cpfile.c 24.5 KB
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 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 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969
/*
 * cpfile.c - NILFS checkpoint file.
 *
 * Copyright (C) 2006-2008 Nippon Telegraph and Telephone Corporation.
 *
 * 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
 * the Free Software Foundation; either version 2 of the License, or
 * (at your option) any later version.
 *
 * 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
 * along with this program; if not, write to the Free Software
 * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
 *
 * Written by Koji Sato <koji@osrg.net>.
 */

#include <linux/kernel.h>
#include <linux/fs.h>
#include <linux/string.h>
#include <linux/buffer_head.h>
#include <linux/errno.h>
#include <linux/nilfs2_fs.h>
#include "mdt.h"
#include "cpfile.h"


static inline unsigned long
nilfs_cpfile_checkpoints_per_block(const struct inode *cpfile)
{
	return NILFS_MDT(cpfile)->mi_entries_per_block;
}

/* block number from the beginning of the file */
static unsigned long
nilfs_cpfile_get_blkoff(const struct inode *cpfile, __u64 cno)
{
	__u64 tcno = cno + NILFS_MDT(cpfile)->mi_first_entry_offset - 1;
	do_div(tcno, nilfs_cpfile_checkpoints_per_block(cpfile));
	return (unsigned long)tcno;
}

/* offset in block */
static unsigned long
nilfs_cpfile_get_offset(const struct inode *cpfile, __u64 cno)
{
	__u64 tcno = cno + NILFS_MDT(cpfile)->mi_first_entry_offset - 1;
	return do_div(tcno, nilfs_cpfile_checkpoints_per_block(cpfile));
}

static unsigned long
nilfs_cpfile_checkpoints_in_block(const struct inode *cpfile,
				  __u64 curr,
				  __u64 max)
{
	return min_t(__u64,
		     nilfs_cpfile_checkpoints_per_block(cpfile) -
		     nilfs_cpfile_get_offset(cpfile, curr),
		     max - curr);
}

static inline int nilfs_cpfile_is_in_first(const struct inode *cpfile,
					   __u64 cno)
{
	return nilfs_cpfile_get_blkoff(cpfile, cno) == 0;
}

static unsigned int
nilfs_cpfile_block_add_valid_checkpoints(const struct inode *cpfile,
					 struct buffer_head *bh,
					 void *kaddr,
					 unsigned int n)
{
	struct nilfs_checkpoint *cp = kaddr + bh_offset(bh);
	unsigned int count;

	count = le32_to_cpu(cp->cp_checkpoints_count) + n;
	cp->cp_checkpoints_count = cpu_to_le32(count);
	return count;
}

static unsigned int
nilfs_cpfile_block_sub_valid_checkpoints(const struct inode *cpfile,
					 struct buffer_head *bh,
					 void *kaddr,
					 unsigned int n)
{
	struct nilfs_checkpoint *cp = kaddr + bh_offset(bh);
	unsigned int count;

	WARN_ON(le32_to_cpu(cp->cp_checkpoints_count) < n);
	count = le32_to_cpu(cp->cp_checkpoints_count) - n;
	cp->cp_checkpoints_count = cpu_to_le32(count);
	return count;
}

static inline struct nilfs_cpfile_header *
nilfs_cpfile_block_get_header(const struct inode *cpfile,
			      struct buffer_head *bh,
			      void *kaddr)
{
	return kaddr + bh_offset(bh);
}

static struct nilfs_checkpoint *
nilfs_cpfile_block_get_checkpoint(const struct inode *cpfile, __u64 cno,
				  struct buffer_head *bh,
				  void *kaddr)
{
	return kaddr + bh_offset(bh) + nilfs_cpfile_get_offset(cpfile, cno) *
		NILFS_MDT(cpfile)->mi_entry_size;
}

static void nilfs_cpfile_block_init(struct inode *cpfile,
				    struct buffer_head *bh,
				    void *kaddr)
{
	struct nilfs_checkpoint *cp = kaddr + bh_offset(bh);
	size_t cpsz = NILFS_MDT(cpfile)->mi_entry_size;
	int n = nilfs_cpfile_checkpoints_per_block(cpfile);

	while (n-- > 0) {
		nilfs_checkpoint_set_invalid(cp);
		cp = (void *)cp + cpsz;
	}
}

static inline int nilfs_cpfile_get_header_block(struct inode *cpfile,
						struct buffer_head **bhp)
{
	return nilfs_mdt_get_block(cpfile, 0, 0, NULL, bhp);
}

static inline int nilfs_cpfile_get_checkpoint_block(struct inode *cpfile,
						    __u64 cno,
						    int create,
						    struct buffer_head **bhp)
{
	return nilfs_mdt_get_block(cpfile,
				   nilfs_cpfile_get_blkoff(cpfile, cno),
				   create, nilfs_cpfile_block_init, bhp);
}

static inline int nilfs_cpfile_delete_checkpoint_block(struct inode *cpfile,
						       __u64 cno)
{
	return nilfs_mdt_delete_block(cpfile,
				      nilfs_cpfile_get_blkoff(cpfile, cno));
}

/**
 * nilfs_cpfile_get_checkpoint - get a checkpoint
 * @cpfile: inode of checkpoint file
 * @cno: checkpoint number
 * @create: create flag
 * @cpp: pointer to a checkpoint
 * @bhp: pointer to a buffer head
 *
 * Description: nilfs_cpfile_get_checkpoint() acquires the checkpoint
 * specified by @cno. A new checkpoint will be created if @cno is the current
 * checkpoint number and @create is nonzero.
 *
 * Return Value: On success, 0 is returned, and the checkpoint and the
 * buffer head of the buffer on which the checkpoint is located are stored in
 * the place pointed by @cpp and @bhp, respectively. On error, one of the
 * following negative error codes is returned.
 *
 * %-EIO - I/O error.
 *
 * %-ENOMEM - Insufficient amount of memory available.
 *
 * %-ENOENT - No such checkpoint.
 *
 * %-EINVAL - invalid checkpoint.
 */
int nilfs_cpfile_get_checkpoint(struct inode *cpfile,
				__u64 cno,
				int create,
				struct nilfs_checkpoint **cpp,
				struct buffer_head **bhp)
{
	struct buffer_head *header_bh, *cp_bh;
	struct nilfs_cpfile_header *header;
	struct nilfs_checkpoint *cp;
	void *kaddr;
	int ret;

	if (unlikely(cno < 1 || cno > nilfs_mdt_cno(cpfile) ||
		     (cno < nilfs_mdt_cno(cpfile) && create)))
		return -EINVAL;

	down_write(&NILFS_MDT(cpfile)->mi_sem);

	ret = nilfs_cpfile_get_header_block(cpfile, &header_bh);
	if (ret < 0)
		goto out_sem;
	ret = nilfs_cpfile_get_checkpoint_block(cpfile, cno, create, &cp_bh);
	if (ret < 0)
		goto out_header;
	kaddr = kmap(cp_bh->b_page);
	cp = nilfs_cpfile_block_get_checkpoint(cpfile, cno, cp_bh, kaddr);
	if (nilfs_checkpoint_invalid(cp)) {
		if (!create) {
			kunmap(cp_bh->b_page);
			brelse(cp_bh);
			ret = -ENOENT;
			goto out_header;
		}
		/* a newly-created checkpoint */
		nilfs_checkpoint_clear_invalid(cp);
		if (!nilfs_cpfile_is_in_first(cpfile, cno))
			nilfs_cpfile_block_add_valid_checkpoints(cpfile, cp_bh,
								 kaddr, 1);
		mark_buffer_dirty(cp_bh);

		kaddr = kmap_atomic(header_bh->b_page);
		header = nilfs_cpfile_block_get_header(cpfile, header_bh,
						       kaddr);
		le64_add_cpu(&header->ch_ncheckpoints, 1);
		kunmap_atomic(kaddr);
		mark_buffer_dirty(header_bh);
		nilfs_mdt_mark_dirty(cpfile);
	}

	if (cpp != NULL)
		*cpp = cp;
	*bhp = cp_bh;

 out_header:
	brelse(header_bh);

 out_sem:
	up_write(&NILFS_MDT(cpfile)->mi_sem);
	return ret;
}

/**
 * nilfs_cpfile_put_checkpoint - put a checkpoint
 * @cpfile: inode of checkpoint file
 * @cno: checkpoint number
 * @bh: buffer head
 *
 * Description: nilfs_cpfile_put_checkpoint() releases the checkpoint
 * specified by @cno. @bh must be the buffer head which has been returned by
 * a previous call to nilfs_cpfile_get_checkpoint() with @cno.
 */
void nilfs_cpfile_put_checkpoint(struct inode *cpfile, __u64 cno,
				 struct buffer_head *bh)
{
	kunmap(bh->b_page);
	brelse(bh);
}

/**
 * nilfs_cpfile_delete_checkpoints - delete checkpoints
 * @cpfile: inode of checkpoint file
 * @start: start checkpoint number
 * @end: end checkpoint numer
 *
 * Description: nilfs_cpfile_delete_checkpoints() deletes the checkpoints in
 * the period from @start to @end, excluding @end itself. The checkpoints
 * which have been already deleted are ignored.
 *
 * Return Value: On success, 0 is returned. On error, one of the following
 * negative error codes is returned.
 *
 * %-EIO - I/O error.
 *
 * %-ENOMEM - Insufficient amount of memory available.
 *
 * %-EINVAL - invalid checkpoints.
 */
int nilfs_cpfile_delete_checkpoints(struct inode *cpfile,
				    __u64 start,
				    __u64 end)
{
	struct buffer_head *header_bh, *cp_bh;
	struct nilfs_cpfile_header *header;
	struct nilfs_checkpoint *cp;
	size_t cpsz = NILFS_MDT(cpfile)->mi_entry_size;
	__u64 cno;
	void *kaddr;
	unsigned long tnicps;
	int ret, ncps, nicps, nss, count, i;

	if (unlikely(start == 0 || start > end)) {
		printk(KERN_ERR "%s: invalid range of checkpoint numbers: "
		       "[%llu, %llu)\n", __func__,
		       (unsigned long long)start, (unsigned long long)end);
		return -EINVAL;
	}

	down_write(&NILFS_MDT(cpfile)->mi_sem);

	ret = nilfs_cpfile_get_header_block(cpfile, &header_bh);
	if (ret < 0)
		goto out_sem;
	tnicps = 0;
	nss = 0;

	for (cno = start; cno < end; cno += ncps) {
		ncps = nilfs_cpfile_checkpoints_in_block(cpfile, cno, end);
		ret = nilfs_cpfile_get_checkpoint_block(cpfile, cno, 0, &cp_bh);
		if (ret < 0) {
			if (ret != -ENOENT)
				break;
			/* skip hole */
			ret = 0;
			continue;
		}

		kaddr = kmap_atomic(cp_bh->b_page);
		cp = nilfs_cpfile_block_get_checkpoint(
			cpfile, cno, cp_bh, kaddr);
		nicps = 0;
		for (i = 0; i < ncps; i++, cp = (void *)cp + cpsz) {
			if (nilfs_checkpoint_snapshot(cp)) {
				nss++;
			} else if (!nilfs_checkpoint_invalid(cp)) {
				nilfs_checkpoint_set_invalid(cp);
				nicps++;
			}
		}
		if (nicps > 0) {
			tnicps += nicps;
			mark_buffer_dirty(cp_bh);
			nilfs_mdt_mark_dirty(cpfile);
			if (!nilfs_cpfile_is_in_first(cpfile, cno)) {
				count =
				  nilfs_cpfile_block_sub_valid_checkpoints(
						cpfile, cp_bh, kaddr, nicps);
				if (count == 0) {
					/* make hole */
					kunmap_atomic(kaddr);
					brelse(cp_bh);
					ret =
					  nilfs_cpfile_delete_checkpoint_block(
								   cpfile, cno);
					if (ret == 0)
						continue;
					printk(KERN_ERR
					       "%s: cannot delete block\n",
					       __func__);
					break;
				}
			}
		}

		kunmap_atomic(kaddr);
		brelse(cp_bh);
	}

	if (tnicps > 0) {
		kaddr = kmap_atomic(header_bh->b_page);
		header = nilfs_cpfile_block_get_header(cpfile, header_bh,
						       kaddr);
		le64_add_cpu(&header->ch_ncheckpoints, -(u64)tnicps);
		mark_buffer_dirty(header_bh);
		nilfs_mdt_mark_dirty(cpfile);
		kunmap_atomic(kaddr);
	}

	brelse(header_bh);
	if (nss > 0)
		ret = -EBUSY;

 out_sem:
	up_write(&NILFS_MDT(cpfile)->mi_sem);
	return ret;
}

static void nilfs_cpfile_checkpoint_to_cpinfo(struct inode *cpfile,
					      struct nilfs_checkpoint *cp,
					      struct nilfs_cpinfo *ci)
{
	ci->ci_flags = le32_to_cpu(cp->cp_flags);
	ci->ci_cno = le64_to_cpu(cp->cp_cno);
	ci->ci_create = le64_to_cpu(cp->cp_create);
	ci->ci_nblk_inc = le64_to_cpu(cp->cp_nblk_inc);
	ci->ci_inodes_count = le64_to_cpu(cp->cp_inodes_count);
	ci->ci_blocks_count = le64_to_cpu(cp->cp_blocks_count);
	ci->ci_next = le64_to_cpu(cp->cp_snapshot_list.ssl_next);
}

static ssize_t nilfs_cpfile_do_get_cpinfo(struct inode *cpfile, __u64 *cnop,
					  void *buf, unsigned cisz, size_t nci)
{
	struct nilfs_checkpoint *cp;
	struct nilfs_cpinfo *ci = buf;
	struct buffer_head *bh;
	size_t cpsz = NILFS_MDT(cpfile)->mi_entry_size;
	__u64 cur_cno = nilfs_mdt_cno(cpfile), cno = *cnop;
	void *kaddr;
	int n, ret;
	int ncps, i;

	if (cno == 0)
		return -ENOENT; /* checkpoint number 0 is invalid */
	down_read(&NILFS_MDT(cpfile)->mi_sem);

	for (n = 0; cno < cur_cno && n < nci; cno += ncps) {
		ncps = nilfs_cpfile_checkpoints_in_block(cpfile, cno, cur_cno);
		ret = nilfs_cpfile_get_checkpoint_block(cpfile, cno, 0, &bh);
		if (ret < 0) {
			if (ret != -ENOENT)
				goto out;
			continue; /* skip hole */
		}

		kaddr = kmap_atomic(bh->b_page);
		cp = nilfs_cpfile_block_get_checkpoint(cpfile, cno, bh, kaddr);
		for (i = 0; i < ncps && n < nci; i++, cp = (void *)cp + cpsz) {
			if (!nilfs_checkpoint_invalid(cp)) {
				nilfs_cpfile_checkpoint_to_cpinfo(cpfile, cp,
								  ci);
				ci = (void *)ci + cisz;
				n++;
			}
		}
		kunmap_atomic(kaddr);
		brelse(bh);
	}

	ret = n;
	if (n > 0) {
		ci = (void *)ci - cisz;
		*cnop = ci->ci_cno + 1;
	}

 out:
	up_read(&NILFS_MDT(cpfile)->mi_sem);
	return ret;
}

static ssize_t nilfs_cpfile_do_get_ssinfo(struct inode *cpfile, __u64 *cnop,
					  void *buf, unsigned cisz, size_t nci)
{
	struct buffer_head *bh;
	struct nilfs_cpfile_header *header;
	struct nilfs_checkpoint *cp;
	struct nilfs_cpinfo *ci = buf;
	__u64 curr = *cnop, next;
	unsigned long curr_blkoff, next_blkoff;
	void *kaddr;
	int n = 0, ret;

	down_read(&NILFS_MDT(cpfile)->mi_sem);

	if (curr == 0) {
		ret = nilfs_cpfile_get_header_block(cpfile, &bh);
		if (ret < 0)
			goto out;
		kaddr = kmap_atomic(bh->b_page);
		header = nilfs_cpfile_block_get_header(cpfile, bh, kaddr);
		curr = le64_to_cpu(header->ch_snapshot_list.ssl_next);
		kunmap_atomic(kaddr);
		brelse(bh);
		if (curr == 0) {
			ret = 0;
			goto out;
		}
	} else if (unlikely(curr == ~(__u64)0)) {
		ret = 0;
		goto out;
	}

	curr_blkoff = nilfs_cpfile_get_blkoff(cpfile, curr);
	ret = nilfs_cpfile_get_checkpoint_block(cpfile, curr, 0, &bh);
	if (unlikely(ret < 0)) {
		if (ret == -ENOENT)
			ret = 0; /* No snapshots (started from a hole block) */
		goto out;
	}
	kaddr = kmap_atomic(bh->b_page);
	while (n < nci) {
		cp = nilfs_cpfile_block_get_checkpoint(cpfile, curr, bh, kaddr);
		curr = ~(__u64)0; /* Terminator */
		if (unlikely(nilfs_checkpoint_invalid(cp) ||
			     !nilfs_checkpoint_snapshot(cp)))
			break;
		nilfs_cpfile_checkpoint_to_cpinfo(cpfile, cp, ci);
		ci = (void *)ci + cisz;
		n++;
		next = le64_to_cpu(cp->cp_snapshot_list.ssl_next);
		if (next == 0)
			break; /* reach end of the snapshot list */

		next_blkoff = nilfs_cpfile_get_blkoff(cpfile, next);
		if (curr_blkoff != next_blkoff) {
			kunmap_atomic(kaddr);
			brelse(bh);
			ret = nilfs_cpfile_get_checkpoint_block(cpfile, next,
								0, &bh);
			if (unlikely(ret < 0)) {
				WARN_ON(ret == -ENOENT);
				goto out;
			}
			kaddr = kmap_atomic(bh->b_page);
		}
		curr = next;
		curr_blkoff = next_blkoff;
	}
	kunmap_atomic(kaddr);
	brelse(bh);
	*cnop = curr;
	ret = n;

 out:
	up_read(&NILFS_MDT(cpfile)->mi_sem);
	return ret;
}

/**
 * nilfs_cpfile_get_cpinfo -
 * @cpfile:
 * @cno:
 * @ci:
 * @nci:
 */

ssize_t nilfs_cpfile_get_cpinfo(struct inode *cpfile, __u64 *cnop, int mode,
				void *buf, unsigned cisz, size_t nci)
{
	switch (mode) {
	case NILFS_CHECKPOINT:
		return nilfs_cpfile_do_get_cpinfo(cpfile, cnop, buf, cisz, nci);
	case NILFS_SNAPSHOT:
		return nilfs_cpfile_do_get_ssinfo(cpfile, cnop, buf, cisz, nci);
	default:
		return -EINVAL;
	}
}

/**
 * nilfs_cpfile_delete_checkpoint -
 * @cpfile:
 * @cno:
 */
int nilfs_cpfile_delete_checkpoint(struct inode *cpfile, __u64 cno)
{
	struct nilfs_cpinfo ci;
	__u64 tcno = cno;
	ssize_t nci;

	nci = nilfs_cpfile_do_get_cpinfo(cpfile, &tcno, &ci, sizeof(ci), 1);
	if (nci < 0)
		return nci;
	else if (nci == 0 || ci.ci_cno != cno)
		return -ENOENT;
	else if (nilfs_cpinfo_snapshot(&ci))
		return -EBUSY;

	return nilfs_cpfile_delete_checkpoints(cpfile, cno, cno + 1);
}

static struct nilfs_snapshot_list *
nilfs_cpfile_block_get_snapshot_list(const struct inode *cpfile,
				     __u64 cno,
				     struct buffer_head *bh,
				     void *kaddr)
{
	struct nilfs_cpfile_header *header;
	struct nilfs_checkpoint *cp;
	struct nilfs_snapshot_list *list;

	if (cno != 0) {
		cp = nilfs_cpfile_block_get_checkpoint(cpfile, cno, bh, kaddr);
		list = &cp->cp_snapshot_list;
	} else {
		header = nilfs_cpfile_block_get_header(cpfile, bh, kaddr);
		list = &header->ch_snapshot_list;
	}
	return list;
}

static int nilfs_cpfile_set_snapshot(struct inode *cpfile, __u64 cno)
{
	struct buffer_head *header_bh, *curr_bh, *prev_bh, *cp_bh;
	struct nilfs_cpfile_header *header;
	struct nilfs_checkpoint *cp;
	struct nilfs_snapshot_list *list;
	__u64 curr, prev;
	unsigned long curr_blkoff, prev_blkoff;
	void *kaddr;
	int ret;

	if (cno == 0)
		return -ENOENT; /* checkpoint number 0 is invalid */
	down_write(&NILFS_MDT(cpfile)->mi_sem);

	ret = nilfs_cpfile_get_checkpoint_block(cpfile, cno, 0, &cp_bh);
	if (ret < 0)
		goto out_sem;
	kaddr = kmap_atomic(cp_bh->b_page);
	cp = nilfs_cpfile_block_get_checkpoint(cpfile, cno, cp_bh, kaddr);
	if (nilfs_checkpoint_invalid(cp)) {
		ret = -ENOENT;
		kunmap_atomic(kaddr);
		goto out_cp;
	}
	if (nilfs_checkpoint_snapshot(cp)) {
		ret = 0;
		kunmap_atomic(kaddr);
		goto out_cp;
	}
	kunmap_atomic(kaddr);

	ret = nilfs_cpfile_get_header_block(cpfile, &header_bh);
	if (ret < 0)
		goto out_cp;
	kaddr = kmap_atomic(header_bh->b_page);
	header = nilfs_cpfile_block_get_header(cpfile, header_bh, kaddr);
	list = &header->ch_snapshot_list;
	curr_bh = header_bh;
	get_bh(curr_bh);
	curr = 0;
	curr_blkoff = 0;
	prev = le64_to_cpu(list->ssl_prev);
	while (prev > cno) {
		prev_blkoff = nilfs_cpfile_get_blkoff(cpfile, prev);
		curr = prev;
		if (curr_blkoff != prev_blkoff) {
			kunmap_atomic(kaddr);
			brelse(curr_bh);
			ret = nilfs_cpfile_get_checkpoint_block(cpfile, curr,
								0, &curr_bh);
			if (ret < 0)
				goto out_header;
			kaddr = kmap_atomic(curr_bh->b_page);
		}
		curr_blkoff = prev_blkoff;
		cp = nilfs_cpfile_block_get_checkpoint(
			cpfile, curr, curr_bh, kaddr);
		list = &cp->cp_snapshot_list;
		prev = le64_to_cpu(list->ssl_prev);
	}
	kunmap_atomic(kaddr);

	if (prev != 0) {
		ret = nilfs_cpfile_get_checkpoint_block(cpfile, prev, 0,
							&prev_bh);
		if (ret < 0)
			goto out_curr;
	} else {
		prev_bh = header_bh;
		get_bh(prev_bh);
	}

	kaddr = kmap_atomic(curr_bh->b_page);
	list = nilfs_cpfile_block_get_snapshot_list(
		cpfile, curr, curr_bh, kaddr);
	list->ssl_prev = cpu_to_le64(cno);
	kunmap_atomic(kaddr);

	kaddr = kmap_atomic(cp_bh->b_page);
	cp = nilfs_cpfile_block_get_checkpoint(cpfile, cno, cp_bh, kaddr);
	cp->cp_snapshot_list.ssl_next = cpu_to_le64(curr);
	cp->cp_snapshot_list.ssl_prev = cpu_to_le64(prev);
	nilfs_checkpoint_set_snapshot(cp);
	kunmap_atomic(kaddr);

	kaddr = kmap_atomic(prev_bh->b_page);
	list = nilfs_cpfile_block_get_snapshot_list(
		cpfile, prev, prev_bh, kaddr);
	list->ssl_next = cpu_to_le64(cno);
	kunmap_atomic(kaddr);

	kaddr = kmap_atomic(header_bh->b_page);
	header = nilfs_cpfile_block_get_header(cpfile, header_bh, kaddr);
	le64_add_cpu(&header->ch_nsnapshots, 1);
	kunmap_atomic(kaddr);

	mark_buffer_dirty(prev_bh);
	mark_buffer_dirty(curr_bh);
	mark_buffer_dirty(cp_bh);
	mark_buffer_dirty(header_bh);
	nilfs_mdt_mark_dirty(cpfile);

	brelse(prev_bh);

 out_curr:
	brelse(curr_bh);

 out_header:
	brelse(header_bh);

 out_cp:
	brelse(cp_bh);

 out_sem:
	up_write(&NILFS_MDT(cpfile)->mi_sem);
	return ret;
}

static int nilfs_cpfile_clear_snapshot(struct inode *cpfile, __u64 cno)
{
	struct buffer_head *header_bh, *next_bh, *prev_bh, *cp_bh;
	struct nilfs_cpfile_header *header;
	struct nilfs_checkpoint *cp;
	struct nilfs_snapshot_list *list;
	__u64 next, prev;
	void *kaddr;
	int ret;

	if (cno == 0)
		return -ENOENT; /* checkpoint number 0 is invalid */
	down_write(&NILFS_MDT(cpfile)->mi_sem);

	ret = nilfs_cpfile_get_checkpoint_block(cpfile, cno, 0, &cp_bh);
	if (ret < 0)
		goto out_sem;
	kaddr = kmap_atomic(cp_bh->b_page);
	cp = nilfs_cpfile_block_get_checkpoint(cpfile, cno, cp_bh, kaddr);
	if (nilfs_checkpoint_invalid(cp)) {
		ret = -ENOENT;
		kunmap_atomic(kaddr);
		goto out_cp;
	}
	if (!nilfs_checkpoint_snapshot(cp)) {
		ret = 0;
		kunmap_atomic(kaddr);
		goto out_cp;
	}

	list = &cp->cp_snapshot_list;
	next = le64_to_cpu(list->ssl_next);
	prev = le64_to_cpu(list->ssl_prev);
	kunmap_atomic(kaddr);

	ret = nilfs_cpfile_get_header_block(cpfile, &header_bh);
	if (ret < 0)
		goto out_cp;
	if (next != 0) {
		ret = nilfs_cpfile_get_checkpoint_block(cpfile, next, 0,
							&next_bh);
		if (ret < 0)
			goto out_header;
	} else {
		next_bh = header_bh;
		get_bh(next_bh);
	}
	if (prev != 0) {
		ret = nilfs_cpfile_get_checkpoint_block(cpfile, prev, 0,
							&prev_bh);
		if (ret < 0)
			goto out_next;
	} else {
		prev_bh = header_bh;
		get_bh(prev_bh);
	}

	kaddr = kmap_atomic(next_bh->b_page);
	list = nilfs_cpfile_block_get_snapshot_list(
		cpfile, next, next_bh, kaddr);
	list->ssl_prev = cpu_to_le64(prev);
	kunmap_atomic(kaddr);

	kaddr = kmap_atomic(prev_bh->b_page);
	list = nilfs_cpfile_block_get_snapshot_list(
		cpfile, prev, prev_bh, kaddr);
	list->ssl_next = cpu_to_le64(next);
	kunmap_atomic(kaddr);

	kaddr = kmap_atomic(cp_bh->b_page);
	cp = nilfs_cpfile_block_get_checkpoint(cpfile, cno, cp_bh, kaddr);
	cp->cp_snapshot_list.ssl_next = cpu_to_le64(0);
	cp->cp_snapshot_list.ssl_prev = cpu_to_le64(0);
	nilfs_checkpoint_clear_snapshot(cp);
	kunmap_atomic(kaddr);

	kaddr = kmap_atomic(header_bh->b_page);
	header = nilfs_cpfile_block_get_header(cpfile, header_bh, kaddr);
	le64_add_cpu(&header->ch_nsnapshots, -1);
	kunmap_atomic(kaddr);

	mark_buffer_dirty(next_bh);
	mark_buffer_dirty(prev_bh);
	mark_buffer_dirty(cp_bh);
	mark_buffer_dirty(header_bh);
	nilfs_mdt_mark_dirty(cpfile);

	brelse(prev_bh);

 out_next:
	brelse(next_bh);

 out_header:
	brelse(header_bh);

 out_cp:
	brelse(cp_bh);

 out_sem:
	up_write(&NILFS_MDT(cpfile)->mi_sem);
	return ret;
}

/**
 * nilfs_cpfile_is_snapshot -
 * @cpfile: inode of checkpoint file
 * @cno: checkpoint number
 *
 * Description:
 *
 * Return Value: On success, 1 is returned if the checkpoint specified by
 * @cno is a snapshot, or 0 if not. On error, one of the following negative
 * error codes is returned.
 *
 * %-EIO - I/O error.
 *
 * %-ENOMEM - Insufficient amount of memory available.
 *
 * %-ENOENT - No such checkpoint.
 */
int nilfs_cpfile_is_snapshot(struct inode *cpfile, __u64 cno)
{
	struct buffer_head *bh;
	struct nilfs_checkpoint *cp;
	void *kaddr;
	int ret;

	/* CP number is invalid if it's zero or larger than the
	largest	exist one.*/
	if (cno == 0 || cno >= nilfs_mdt_cno(cpfile))
		return -ENOENT;
	down_read(&NILFS_MDT(cpfile)->mi_sem);

	ret = nilfs_cpfile_get_checkpoint_block(cpfile, cno, 0, &bh);
	if (ret < 0)
		goto out;
	kaddr = kmap_atomic(bh->b_page);
	cp = nilfs_cpfile_block_get_checkpoint(cpfile, cno, bh, kaddr);
	if (nilfs_checkpoint_invalid(cp))
		ret = -ENOENT;
	else
		ret = nilfs_checkpoint_snapshot(cp);
	kunmap_atomic(kaddr);
	brelse(bh);

 out:
	up_read(&NILFS_MDT(cpfile)->mi_sem);
	return ret;
}

/**
 * nilfs_cpfile_change_cpmode - change checkpoint mode
 * @cpfile: inode of checkpoint file
 * @cno: checkpoint number
 * @status: mode of checkpoint
 *
 * Description: nilfs_change_cpmode() changes the mode of the checkpoint
 * specified by @cno. The mode @mode is NILFS_CHECKPOINT or NILFS_SNAPSHOT.
 *
 * Return Value: On success, 0 is returned. On error, one of the following
 * negative error codes is returned.
 *
 * %-EIO - I/O error.
 *
 * %-ENOMEM - Insufficient amount of memory available.
 *
 * %-ENOENT - No such checkpoint.
 */
int nilfs_cpfile_change_cpmode(struct inode *cpfile, __u64 cno, int mode)
{
	int ret;

	switch (mode) {
	case NILFS_CHECKPOINT:
		if (nilfs_checkpoint_is_mounted(cpfile->i_sb, cno))
			/*
			 * Current implementation does not have to protect
			 * plain read-only mounts since they are exclusive
			 * with a read/write mount and are protected from the
			 * cleaner.
			 */
			ret = -EBUSY;
		else
			ret = nilfs_cpfile_clear_snapshot(cpfile, cno);
		return ret;
	case NILFS_SNAPSHOT:
		return nilfs_cpfile_set_snapshot(cpfile, cno);
	default:
		return -EINVAL;
	}
}

/**
 * nilfs_cpfile_get_stat - get checkpoint statistics
 * @cpfile: inode of checkpoint file
 * @stat: pointer to a structure of checkpoint statistics
 *
 * Description: nilfs_cpfile_get_stat() returns information about checkpoints.
 *
 * Return Value: On success, 0 is returned, and checkpoints information is
 * stored in the place pointed by @stat. On error, one of the following
 * negative error codes is returned.
 *
 * %-EIO - I/O error.
 *
 * %-ENOMEM - Insufficient amount of memory available.
 */
int nilfs_cpfile_get_stat(struct inode *cpfile, struct nilfs_cpstat *cpstat)
{
	struct buffer_head *bh;
	struct nilfs_cpfile_header *header;
	void *kaddr;
	int ret;

	down_read(&NILFS_MDT(cpfile)->mi_sem);

	ret = nilfs_cpfile_get_header_block(cpfile, &bh);
	if (ret < 0)
		goto out_sem;
	kaddr = kmap_atomic(bh->b_page);
	header = nilfs_cpfile_block_get_header(cpfile, bh, kaddr);
	cpstat->cs_cno = nilfs_mdt_cno(cpfile);
	cpstat->cs_ncps = le64_to_cpu(header->ch_ncheckpoints);
	cpstat->cs_nsss = le64_to_cpu(header->ch_nsnapshots);
	kunmap_atomic(kaddr);
	brelse(bh);

 out_sem:
	up_read(&NILFS_MDT(cpfile)->mi_sem);
	return ret;
}

/**
 * nilfs_cpfile_read - read or get cpfile inode
 * @sb: super block instance
 * @cpsize: size of a checkpoint entry
 * @raw_inode: on-disk cpfile inode
 * @inodep: buffer to store the inode
 */
int nilfs_cpfile_read(struct super_block *sb, size_t cpsize,
		      struct nilfs_inode *raw_inode, struct inode **inodep)
{
	struct inode *cpfile;
	int err;

	cpfile = nilfs_iget_locked(sb, NULL, NILFS_CPFILE_INO);
	if (unlikely(!cpfile))
		return -ENOMEM;
	if (!(cpfile->i_state & I_NEW))
		goto out;

	err = nilfs_mdt_init(cpfile, NILFS_MDT_GFP, 0);
	if (err)
		goto failed;

	nilfs_mdt_set_entry_size(cpfile, cpsize,
				 sizeof(struct nilfs_cpfile_header));

	err = nilfs_read_inode_common(cpfile, raw_inode);
	if (err)
		goto failed;

	unlock_new_inode(cpfile);
 out:
	*inodep = cpfile;
	return 0;
 failed:
	iget_failed(cpfile);
	return err;
}