_super(sdp->sd_vfs, GFS2_SB_ADDR >> sdp->sd_fsb2bb_shift);
	if (!page) {
D
David Teigland 已提交
249 250
		if (!silent)
			fs_err(sdp, "can't read superblock\n");
251
		return -EIO;
D
David Teigland 已提交
252
	}
253 254 255 256
	sb = kmap(page);
	gfs2_sb_in(&sdp->sd_sb, sb);
	kunmap(page);
	__free_page(page);
D
David Teigland 已提交
257 258 259 260 261 262 263 264 265

	error = gfs2_check_sb(sdp, &sdp->sd_sb, silent);
	if (error)
		return error;

	sdp->sd_fsb2bb_shift = sdp->sd_sb.sb_bsize_shift -
			       GFS2_BASIC_BLOCK_SHIFT;
	sdp->sd_fsb2bb = 1 << sdp->sd_fsb2bb_shift;
	sdp->sd_diptrs = (sdp->sd_sb.sb_bsize -
266
			  sizeof(struct gfs2_dinode)) / sizeof(u64);
D
David Teigland 已提交
267
	sdp->sd_inptrs = (sdp->sd_sb.sb_bsize -
268
			  sizeof(struct gfs2_meta_header)) / sizeof(u64);
D
David Teigland 已提交
269 270 271
	sdp->sd_jbsize = sdp->sd_sb.sb_bsize - sizeof(struct gfs2_meta_header);
	sdp->sd_hash_bsize = sdp->sd_sb.sb_bsize / 2;
	sdp->sd_hash_bsize_shift = sdp->sd_sb.sb_bsize_shift - 1;
272
	sdp->sd_hash_ptrs = sdp->sd_hash_bsize / sizeof(u64);
D
David Teigland 已提交
273 274
	sdp->sd_qc_per_block = (sdp->sd_sb.sb_bsize -
				sizeof(struct gfs2_meta_header)) /
275
			        sizeof(struct gfs2_quota_change);
D
David Teigland 已提交
276 277 278

	/* Compute maximum reservation required to add a entry to a directory */

279
	hash_blocks = DIV_ROUND_UP(sizeof(u64) * (1 << GFS2_DIR_MAX_DEPTH),
D
David Teigland 已提交
280 281 282 283
			     sdp->sd_jbsize);

	ind_blocks = 0;
	for (tmp_blocks = hash_blocks; tmp_blocks > sdp->sd_diptrs;) {
284
		tmp_blocks = DIV_ROUND_UP(tmp_blocks, sdp->sd_inptrs);
D
David Teigland 已提交
285 286 287 288 289 290 291 292 293 294 295
		ind_blocks += tmp_blocks;
	}

	leaf_blocks = 2 + GFS2_DIR_MAX_DEPTH;

	sdp->sd_max_dirres = hash_blocks + ind_blocks + leaf_blocks;

	sdp->sd_heightsize[0] = sdp->sd_sb.sb_bsize -
				sizeof(struct gfs2_dinode);
	sdp->sd_heightsize[1] = sdp->sd_sb.sb_bsize * sdp->sd_diptrs;
	for (x = 2;; x++) {
296 297
		u64 space, d;
		u32 m;
D
David Teigland 已提交
298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313

		space = sdp->sd_heightsize[x - 1] * sdp->sd_inptrs;
		d = space;
		m = do_div(d, sdp->sd_inptrs);

		if (d != sdp->sd_heightsize[x - 1] || m)
			break;
		sdp->sd_heightsize[x] = space;
	}
	sdp->sd_max_height = x;
	gfs2_assert(sdp, sdp->sd_max_height <= GFS2_MAX_META_HEIGHT);

	sdp->sd_jheightsize[0] = sdp->sd_sb.sb_bsize -
				 sizeof(struct gfs2_dinode);
	sdp->sd_jheightsize[1] = sdp->sd_jbsize * sdp->sd_diptrs;
	for (x = 2;; x++) {
314 315
		u64 space, d;
		u32 m;
D
David Teigland 已提交
316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344

		space = sdp->sd_jheightsize[x - 1] * sdp->sd_inptrs;
		d = space;
		m = do_div(d, sdp->sd_inptrs);

		if (d != sdp->sd_jheightsize[x - 1] || m)
			break;
		sdp->sd_jheightsize[x] = space;
	}
	sdp->sd_max_jheight = x;
	gfs2_assert(sdp, sdp->sd_max_jheight <= GFS2_MAX_META_HEIGHT);

	return 0;
}

/**
 * gfs2_jindex_hold - Grab a lock on the jindex
 * @sdp: The GFS2 superblock
 * @ji_gh: the holder for the jindex glock
 *
 * This is very similar to the gfs2_rindex_hold() function, except that
 * in general we hold the jindex lock for longer periods of time and
 * we grab it far less frequently (in general) then the rgrp lock.
 *
 * Returns: errno
 */

int gfs2_jindex_hold(struct gfs2_sbd *sdp, struct gfs2_holder *ji_gh)
{
345
	struct gfs2_inode *dip = GFS2_I(sdp->sd_jindex);
D
David Teigland 已提交
346 347 348 349 350 351 352
	struct qstr name;
	char buf[20];
	struct gfs2_jdesc *jd;
	int error;

	name.name = buf;

353
	mutex_lock(&sdp->sd_jindex_mutex);
D
David Teigland 已提交
354 355

	for (;;) {
356
		error = gfs2_glock_nq_init(dip->i_gl, LM_ST_SHARED, 0, ji_gh);
D
David Teigland 已提交
357 358 359 360
		if (error)
			break;

		name.len = sprintf(buf, "journal%u", sdp->sd_journals);
361
		name.hash = gfs2_disk_hash(name.name, name.len);
D
David Teigland 已提交
362

363
		error = gfs2_dir_check(sdp->sd_jindex, &name, NULL);
D
David Teigland 已提交
364 365 366 367 368 369 370 371 372 373 374 375 376 377 378
		if (error == -ENOENT) {
			error = 0;
			break;
		}

		gfs2_glock_dq_uninit(ji_gh);

		if (error)
			break;

		error = -ENOMEM;
		jd = kzalloc(sizeof(struct gfs2_jdesc), GFP_KERNEL);
		if (!jd)
			break;

379 380 381 382 383 384
		jd->jd_inode = gfs2_lookupi(sdp->sd_jindex, &name, 1, NULL);
		if (!jd->jd_inode || IS_ERR(jd->jd_inode)) {
			if (!jd->jd_inode)
				error = -ENOENT;
			else
				error = PTR_ERR(jd->jd_inode);
D
David Teigland 已提交
385 386 387 388 389 390 391 392 393 394
			kfree(jd);
			break;
		}

		spin_lock(&sdp->sd_jindex_spin);
		jd->jd_jid = sdp->sd_journals++;
		list_add_tail(&jd->jd_list, &sdp->sd_jindex_list);
		spin_unlock(&sdp->sd_jindex_spin);
	}

395
	mutex_unlock(&sdp->sd_jindex_mutex);
D
David Teigland 已提交
396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419

	return error;
}

/**
 * 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;
	struct gfs2_jdesc *jd;

	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);
		list_del(&jd->jd_list);
420
		iput(jd->jd_inode);
D
David Teigland 已提交
421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488
		kfree(jd);
	}
}

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;
}

void gfs2_jdesc_make_dirty(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);
	if (jd)
		jd->jd_dirty = 1;
	spin_unlock(&sdp->sd_jindex_spin);
}

struct gfs2_jdesc *gfs2_jdesc_find_dirty(struct gfs2_sbd *sdp)
{
	struct gfs2_jdesc *jd;
	int found = 0;

	spin_lock(&sdp->sd_jindex_spin);

	list_for_each_entry(jd, &sdp->sd_jindex_list, jd_list) {
		if (jd->jd_dirty) {
			jd->jd_dirty = 0;
			found = 1;
			break;
		}
	}
	spin_unlock(&sdp->sd_jindex_spin);

	if (!found)
		jd = NULL;

	return jd;
}

int gfs2_jdesc_check(struct gfs2_jdesc *jd)
{
489 490
	struct gfs2_inode *ip = GFS2_I(jd->jd_inode);
	struct gfs2_sbd *sdp = GFS2_SB(jd->jd_inode);
D
David Teigland 已提交
491 492 493
	int ar;
	int error;

494
	if (ip->i_di.di_size < (8 << 20) || ip->i_di.di_size > (1 << 30) ||
D
David Teigland 已提交
495 496 497 498 499 500
	    (ip->i_di.di_size & (sdp->sd_sb.sb_bsize - 1))) {
		gfs2_consist_inode(ip);
		return -EIO;
	}
	jd->jd_blocks = ip->i_di.di_size >> sdp->sd_sb.sb_bsize_shift;

501
	error = gfs2_write_alloc_required(ip, 0, ip->i_di.di_size, &ar);
D
David Teigland 已提交
502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518
	if (!error && ar) {
		gfs2_consist_inode(ip);
		error = -EIO;
	}

	return error;
}

/**
 * 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)
{
519
	struct gfs2_inode *ip = GFS2_I(sdp->sd_jdesc->jd_inode);
520
	struct gfs2_glock *j_gl = ip->i_gl;
D
David Teigland 已提交
521
	struct gfs2_holder t_gh;
A
Al Viro 已提交
522
	struct gfs2_log_header_host head;
D
David Teigland 已提交
523 524
	int error;

525
	error = gfs2_glock_nq_init(sdp->sd_trans_gl, LM_ST_SHARED, 0, &t_gh);
D
David Teigland 已提交
526 527 528
	if (error)
		return error;

529
	gfs2_meta_cache_flush(ip);
530
	j_gl->gl_ops->go_inval(j_gl, DIO_METADATA);
D
David Teigland 已提交
531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547

	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)
548
		goto fail;
D
David Teigland 已提交
549 550 551 552 553 554 555

	set_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags);

	gfs2_glock_dq_uninit(&t_gh);

	return 0;

556
fail:
D
David Teigland 已提交
557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577
	t_gh.gh_flags |= GL_NOCACHE;
	gfs2_glock_dq_uninit(&t_gh);

	return error;
}

/**
 * gfs2_make_fs_ro - Turn a Read-Write FS into a Read-Only one
 * @sdp: the filesystem
 *
 * Returns: errno
 */

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

	gfs2_quota_sync(sdp);
	gfs2_statfs_sync(sdp);

578 579
	error = gfs2_glock_nq_init(sdp->sd_trans_gl, LM_ST_SHARED, GL_NOCACHE,
				   &t_gh);
D
David Teigland 已提交
580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597
	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;
}

int gfs2_statfs_init(struct gfs2_sbd *sdp)
{
598
	struct gfs2_inode *m_ip = GFS2_I(sdp->sd_statfs_inode);
599
	struct gfs2_statfs_change_host *m_sc = &sdp->sd_statfs_master;
600
	struct gfs2_inode *l_ip = GFS2_I(sdp->sd_sc_inode);
601
	struct gfs2_statfs_change_host *l_sc = &sdp->sd_statfs_local;
D
David Teigland 已提交
602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634
	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);
	}

635
out_m_bh:
D
David Teigland 已提交
636
	brelse(m_bh);
637
out:
D
David Teigland 已提交
638 639 640 641
	gfs2_glock_dq_uninit(&gh);
	return 0;
}

642 643
void gfs2_statfs_change(struct gfs2_sbd *sdp, s64 total, s64 free,
			s64 dinodes)
D
David Teigland 已提交
644
{
645
	struct gfs2_inode *l_ip = GFS2_I(sdp->sd_sc_inode);
646
	struct gfs2_statfs_change_host *l_sc = &sdp->sd_statfs_local;
D
David Teigland 已提交
647 648 649 650 651 652 653
	struct buffer_head *l_bh;
	int error;

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

654
	mutex_lock(&sdp->sd_statfs_mutex);
655
	gfs2_trans_add_bh(l_ip->i_gl, l_bh, 1);
656
	mutex_unlock(&sdp->sd_statfs_mutex);
D
David Teigland 已提交
657 658 659 660 661

	spin_lock(&sdp->sd_statfs_spin);
	l_sc->sc_total += total;
	l_sc->sc_free += free;
	l_sc->sc_dinodes += dinodes;
662
	gfs2_statfs_change_out(l_sc, l_bh->b_data + sizeof(struct gfs2_dinode));
D
David Teigland 已提交
663 664 665 666 667 668 669
	spin_unlock(&sdp->sd_statfs_spin);

	brelse(l_bh);
}

int gfs2_statfs_sync(struct gfs2_sbd *sdp)
{
670 671
	struct gfs2_inode *m_ip = GFS2_I(sdp->sd_statfs_inode);
	struct gfs2_inode *l_ip = GFS2_I(sdp->sd_sc_inode);
672 673
	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
674 675 676 677 678 679 680 681 682 683 684 685 686 687 688
	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 +
689
			      sizeof(struct gfs2_dinode));
D
David Teigland 已提交
690 691 692 693 694 695 696 697 698 699 700 701 702 703
	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;

704
	mutex_lock(&sdp->sd_statfs_mutex);
705
	gfs2_trans_add_bh(l_ip->i_gl, l_bh, 1);
706
	mutex_unlock(&sdp->sd_statfs_mutex);
D
David Teigland 已提交
707 708 709 710 711 712 713 714 715 716

	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);

717
	gfs2_trans_add_bh(m_ip->i_gl, m_bh, 1);
D
David Teigland 已提交
718 719 720 721
	gfs2_statfs_change_out(m_sc, m_bh->b_data + sizeof(struct gfs2_dinode));

	gfs2_trans_end(sdp);

722
out_bh2:
D
David Teigland 已提交
723
	brelse(l_bh);
724
out_bh:
D
David Teigland 已提交
725
	brelse(m_bh);
726
out:
D
David Teigland 已提交
727 728 729 730 731 732 733 734 735 736 737 738
	gfs2_glock_dq_uninit(&gh);
	return error;
}

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

739
int gfs2_statfs_i(struct gfs2_sbd *sdp, struct gfs2_statfs_change_host *sc)
D
David Teigland 已提交
740
{
741 742
	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
743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771

	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;
}

/**
 * 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,
772
			    struct gfs2_statfs_change_host *sc)
D
David Teigland 已提交
773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793
{
	gfs2_rgrp_verify(rgd);
	sc->sc_total += rgd->rd_ri.ri_data;
	sc->sc_free += rgd->rd_rg.rg_free;
	sc->sc_dinodes += rgd->rd_rg.rg_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
 */

794
int gfs2_statfs_slow(struct gfs2_sbd *sdp, struct gfs2_statfs_change_host *sc)
D
David Teigland 已提交
795 796 797 798 799 800 801 802 803
{
	struct gfs2_holder ri_gh;
	struct gfs2_rgrpd *rgd_next;
	struct gfs2_holder *gha, *gh;
	unsigned int slots = 64;
	unsigned int x;
	int done;
	int error = 0, err;

804
	memset(sc, 0, sizeof(struct gfs2_statfs_change_host));
D
David Teigland 已提交
805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827
	gha = kcalloc(slots, sizeof(struct gfs2_holder), GFP_KERNEL);
	if (!gha)
		return -ENOMEM;

	error = gfs2_rindex_hold(sdp, &ri_gh);
	if (error)
		goto out;

	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)
828 829
						error = statfs_slow_fill(
							gh->gh_gl->gl_object, sc);
D
David Teigland 已提交
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
					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();
	}

	gfs2_glock_dq_uninit(&ri_gh);

857
out:
D
David Teigland 已提交
858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876
	kfree(gha);
	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
 */

877 878
static int gfs2_lock_fs_check_clean(struct gfs2_sbd *sdp,
				    struct gfs2_holder *t_gh)
D
David Teigland 已提交
879
{
880
	struct gfs2_inode *ip;
D
David Teigland 已提交
881 882 883 884
	struct gfs2_holder ji_gh;
	struct gfs2_jdesc *jd;
	struct lfcc *lfcc;
	LIST_HEAD(list);
A
Al Viro 已提交
885
	struct gfs2_log_header_host lh;
D
David Teigland 已提交
886 887 888 889 890 891 892 893 894 895 896 897
	int error;

	error = gfs2_jindex_hold(sdp, &ji_gh);
	if (error)
		return 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;
		}
898 899
		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
900 901 902 903 904 905 906 907
		if (error) {
			kfree(lfcc);
			goto out;
		}
		list_add(&lfcc->list, &list);
	}

	error = gfs2_glock_nq_init(sdp->sd_trans_gl, LM_ST_DEFERRED,
908
			       LM_FLAG_PRIORITY | GL_NOCACHE,
D
David Teigland 已提交
909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926
			       t_gh);

	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);

927
out:
D
David Teigland 已提交
928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952
	while (!list_empty(&list)) {
		lfcc = list_entry(list.next, struct lfcc, list);
		list_del(&lfcc->list);
		gfs2_glock_dq_uninit(&lfcc->gh);
		kfree(lfcc);
	}
	gfs2_glock_dq_uninit(&ji_gh);
	return error;
}

/**
 * gfs2_freeze_fs - freezes the file system
 * @sdp: the file system
 *
 * This function flushes data and meta data for all machines by
 * aquiring the transaction log exclusively.  All journals are
 * ensured to be in a clean state as well.
 *
 * Returns: errno
 */

int gfs2_freeze_fs(struct gfs2_sbd *sdp)
{
	int error = 0;

953
	mutex_lock(&sdp->sd_freeze_lock);
D
David Teigland 已提交
954 955 956 957 958 959 960

	if (!sdp->sd_freeze_count++) {
		error = gfs2_lock_fs_check_clean(sdp, &sdp->sd_freeze_gh);
		if (error)
			sdp->sd_freeze_count--;
	}

961
	mutex_unlock(&sdp->sd_freeze_lock);
D
David Teigland 已提交
962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977

	return error;
}

/**
 * gfs2_unfreeze_fs - unfreezes the file system
 * @sdp: the file system
 *
 * This function allows the file system to proceed by unlocking
 * the exclusively held transaction lock.  Other GFS2 nodes are
 * now free to acquire the lock shared and go on with their lives.
 *
 */

void gfs2_unfreeze_fs(struct gfs2_sbd *sdp)
{
978
	mutex_lock(&sdp->sd_freeze_lock);
D
David Teigland 已提交
979 980 981 982

	if (sdp->sd_freeze_count && !--sdp->sd_freeze_count)
		gfs2_glock_dq_uninit(&sdp->sd_freeze_gh);

983
	mutex_unlock(&sdp->sd_freeze_lock);
D
David Teigland 已提交
984 985
}