super.c

嵌入式系统设计与实例开发源码

		fsync_no_super(j_dev);
		invalidate_buffers(j_dev);
		ext3_blkdev_remove(sbi);
	}
	clear_ro_after(sb);

	return;
}

static struct super_operations ext3_sops = {
	read_inode:	ext3_read_inode,	/* BKL held */
	write_inode:	ext3_write_inode,	/* BKL not held.  Don't need */
	dirty_inode:	ext3_dirty_inode,	/* BKL not held.  We take it */
	put_inode:	ext3_put_inode,		/* BKL not held.  Don't need */
	delete_inode:	ext3_delete_inode,	/* BKL not held.  We take it */
	put_super:	ext3_put_super,		/* BKL held */
	write_super:	ext3_write_super,	/* BKL held */
	write_super_lockfs: ext3_write_super_lockfs, /* BKL not held. Take it */
	unlockfs:	ext3_unlockfs,		/* BKL not held.  We take it */
	statfs:		ext3_statfs,		/* BKL held */
	remount_fs:	ext3_remount,		/* BKL held */
};

static int want_value(char *value, char *option)
{
	if (!value || !*value) {
		printk(KERN_NOTICE "EXT3-fs: the %s option needs an argument\n",
		       option);
		return -1;
	}
	return 0;
}

static int want_null_value(char *value, char *option)
{
	if (*value) {
		printk(KERN_NOTICE "EXT3-fs: Invalid %s argument: %s\n",
		       option, value);
		return -1;
	}
	return 0;
}

static int want_numeric(char *value, char *option, unsigned long *number)
{
	if (want_value(value, option))
		return -1;
	*number = simple_strtoul(value, &value, 0);
	if (want_null_value(value, option))
		return -1;
	return 0;
}

/*
 * This function has been shamelessly adapted from the msdos fs
 */
static int parse_options (char * options, unsigned long * sb_block,
			  struct ext3_sb_info *sbi,
			  unsigned long * inum,
			  int is_remount)
{
	unsigned long *mount_options = &sbi->s_mount_opt;
	uid_t *resuid = &sbi->s_resuid;
	gid_t *resgid = &sbi->s_resgid;
	char * this_char;
	char * value;

	if (!options)
		return 1;
	for (this_char = strtok (options, ",");
	     this_char != NULL;
	     this_char = strtok (NULL, ",")) {
		if ((value = strchr (this_char, '=')) != NULL)
			*value++ = 0;
		if (!strcmp (this_char, "bsddf"))
			clear_opt (*mount_options, MINIX_DF);
		else if (!strcmp (this_char, "nouid32")) {
			set_opt (*mount_options, NO_UID32);
		}
		else if (!strcmp (this_char, "abort"))
			set_opt (*mount_options, ABORT);
		else if (!strcmp (this_char, "check")) {
			if (!value || !*value || !strcmp (value, "none"))
				clear_opt (*mount_options, CHECK);
			else
#ifdef CONFIG_EXT3_CHECK
				set_opt (*mount_options, CHECK);
#else
				printk(KERN_ERR 
				       "EXT3 Check option not supported\n");
#endif
		}
		else if (!strcmp (this_char, "debug"))
			set_opt (*mount_options, DEBUG);
		else if (!strcmp (this_char, "errors")) {
			if (want_value(value, "errors"))
				return 0;
			if (!strcmp (value, "continue")) {
				clear_opt (*mount_options, ERRORS_RO);
				clear_opt (*mount_options, ERRORS_PANIC);
				set_opt (*mount_options, ERRORS_CONT);
			}
			else if (!strcmp (value, "remount-ro")) {
				clear_opt (*mount_options, ERRORS_CONT);
				clear_opt (*mount_options, ERRORS_PANIC);
				set_opt (*mount_options, ERRORS_RO);
			}
			else if (!strcmp (value, "panic")) {
				clear_opt (*mount_options, ERRORS_CONT);
				clear_opt (*mount_options, ERRORS_RO);
				set_opt (*mount_options, ERRORS_PANIC);
			}
			else {
				printk (KERN_ERR
					"EXT3-fs: Invalid errors option: %s\n",
					value);
				return 0;
			}
		}
		else if (!strcmp (this_char, "grpid") ||
			 !strcmp (this_char, "bsdgroups"))
			set_opt (*mount_options, GRPID);
		else if (!strcmp (this_char, "minixdf"))
			set_opt (*mount_options, MINIX_DF);
		else if (!strcmp (this_char, "nocheck"))
			clear_opt (*mount_options, CHECK);
		else if (!strcmp (this_char, "nogrpid") ||
			 !strcmp (this_char, "sysvgroups"))
			clear_opt (*mount_options, GRPID);
		else if (!strcmp (this_char, "resgid")) {
			unsigned long v;
			if (want_numeric(value, "resgid", &v))
				return 0;
			*resgid = v;
		}
		else if (!strcmp (this_char, "resuid")) {
			unsigned long v;
			if (want_numeric(value, "resuid", &v))
				return 0;
			*resuid = v;
		}
		else if (!strcmp (this_char, "sb")) {
			if (want_numeric(value, "sb", sb_block))
				return 0;
		}
#ifdef CONFIG_JBD_DEBUG
		else if (!strcmp (this_char, "ro-after")) {
			unsigned long v;
			if (want_numeric(value, "ro-after", &v))
				return 0;
			ext3_ro_after = v;
		}
#endif
		/* Silently ignore the quota options */
		else if (!strcmp (this_char, "grpquota")
		         || !strcmp (this_char, "noquota")
		         || !strcmp (this_char, "quota")
		         || !strcmp (this_char, "usrquota"))
			/* Don't do anything ;-) */ ;
		else if (!strcmp (this_char, "journal")) {
			/* @@@ FIXME */
			/* Eventually we will want to be able to create
                           a journal file here.  For now, only allow the
                           user to specify an existing inode to be the
                           journal file. */
			if (is_remount) {
				printk(KERN_ERR "EXT3-fs: cannot specify "
				       "journal on remount\n");
				return 0;
			}

			if (want_value(value, "journal"))
				return 0;
			if (!strcmp (value, "update"))
				set_opt (*mount_options, UPDATE_JOURNAL);
			else if (want_numeric(value, "journal", inum))
				return 0;
		}
		else if (!strcmp (this_char, "noload"))
			set_opt (*mount_options, NOLOAD);
		else if (!strcmp (this_char, "data")) {
			int data_opt = 0;

			if (want_value(value, "data"))
				return 0;
			if (!strcmp (value, "journal"))
				data_opt = EXT3_MOUNT_JOURNAL_DATA;
			else if (!strcmp (value, "ordered"))
				data_opt = EXT3_MOUNT_ORDERED_DATA;
			else if (!strcmp (value, "writeback"))
				data_opt = EXT3_MOUNT_WRITEBACK_DATA;
			else {
				printk (KERN_ERR 
					"EXT3-fs: Invalid data option: %s\n",
					value);
				return 0;
			}
			if (is_remount) {
				if ((*mount_options & EXT3_MOUNT_DATA_FLAGS) !=
							data_opt) {
					printk(KERN_ERR
					       "EXT3-fs: cannot change data "
					       "mode on remount\n");
					return 0;
				}
			} else {
				*mount_options &= ~EXT3_MOUNT_DATA_FLAGS;
				*mount_options |= data_opt;
			}
		} else {
			printk (KERN_ERR 
				"EXT3-fs: Unrecognized mount option %s\n",
				this_char);
			return 0;
		}
	}
	return 1;
}

static int ext3_setup_super(struct super_block *sb, struct ext3_super_block *es,
			    int read_only)
{
	struct ext3_sb_info *sbi = EXT3_SB(sb);
	int res = 0;

	if (le32_to_cpu(es->s_rev_level) > EXT3_MAX_SUPP_REV) {
		printk (KERN_ERR "EXT3-fs warning: revision level too high, "
			"forcing read-only mode\n");
		res = MS_RDONLY;
	}
	if (read_only)
		return res;
	if (!(sbi->s_mount_state & EXT3_VALID_FS))
		printk (KERN_WARNING "EXT3-fs warning: mounting unchecked fs, "
			"running e2fsck is recommended\n");
	else if ((sbi->s_mount_state & EXT3_ERROR_FS))
		printk (KERN_WARNING
			"EXT3-fs warning: mounting fs with errors, "
			"running e2fsck is recommended\n");
	else if ((__s16) le16_to_cpu(es->s_max_mnt_count) >= 0 &&
		 le16_to_cpu(es->s_mnt_count) >=
		 (unsigned short) (__s16) le16_to_cpu(es->s_max_mnt_count))
		printk (KERN_WARNING
			"EXT3-fs warning: maximal mount count reached, "
			"running e2fsck is recommended\n");
	else if (le32_to_cpu(es->s_checkinterval) &&
		(le32_to_cpu(es->s_lastcheck) +
			le32_to_cpu(es->s_checkinterval) <= CURRENT_TIME))
		printk (KERN_WARNING
			"EXT3-fs warning: checktime reached, "
			"running e2fsck is recommended\n");
#if 0
		/* @@@ We _will_ want to clear the valid bit if we find
                   inconsistencies, to force a fsck at reboot.  But for
                   a plain journaled filesystem we can keep it set as
                   valid forever! :) */
	es->s_state = cpu_to_le16(le16_to_cpu(es->s_state) & ~EXT3_VALID_FS);
#endif
	if (!(__s16) le16_to_cpu(es->s_max_mnt_count))
		es->s_max_mnt_count =
			(__s16) cpu_to_le16(EXT3_DFL_MAX_MNT_COUNT);
	es->s_mnt_count=cpu_to_le16(le16_to_cpu(es->s_mnt_count) + 1);
	es->s_mtime = cpu_to_le32(CURRENT_TIME);
	ext3_update_dynamic_rev(sb);
	EXT3_SET_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_RECOVER);
	ext3_commit_super (sb, es, 1);
	if (test_opt (sb, DEBUG))
		printk (KERN_INFO
			"[EXT3 FS %s, %s, bs=%lu, gc=%lu, "
			"bpg=%lu, ipg=%lu, mo=%04lx]\n",
			EXT3FS_VERSION, EXT3FS_DATE, sb->s_blocksize,
			sbi->s_groups_count,
			EXT3_BLOCKS_PER_GROUP(sb),
			EXT3_INODES_PER_GROUP(sb),
			sbi->s_mount_opt);
	printk(KERN_INFO "EXT3 FS " EXT3FS_VERSION ", " EXT3FS_DATE " on %s, ",
				bdevname(sb->s_dev));
	if (EXT3_SB(sb)->s_journal->j_inode == NULL) {
		printk("external journal on %s\n",
				bdevname(EXT3_SB(sb)->s_journal->j_dev));
	} else {
		printk("internal journal\n");
	}
#ifdef CONFIG_EXT3_CHECK
	if (test_opt (sb, CHECK)) {
		ext3_check_blocks_bitmap (sb);
		ext3_check_inodes_bitmap (sb);
	}
#endif
	setup_ro_after(sb);
	return res;
}

static int ext3_check_descriptors (struct super_block * sb)
{
	struct ext3_sb_info *sbi = EXT3_SB(sb);
	unsigned long block = le32_to_cpu(sbi->s_es->s_first_data_block);
	struct ext3_group_desc * gdp = NULL;
	int desc_block = 0;
	int i;

	ext3_debug ("Checking group descriptors");

	for (i = 0; i < sbi->s_groups_count; i++)
	{
		if ((i % EXT3_DESC_PER_BLOCK(sb)) == 0)
			gdp = (struct ext3_group_desc *)
					sbi->s_group_desc[desc_block++]->b_data;
		if (le32_to_cpu(gdp->bg_block_bitmap) < block ||
		    le32_to_cpu(gdp->bg_block_bitmap) >=
				block + EXT3_BLOCKS_PER_GROUP(sb))
		{
			ext3_error (sb, "ext3_check_descriptors",
				    "Block bitmap for group %d"
				    " not in group (block %lu)!",
				    i, (unsigned long)
					le32_to_cpu(gdp->bg_block_bitmap));
			return 0;
		}
		if (le32_to_cpu(gdp->bg_inode_bitmap) < block ||
		    le32_to_cpu(gdp->bg_inode_bitmap) >=
				block + EXT3_BLOCKS_PER_GROUP(sb))
		{
			ext3_error (sb, "ext3_check_descriptors",
				    "Inode bitmap for group %d"
				    " not in group (block %lu)!",
				    i, (unsigned long)
					le32_to_cpu(gdp->bg_inode_bitmap));
			return 0;
		}
		if (le32_to_cpu(gdp->bg_inode_table) < block ||
		    le32_to_cpu(gdp->bg_inode_table) + sbi->s_itb_per_group >=
		    block + EXT3_BLOCKS_PER_GROUP(sb))
		{
			ext3_error (sb, "ext3_check_descriptors",
				    "Inode table for group %d"
				    " not in group (block %lu)!",
				    i, (unsigned long)
					le32_to_cpu(gdp->bg_inode_table));
			return 0;
		}
		block += EXT3_BLOCKS_PER_GROUP(sb);
		gdp++;
	}
	return 1;
}


/* ext3_orphan_cleanup() walks a singly-linked list of inodes (starting at
 * the superblock) which were deleted from all directories, but held open by
 * a process at the time of a crash.  We walk the list and try to delete these
 * inodes at recovery time (only with a read-write filesystem).
 *
 * In order to keep the orphan inode chain consistent during traversal (in
 * case of crash during recovery), we link each inode into the superblock
 * orphan list_head and handle it the same way as an inode deletion during
 * normal operation (which journals the operations for us).
 *
 * We only do an iget() and an iput() on each inode, which is very safe if we
 * accidentally point at an in-use or already deleted inode.  The worst that
 * can happen in this case is that we get a "bit already cleared" message from
 * ext3_free_inode().  The only reason we would point at a wrong inode is if
 * e2fsck was run on this filesystem, and it must have already done the orphan
 * inode cleanup for us, so we can safely abort without any further action.
 */
static void ext3_orphan_cleanup (struct super_block * sb,
				 struct ext3_super_block * es)
{
	unsigned int s_flags = sb->s_flags;
	int nr_orphans = 0, nr_truncates = 0;
	if (!es->s_last_orphan) {
		jbd_debug(4, "no orphan inodes to clean up\n");
		return;
	}

	if (s_flags & MS_RDONLY) {
		printk(KERN_INFO "EXT3-fs: %s: orphan cleanup on readonly fs\n",
		       bdevname(sb->s_dev));
		sb->s_flags &= ~MS_RDONLY;
	}

	if (sb->u.ext3_sb.s_mount_state & EXT3_ERROR_FS) {
		if (es->s_last_orphan)
			jbd_debug(1, "Errors on filesystem, "
				  "clearing orphan list.\n");
		es->s_last_orphan = 0;
		jbd_debug(1, "Skipping orphan recovery on fs with errors.\n");
		return;
	}

	while (es->s_last_orphan) {
		struct inode *inode;

		if (!(inode =
		      ext3_orphan_get(sb, le32_to_cpu(es->s_last_orphan)))) {
			es->s_last_orphan = 0;
			break;
		}

		list_add(&EXT3_I(inode)->i_orphan, &EXT3_SB(sb)->s_orphan);
		if (inode->i_nlink) {
			printk(KERN_DEBUG __FUNCTION__
				": truncating inode %ld to %Ld bytes\n",
				inode->i_ino, inode->i_size);
			jbd_debug(2, "truncating inode %ld to %Ld bytes\n",
				  inode->i_ino, inode->i_size);
			ext3_truncate(inode);
			nr_truncates++;
		} else {
			printk(KERN_DEBUG __FUNCTION__
				": deleting unreferenced inode %ld\n",
				inode->i_ino);
			jbd_debug(2, "deleting unreferenced inode %ld\n",
				  inode->i_ino);
			nr_orphans++;
		}
		iput(inode);  /* The delete magic happens here! */
	}

#define PLURAL(x) (x), ((x)==1) ? "" : "s"

	if (nr_orphans)
		printk(KERN_INFO "EXT3-fs: %s: %d orphan inode%s deleted\n",
		       bdevname(sb->s_dev), PLURAL(nr_orphans));
	if (nr_truncates)
		printk(KERN_INFO "EXT3-fs: %s: %d truncate%s cleaned up\n",
		       bdevname(sb->s_dev), PLURAL(nr_truncates));
	sb->s_flags = s_flags; /* Restore MS_RDONLY status */
}

#define log2(n) ffz(~(n))

/*
 * Maximal file size.  There is a direct, and {,double-,triple-}indirect
 * block limit, and also a limit of (2^32 - 1) 512-byte sectors in i_blocks.
 * We need to be 1 filesystem block less than the 2^32 sector limit.
 */
static loff_t ext3_max_size(int bits)
{