super.c 20.5 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6
/*
 *   Copyright (C) International Business Machines Corp., 2000-2004
 *   Portions Copyright (C) Christoph Hellwig, 2001-2002
 *
 *   This program is free software;  you can redistribute it and/or modify
 *   it under the terms of the GNU General Public License as published by
D
Dave Kleikamp 已提交
7
 *   the Free Software Foundation; either version 2 of the License, or
L
Linus Torvalds 已提交
8
 *   (at your option) any later version.
D
Dave Kleikamp 已提交
9
 *
L
Linus Torvalds 已提交
10 11 12 13 14 15
 *   This program is distributed in the hope that it will be useful,
 *   but WITHOUT ANY WARRANTY;  without even the implied warranty of
 *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See
 *   the GNU General Public License for more details.
 *
 *   You should have received a copy of the GNU General Public License
D
Dave Kleikamp 已提交
16
 *   along with this program;  if not, write to the Free Software
L
Linus Torvalds 已提交
17 18 19 20 21 22 23 24
 *   Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
 */

#include <linux/fs.h>
#include <linux/module.h>
#include <linux/parser.h>
#include <linux/completion.h>
#include <linux/vfs.h>
25
#include <linux/mount.h>
L
Linus Torvalds 已提交
26
#include <linux/moduleparam.h>
C
Christoph Hellwig 已提交
27
#include <linux/kthread.h>
28
#include <linux/posix_acl.h>
29
#include <linux/buffer_head.h>
L
Linus Torvalds 已提交
30
#include <asm/uaccess.h>
31
#include <linux/seq_file.h>
L
Linus Torvalds 已提交
32 33 34

#include "jfs_incore.h"
#include "jfs_filsys.h"
35
#include "jfs_inode.h"
L
Linus Torvalds 已提交
36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57
#include "jfs_metapage.h"
#include "jfs_superblock.h"
#include "jfs_dmap.h"
#include "jfs_imap.h"
#include "jfs_acl.h"
#include "jfs_debug.h"

MODULE_DESCRIPTION("The Journaled Filesystem (JFS)");
MODULE_AUTHOR("Steve Best/Dave Kleikamp/Barry Arndt, IBM");
MODULE_LICENSE("GPL");

static kmem_cache_t * jfs_inode_cachep;

static struct super_operations jfs_super_operations;
static struct export_operations jfs_export_operations;
static struct file_system_type jfs_fs_type;

#define MAX_COMMIT_THREADS 64
static int commit_threads = 0;
module_param(commit_threads, int, 0);
MODULE_PARM_DESC(commit_threads, "Number of commit threads");

C
Christoph Hellwig 已提交
58 59 60
static struct task_struct *jfsCommitThread[MAX_COMMIT_THREADS];
struct task_struct *jfsIOthread;
struct task_struct *jfsSyncThread;
L
Linus Torvalds 已提交
61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84

#ifdef CONFIG_JFS_DEBUG
int jfsloglevel = JFS_LOGLEVEL_WARN;
module_param(jfsloglevel, int, 0644);
MODULE_PARM_DESC(jfsloglevel, "Specify JFS loglevel (0, 1 or 2)");
#endif

static void jfs_handle_error(struct super_block *sb)
{
	struct jfs_sb_info *sbi = JFS_SBI(sb);

	if (sb->s_flags & MS_RDONLY)
		return;

	updateSuper(sb, FM_DIRTY);

	if (sbi->flag & JFS_ERR_PANIC)
		panic("JFS (device %s): panic forced after error\n",
			sb->s_id);
	else if (sbi->flag & JFS_ERR_REMOUNT_RO) {
		jfs_err("ERROR: (device %s): remounting filesystem "
			"as read-only\n",
			sb->s_id);
		sb->s_flags |= MS_RDONLY;
D
Dave Kleikamp 已提交
85
	}
L
Linus Torvalds 已提交
86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117

	/* nothing is done for continue beyond marking the superblock dirty */
}

void jfs_error(struct super_block *sb, const char * function, ...)
{
	static char error_buf[256];
	va_list args;

	va_start(args, function);
	vsprintf(error_buf, function, args);
	va_end(args);

	printk(KERN_ERR "ERROR: (device %s): %s\n", sb->s_id, error_buf);

	jfs_handle_error(sb);
}

static struct inode *jfs_alloc_inode(struct super_block *sb)
{
	struct jfs_inode_info *jfs_inode;

	jfs_inode = kmem_cache_alloc(jfs_inode_cachep, GFP_NOFS);
	if (!jfs_inode)
		return NULL;
	return &jfs_inode->vfs_inode;
}

static void jfs_destroy_inode(struct inode *inode)
{
	struct jfs_inode_info *ji = JFS_IP(inode);

D
Dave Kleikamp 已提交
118 119
	BUG_ON(!list_empty(&ji->anon_inode_list));

L
Linus Torvalds 已提交
120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141
	spin_lock_irq(&ji->ag_lock);
	if (ji->active_ag != -1) {
		struct bmap *bmap = JFS_SBI(inode->i_sb)->bmap;
		atomic_dec(&bmap->db_active[ji->active_ag]);
		ji->active_ag = -1;
	}
	spin_unlock_irq(&ji->ag_lock);

#ifdef CONFIG_JFS_POSIX_ACL
	if (ji->i_acl != JFS_ACL_NOT_CACHED) {
		posix_acl_release(ji->i_acl);
		ji->i_acl = JFS_ACL_NOT_CACHED;
	}
	if (ji->i_default_acl != JFS_ACL_NOT_CACHED) {
		posix_acl_release(ji->i_default_acl);
		ji->i_default_acl = JFS_ACL_NOT_CACHED;
	}
#endif

	kmem_cache_free(jfs_inode_cachep, ji);
}

142
static int jfs_statfs(struct dentry *dentry, struct kstatfs *buf)
L
Linus Torvalds 已提交
143
{
144
	struct jfs_sb_info *sbi = JFS_SBI(dentry->d_sb);
L
Linus Torvalds 已提交
145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186
	s64 maxinodes;
	struct inomap *imap = JFS_IP(sbi->ipimap)->i_imap;

	jfs_info("In jfs_statfs");
	buf->f_type = JFS_SUPER_MAGIC;
	buf->f_bsize = sbi->bsize;
	buf->f_blocks = sbi->bmap->db_mapsize;
	buf->f_bfree = sbi->bmap->db_nfree;
	buf->f_bavail = sbi->bmap->db_nfree;
	/*
	 * If we really return the number of allocated & free inodes, some
	 * applications will fail because they won't see enough free inodes.
	 * We'll try to calculate some guess as to how may inodes we can
	 * really allocate
	 *
	 * buf->f_files = atomic_read(&imap->im_numinos);
	 * buf->f_ffree = atomic_read(&imap->im_numfree);
	 */
	maxinodes = min((s64) atomic_read(&imap->im_numinos) +
			((sbi->bmap->db_nfree >> imap->im_l2nbperiext)
			 << L2INOSPEREXT), (s64) 0xffffffffLL);
	buf->f_files = maxinodes;
	buf->f_ffree = maxinodes - (atomic_read(&imap->im_numinos) -
				    atomic_read(&imap->im_numfree));

	buf->f_namelen = JFS_NAME_MAX;
	return 0;
}

static void jfs_put_super(struct super_block *sb)
{
	struct jfs_sb_info *sbi = JFS_SBI(sb);
	int rc;

	jfs_info("In jfs_put_super");
	rc = jfs_umount(sb);
	if (rc)
		jfs_err("jfs_umount failed with return code %d", rc);
	if (sbi->nls_tab)
		unload_nls(sbi->nls_tab);
	sbi->nls_tab = NULL;

187 188 189 190
	truncate_inode_pages(sbi->direct_inode->i_mapping, 0);
	iput(sbi->direct_inode);
	sbi->direct_inode = NULL;

L
Linus Torvalds 已提交
191 192 193 194 195
	kfree(sbi);
}

enum {
	Opt_integrity, Opt_nointegrity, Opt_iocharset, Opt_resize,
196
	Opt_resize_nosize, Opt_errors, Opt_ignore, Opt_err, Opt_quota,
197
	Opt_usrquota, Opt_grpquota, Opt_uid, Opt_gid, Opt_umask
L
Linus Torvalds 已提交
198 199 200 201 202 203 204 205 206 207 208
};

static match_table_t tokens = {
	{Opt_integrity, "integrity"},
	{Opt_nointegrity, "nointegrity"},
	{Opt_iocharset, "iocharset=%s"},
	{Opt_resize, "resize=%u"},
	{Opt_resize_nosize, "resize"},
	{Opt_errors, "errors=%s"},
	{Opt_ignore, "noquota"},
	{Opt_ignore, "quota"},
209 210
	{Opt_usrquota, "usrquota"},
	{Opt_grpquota, "grpquota"},
211 212 213
	{Opt_uid, "uid=%u"},
	{Opt_gid, "gid=%u"},
	{Opt_umask, "umask=%u"},
L
Linus Torvalds 已提交
214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300
	{Opt_err, NULL}
};

static int parse_options(char *options, struct super_block *sb, s64 *newLVSize,
			 int *flag)
{
	void *nls_map = (void *)-1;	/* -1: no change;  NULL: none */
	char *p;
	struct jfs_sb_info *sbi = JFS_SBI(sb);

	*newLVSize = 0;

	if (!options)
		return 1;

	while ((p = strsep(&options, ",")) != NULL) {
		substring_t args[MAX_OPT_ARGS];
		int token;
		if (!*p)
			continue;

		token = match_token(p, tokens, args);
		switch (token) {
		case Opt_integrity:
			*flag &= ~JFS_NOINTEGRITY;
			break;
		case Opt_nointegrity:
			*flag |= JFS_NOINTEGRITY;
			break;
		case Opt_ignore:
			/* Silently ignore the quota options */
			/* Don't do anything ;-) */
			break;
		case Opt_iocharset:
			if (nls_map && nls_map != (void *) -1)
				unload_nls(nls_map);
			if (!strcmp(args[0].from, "none"))
				nls_map = NULL;
			else {
				nls_map = load_nls(args[0].from);
				if (!nls_map) {
					printk(KERN_ERR
					       "JFS: charset not found\n");
					goto cleanup;
				}
			}
			break;
		case Opt_resize:
		{
			char *resize = args[0].from;
			*newLVSize = simple_strtoull(resize, &resize, 0);
			break;
		}
		case Opt_resize_nosize:
		{
			*newLVSize = sb->s_bdev->bd_inode->i_size >>
				sb->s_blocksize_bits;
			if (*newLVSize == 0)
				printk(KERN_ERR
				       "JFS: Cannot determine volume size\n");
			break;
		}
		case Opt_errors:
		{
			char *errors = args[0].from;
			if (!errors || !*errors)
				goto cleanup;
			if (!strcmp(errors, "continue")) {
				*flag &= ~JFS_ERR_REMOUNT_RO;
				*flag &= ~JFS_ERR_PANIC;
				*flag |= JFS_ERR_CONTINUE;
			} else if (!strcmp(errors, "remount-ro")) {
				*flag &= ~JFS_ERR_CONTINUE;
				*flag &= ~JFS_ERR_PANIC;
				*flag |= JFS_ERR_REMOUNT_RO;
			} else if (!strcmp(errors, "panic")) {
				*flag &= ~JFS_ERR_CONTINUE;
				*flag &= ~JFS_ERR_REMOUNT_RO;
				*flag |= JFS_ERR_PANIC;
			} else {
				printk(KERN_ERR
				       "JFS: %s is an invalid error handler\n",
				       errors);
				goto cleanup;
			}
			break;
		}
301

302
#ifdef CONFIG_QUOTA
303 304 305 306 307 308 309 310 311 312 313 314 315 316 317
		case Opt_quota:
		case Opt_usrquota:
			*flag |= JFS_USRQUOTA;
			break;
		case Opt_grpquota:
			*flag |= JFS_GRPQUOTA;
			break;
#else
		case Opt_usrquota:
		case Opt_grpquota:
		case Opt_quota:
			printk(KERN_ERR
			       "JFS: quota operations not supported\n");
			break;
#endif
318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340
		case Opt_uid:
		{
			char *uid = args[0].from;
			sbi->uid = simple_strtoul(uid, &uid, 0);
			break;
		}
		case Opt_gid:
		{
			char *gid = args[0].from;
			sbi->gid = simple_strtoul(gid, &gid, 0);
			break;
		}
		case Opt_umask:
		{
			char *umask = args[0].from;
			sbi->umask = simple_strtoul(umask, &umask, 8);
			if (sbi->umask & ~0777) {
				printk(KERN_ERR
				       "JFS: Invalid value of umask\n");
				goto cleanup;
			}
			break;
		}
L
Linus Torvalds 已提交
341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382
		default:
			printk("jfs: Unrecognized mount option \"%s\" "
					" or missing value\n", p);
			goto cleanup;
		}
	}

	if (nls_map != (void *) -1) {
		/* Discard old (if remount) */
		if (sbi->nls_tab)
			unload_nls(sbi->nls_tab);
		sbi->nls_tab = nls_map;
	}
	return 1;

cleanup:
	if (nls_map && nls_map != (void *) -1)
		unload_nls(nls_map);
	return 0;
}

static int jfs_remount(struct super_block *sb, int *flags, char *data)
{
	s64 newLVSize = 0;
	int rc = 0;
	int flag = JFS_SBI(sb)->flag;

	if (!parse_options(data, sb, &newLVSize, &flag)) {
		return -EINVAL;
	}
	if (newLVSize) {
		if (sb->s_flags & MS_RDONLY) {
			printk(KERN_ERR
		  "JFS: resize requires volume to be mounted read-write\n");
			return -EROFS;
		}
		rc = jfs_extendfs(sb, newLVSize, 0);
		if (rc)
			return rc;
	}

	if ((sb->s_flags & MS_RDONLY) && !(*flags & MS_RDONLY)) {
383 384 385 386 387 388
		/*
		 * Invalidate any previously read metadata.  fsck may have
		 * changed the on-disk data since we mounted r/o
		 */
		truncate_inode_pages(JFS_SBI(sb)->direct_inode->i_mapping, 0);

L
Linus Torvalds 已提交
389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422
		JFS_SBI(sb)->flag = flag;
		return jfs_mount_rw(sb, 1);
	}
	if ((!(sb->s_flags & MS_RDONLY)) && (*flags & MS_RDONLY)) {
		rc = jfs_umount_rw(sb);
		JFS_SBI(sb)->flag = flag;
		return rc;
	}
	if ((JFS_SBI(sb)->flag & JFS_NOINTEGRITY) != (flag & JFS_NOINTEGRITY))
		if (!(sb->s_flags & MS_RDONLY)) {
			rc = jfs_umount_rw(sb);
			if (rc)
				return rc;
			JFS_SBI(sb)->flag = flag;
			return jfs_mount_rw(sb, 1);
		}
	JFS_SBI(sb)->flag = flag;

	return 0;
}

static int jfs_fill_super(struct super_block *sb, void *data, int silent)
{
	struct jfs_sb_info *sbi;
	struct inode *inode;
	int rc;
	s64 newLVSize = 0;
	int flag;

	jfs_info("In jfs_read_super: s_flags=0x%lx", sb->s_flags);

	if (!new_valid_dev(sb->s_bdev->bd_dev))
		return -EOVERFLOW;

E
Eric Sesterhenn 已提交
423
	sbi = kzalloc(sizeof (struct jfs_sb_info), GFP_KERNEL);
L
Linus Torvalds 已提交
424
	if (!sbi)
425
		return -ENOMEM;
L
Linus Torvalds 已提交
426 427
	sb->s_fs_info = sbi;
	sbi->sb = sb;
428
	sbi->uid = sbi->gid = sbi->umask = -1;
L
Linus Torvalds 已提交
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

	/* initialize the mount flag and determine the default error handler */
	flag = JFS_ERR_REMOUNT_RO;

	if (!parse_options((char *) data, sb, &newLVSize, &flag)) {
		kfree(sbi);
		return -EINVAL;
	}
	sbi->flag = flag;

#ifdef CONFIG_JFS_POSIX_ACL
	sb->s_flags |= MS_POSIXACL;
#endif

	if (newLVSize) {
		printk(KERN_ERR "resize option for remount only\n");
		return -EINVAL;
	}

	/*
	 * Initialize blocksize to 4K.
	 */
	sb_set_blocksize(sb, PSIZE);

	/*
	 * Set method vectors.
	 */
	sb->s_op = &jfs_super_operations;
	sb->s_export_op = &jfs_export_operations;

459 460 461 462 463 464 465 466 467 468
	/*
	 * Initialize direct-mapping inode/address-space
	 */
	inode = new_inode(sb);
	if (inode == NULL)
		goto out_kfree;
	inode->i_ino = 0;
	inode->i_nlink = 1;
	inode->i_size = sb->s_bdev->bd_inode->i_size;
	inode->i_mapping->a_ops = &jfs_metapage_aops;
469
	insert_inode_hash(inode);
470 471 472 473
	mapping_set_gfp_mask(inode->i_mapping, GFP_NOFS);

	sbi->direct_inode = inode;

L
Linus Torvalds 已提交
474 475 476 477 478
	rc = jfs_mount(sb);
	if (rc) {
		if (!silent) {
			jfs_err("jfs_mount failed w/return code = %d", rc);
		}
479
		goto out_mount_failed;
L
Linus Torvalds 已提交
480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527
	}
	if (sb->s_flags & MS_RDONLY)
		sbi->log = NULL;
	else {
		rc = jfs_mount_rw(sb, 0);
		if (rc) {
			if (!silent) {
				jfs_err("jfs_mount_rw failed, return code = %d",
					rc);
			}
			goto out_no_rw;
		}
	}

	sb->s_magic = JFS_SUPER_MAGIC;

	inode = iget(sb, ROOT_I);
	if (!inode || is_bad_inode(inode))
		goto out_no_root;
	sb->s_root = d_alloc_root(inode);
	if (!sb->s_root)
		goto out_no_root;

	if (sbi->mntflag & JFS_OS2)
		sb->s_root->d_op = &jfs_ci_dentry_operations;

	/* logical blocks are represented by 40 bits in pxd_t, etc. */
	sb->s_maxbytes = ((u64) sb->s_blocksize) << 40;
#if BITS_PER_LONG == 32
	/*
	 * Page cache is indexed by long.
	 * I would use MAX_LFS_FILESIZE, but it's only half as big
	 */
	sb->s_maxbytes = min(((u64) PAGE_CACHE_SIZE << 32) - 1, sb->s_maxbytes);
#endif
	sb->s_time_gran = 1;
	return 0;

out_no_root:
	jfs_err("jfs_read_super: get root inode failed");
	if (inode)
		iput(inode);

out_no_rw:
	rc = jfs_umount(sb);
	if (rc) {
		jfs_err("jfs_umount failed with return code %d", rc);
	}
528
out_mount_failed:
529
	filemap_write_and_wait(sbi->direct_inode->i_mapping);
530 531 532 533
	truncate_inode_pages(sbi->direct_inode->i_mapping, 0);
	make_bad_inode(sbi->direct_inode);
	iput(sbi->direct_inode);
	sbi->direct_inode = NULL;
L
Linus Torvalds 已提交
534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567
out_kfree:
	if (sbi->nls_tab)
		unload_nls(sbi->nls_tab);
	kfree(sbi);
	return -EINVAL;
}

static void jfs_write_super_lockfs(struct super_block *sb)
{
	struct jfs_sb_info *sbi = JFS_SBI(sb);
	struct jfs_log *log = sbi->log;

	if (!(sb->s_flags & MS_RDONLY)) {
		txQuiesce(sb);
		lmLogShutdown(log);
		updateSuper(sb, FM_CLEAN);
	}
}

static void jfs_unlockfs(struct super_block *sb)
{
	struct jfs_sb_info *sbi = JFS_SBI(sb);
	struct jfs_log *log = sbi->log;
	int rc = 0;

	if (!(sb->s_flags & MS_RDONLY)) {
		updateSuper(sb, FM_MOUNT);
		if ((rc = lmLogInit(log)))
			jfs_err("jfs_unlock failed with return code %d", rc);
		else
			txResume(sb);
	}
}

568 569
static int jfs_get_sb(struct file_system_type *fs_type,
	int flags, const char *dev_name, void *data, struct vfsmount *mnt)
L
Linus Torvalds 已提交
570
{
571 572
	return get_sb_bdev(fs_type, flags, dev_name, data, jfs_fill_super,
			   mnt);
L
Linus Torvalds 已提交
573 574 575 576 577 578 579
}

static int jfs_sync_fs(struct super_block *sb, int wait)
{
	struct jfs_log *log = JFS_SBI(sb)->log;

	/* log == NULL indicates read-only mount */
580
	if (log) {
L
Linus Torvalds 已提交
581
		jfs_flush_journal(log, wait);
582
		jfs_syncpt(log, 0);
583
	}
L
Linus Torvalds 已提交
584 585 586 587

	return 0;
}

588 589 590 591
static int jfs_show_options(struct seq_file *seq, struct vfsmount *vfs)
{
	struct jfs_sb_info *sbi = JFS_SBI(vfs->mnt_sb);

592 593 594 595 596 597
	if (sbi->uid != -1)
		seq_printf(seq, ",uid=%d", sbi->uid);
	if (sbi->gid != -1)
		seq_printf(seq, ",gid=%d", sbi->gid);
	if (sbi->umask != -1)
		seq_printf(seq, ",umask=%03o", sbi->umask);
598 599 600
	if (sbi->flag & JFS_NOINTEGRITY)
		seq_puts(seq, ",nointegrity");

601
#ifdef CONFIG_QUOTA
602 603 604 605 606 607 608 609 610 611
	if (sbi->flag & JFS_USRQUOTA)
		seq_puts(seq, ",usrquota");

	if (sbi->flag & JFS_GRPQUOTA)
		seq_puts(seq, ",grpquota");
#endif

	return 0;
}

612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680
#ifdef CONFIG_QUOTA

/* Read data from quotafile - avoid pagecache and such because we cannot afford
 * acquiring the locks... As quota files are never truncated and quota code
 * itself serializes the operations (and noone else should touch the files)
 * we don't have to be afraid of races */
static ssize_t jfs_quota_read(struct super_block *sb, int type, char *data,
			      size_t len, loff_t off)
{
	struct inode *inode = sb_dqopt(sb)->files[type];
	sector_t blk = off >> sb->s_blocksize_bits;
	int err = 0;
	int offset = off & (sb->s_blocksize - 1);
	int tocopy;
	size_t toread;
	struct buffer_head tmp_bh;
	struct buffer_head *bh;
	loff_t i_size = i_size_read(inode);

	if (off > i_size)
		return 0;
	if (off+len > i_size)
		len = i_size-off;
	toread = len;
	while (toread > 0) {
		tocopy = sb->s_blocksize - offset < toread ?
				sb->s_blocksize - offset : toread;

		tmp_bh.b_state = 0;
		tmp_bh.b_size = 1 << inode->i_blkbits;
		err = jfs_get_block(inode, blk, &tmp_bh, 0);
		if (err)
			return err;
		if (!buffer_mapped(&tmp_bh))	/* A hole? */
			memset(data, 0, tocopy);
		else {
			bh = sb_bread(sb, tmp_bh.b_blocknr);
			if (!bh)
				return -EIO;
			memcpy(data, bh->b_data+offset, tocopy);
			brelse(bh);
		}
		offset = 0;
		toread -= tocopy;
		data += tocopy;
		blk++;
	}
	return len;
}

/* Write to quotafile */
static ssize_t jfs_quota_write(struct super_block *sb, int type,
			       const char *data, size_t len, loff_t off)
{
	struct inode *inode = sb_dqopt(sb)->files[type];
	sector_t blk = off >> sb->s_blocksize_bits;
	int err = 0;
	int offset = off & (sb->s_blocksize - 1);
	int tocopy;
	size_t towrite = len;
	struct buffer_head tmp_bh;
	struct buffer_head *bh;

	mutex_lock(&inode->i_mutex);
	while (towrite > 0) {
		tocopy = sb->s_blocksize - offset < towrite ?
				sb->s_blocksize - offset : towrite;

		tmp_bh.b_state = 0;
681
		tmp_bh.b_size = 1 << inode->i_blkbits;
682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718
		err = jfs_get_block(inode, blk, &tmp_bh, 1);
		if (err)
			goto out;
		if (offset || tocopy != sb->s_blocksize)
			bh = sb_bread(sb, tmp_bh.b_blocknr);
		else
			bh = sb_getblk(sb, tmp_bh.b_blocknr);
		if (!bh) {
			err = -EIO;
			goto out;
		}
		lock_buffer(bh);
		memcpy(bh->b_data+offset, data, tocopy);
		flush_dcache_page(bh->b_page);
		set_buffer_uptodate(bh);
		mark_buffer_dirty(bh);
		unlock_buffer(bh);
		brelse(bh);
		offset = 0;
		towrite -= tocopy;
		data += tocopy;
		blk++;
	}
out:
	if (len == towrite)
		return err;
	if (inode->i_size < off+len-towrite)
		i_size_write(inode, off+len-towrite);
	inode->i_version++;
	inode->i_mtime = inode->i_ctime = CURRENT_TIME;
	mark_inode_dirty(inode);
	mutex_unlock(&inode->i_mutex);
	return len - towrite;
}

#endif

L
Linus Torvalds 已提交
719 720 721 722 723 724 725 726 727 728 729 730 731
static struct super_operations jfs_super_operations = {
	.alloc_inode	= jfs_alloc_inode,
	.destroy_inode	= jfs_destroy_inode,
	.read_inode	= jfs_read_inode,
	.dirty_inode	= jfs_dirty_inode,
	.write_inode	= jfs_write_inode,
	.delete_inode	= jfs_delete_inode,
	.put_super	= jfs_put_super,
	.sync_fs	= jfs_sync_fs,
	.write_super_lockfs = jfs_write_super_lockfs,
	.unlockfs       = jfs_unlockfs,
	.statfs		= jfs_statfs,
	.remount_fs	= jfs_remount,
732 733 734 735 736
	.show_options	= jfs_show_options,
#ifdef CONFIG_QUOTA
	.quota_read	= jfs_quota_read,
	.quota_write	= jfs_quota_write,
#endif
L
Linus Torvalds 已提交
737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759
};

static struct export_operations jfs_export_operations = {
	.get_parent	= jfs_get_parent,
};

static struct file_system_type jfs_fs_type = {
	.owner		= THIS_MODULE,
	.name		= "jfs",
	.get_sb		= jfs_get_sb,
	.kill_sb	= kill_block_super,
	.fs_flags	= FS_REQUIRES_DEV,
};

static void init_once(void *foo, kmem_cache_t * cachep, unsigned long flags)
{
	struct jfs_inode_info *jfs_ip = (struct jfs_inode_info *) foo;

	if ((flags & (SLAB_CTOR_VERIFY | SLAB_CTOR_CONSTRUCTOR)) ==
	    SLAB_CTOR_CONSTRUCTOR) {
		memset(jfs_ip, 0, sizeof(struct jfs_inode_info));
		INIT_LIST_HEAD(&jfs_ip->anon_inode_list);
		init_rwsem(&jfs_ip->rdwrlock);
760
		mutex_init(&jfs_ip->commit_mutex);
L
Linus Torvalds 已提交
761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777
		init_rwsem(&jfs_ip->xattr_sem);
		spin_lock_init(&jfs_ip->ag_lock);
		jfs_ip->active_ag = -1;
#ifdef CONFIG_JFS_POSIX_ACL
		jfs_ip->i_acl = JFS_ACL_NOT_CACHED;
		jfs_ip->i_default_acl = JFS_ACL_NOT_CACHED;
#endif
		inode_init_once(&jfs_ip->vfs_inode);
	}
}

static int __init init_jfs_fs(void)
{
	int i;
	int rc;

	jfs_inode_cachep =
D
Dave Kleikamp 已提交
778
	    kmem_cache_create("jfs_ip", sizeof(struct jfs_inode_info), 0,
779 780
			    SLAB_RECLAIM_ACCOUNT|SLAB_MEM_SPREAD,
			    init_once, NULL);
L
Linus Torvalds 已提交
781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804
	if (jfs_inode_cachep == NULL)
		return -ENOMEM;

	/*
	 * Metapage initialization
	 */
	rc = metapage_init();
	if (rc) {
		jfs_err("metapage_init failed w/rc = %d", rc);
		goto free_slab;
	}

	/*
	 * Transaction Manager initialization
	 */
	rc = txInit();
	if (rc) {
		jfs_err("txInit failed w/rc = %d", rc);
		goto free_metapage;
	}

	/*
	 * I/O completion thread (endio)
	 */
C
Christoph Hellwig 已提交
805 806 807 808
	jfsIOthread = kthread_run(jfsIOWait, NULL, "jfsIO");
	if (IS_ERR(jfsIOthread)) {
		rc = PTR_ERR(jfsIOthread);
		jfs_err("init_jfs_fs: fork failed w/rc = %d", rc);
L
Linus Torvalds 已提交
809 810 811 812 813 814 815 816 817
		goto end_txmngr;
	}

	if (commit_threads < 1)
		commit_threads = num_online_cpus();
	if (commit_threads > MAX_COMMIT_THREADS)
		commit_threads = MAX_COMMIT_THREADS;

	for (i = 0; i < commit_threads; i++) {
C
Christoph Hellwig 已提交
818 819 820 821
		jfsCommitThread[i] = kthread_run(jfs_lazycommit, NULL, "jfsCommit");
		if (IS_ERR(jfsCommitThread[i])) {
			rc = PTR_ERR(jfsCommitThread[i]);
			jfs_err("init_jfs_fs: fork failed w/rc = %d", rc);
L
Linus Torvalds 已提交
822 823 824 825 826
			commit_threads = i;
			goto kill_committask;
		}
	}

C
Christoph Hellwig 已提交
827 828 829 830
	jfsSyncThread = kthread_run(jfs_sync, NULL, "jfsSync");
	if (IS_ERR(jfsSyncThread)) {
		rc = PTR_ERR(jfsSyncThread);
		jfs_err("init_jfs_fs: fork failed w/rc = %d", rc);
L
Linus Torvalds 已提交
831 832 833 834 835 836 837 838 839 840 841
		goto kill_committask;
	}

#ifdef PROC_FS_JFS
	jfs_proc_init();
#endif

	return register_filesystem(&jfs_fs_type);

kill_committask:
	for (i = 0; i < commit_threads; i++)
C
Christoph Hellwig 已提交
842 843
		kthread_stop(jfsCommitThread[i]);
	kthread_stop(jfsIOthread);
L
Linus Torvalds 已提交
844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860
end_txmngr:
	txExit();
free_metapage:
	metapage_exit();
free_slab:
	kmem_cache_destroy(jfs_inode_cachep);
	return rc;
}

static void __exit exit_jfs_fs(void)
{
	int i;

	jfs_info("exit_jfs_fs called");

	txExit();
	metapage_exit();
C
Christoph Hellwig 已提交
861 862

	kthread_stop(jfsIOthread);
L
Linus Torvalds 已提交
863
	for (i = 0; i < commit_threads; i++)
C
Christoph Hellwig 已提交
864 865
		kthread_stop(jfsCommitThread[i]);
	kthread_stop(jfsSyncThread);
L
Linus Torvalds 已提交
866 867 868 869 870 871 872 873 874
#ifdef PROC_FS_JFS
	jfs_proc_clean();
#endif
	unregister_filesystem(&jfs_fs_type);
	kmem_cache_destroy(jfs_inode_cachep);
}

module_init(init_jfs_fs)
module_exit(exit_jfs_fs)