super.c

ocfs1.4.1 oracle分布式文件系统

	bbits = le32_to_cpu(di->id2.i_super.s_blocksize_bits);
	sb->s_maxbytes = ocfs2_max_file_offset(bbits, cbits);

	osb->sb = sb;
	/* Save off for ocfs2_rw_direct */
	osb->s_sectsize_bits = blksize_bits(sector_size);
	BUG_ON(!osb->s_sectsize_bits);

	init_waitqueue_head(&osb->recovery_event);
	spin_lock_init(&osb->dc_task_lock);
	init_waitqueue_head(&osb->dc_event);
	osb->dc_work_sequence = 0;
	osb->dc_wake_sequence = 0;
	INIT_LIST_HEAD(&osb->blocked_lock_list);
	osb->blocked_lock_count = 0;
	spin_lock_init(&osb->osb_lock);
	ocfs2_init_inode_steal_slot(osb);

	atomic_set(&osb->alloc_stats.moves, 0);
	atomic_set(&osb->alloc_stats.local_data, 0);
	atomic_set(&osb->alloc_stats.bitmap_data, 0);
	atomic_set(&osb->alloc_stats.bg_allocs, 0);
	atomic_set(&osb->alloc_stats.bg_extends, 0);

	ocfs2_init_node_maps(osb);

	snprintf(osb->dev_str, sizeof(osb->dev_str), "%u,%u",
		 MAJOR(osb->sb->s_dev), MINOR(osb->sb->s_dev));

	mutex_init(&osb->recovery_lock);

	osb->disable_recovery = 0;
	osb->recovery_thread_task = NULL;

	init_waitqueue_head(&osb->checkpoint_event);
	atomic_set(&osb->needs_checkpoint, 0);

	osb->s_atime_quantum = OCFS2_DEFAULT_ATIME_QUANTUM;

	osb->node_num = O2NM_INVALID_NODE_NUM;
	osb->slot_num = OCFS2_INVALID_SLOT;

	osb->local_alloc_state = OCFS2_LA_UNUSED;
	osb->local_alloc_bh = NULL;

	ocfs2_setup_hb_callbacks(osb);

	init_waitqueue_head(&osb->osb_mount_event);

	osb->vol_label = kmalloc(OCFS2_MAX_VOL_LABEL_LEN, GFP_KERNEL);
	if (!osb->vol_label) {
		mlog(ML_ERROR, "unable to alloc vol label\n");
		status = -ENOMEM;
		goto bail;
	}

	osb->max_slots = le16_to_cpu(di->id2.i_super.s_max_slots);
	if (osb->max_slots > OCFS2_MAX_SLOTS || osb->max_slots == 0) {
		mlog(ML_ERROR, "Invalid number of node slots (%u)\n",
		     osb->max_slots);
		status = -EINVAL;
		goto bail;
	}
	mlog(0, "max_slots for this device: %u\n", osb->max_slots);

	osb->slot_recovery_generations =
		kcalloc(osb->max_slots, sizeof(*osb->slot_recovery_generations),
			GFP_KERNEL);
	if (!osb->slot_recovery_generations) {
		status = -ENOMEM;
		mlog_errno(status);
		goto bail;
	}

	init_waitqueue_head(&osb->osb_wipe_event);
	osb->osb_orphan_wipes = kcalloc(osb->max_slots,
					sizeof(*osb->osb_orphan_wipes),
					GFP_KERNEL);
	if (!osb->osb_orphan_wipes) {
		status = -ENOMEM;
		mlog_errno(status);
		goto bail;
	}

	osb->s_feature_compat =
		le32_to_cpu(OCFS2_RAW_SB(di)->s_feature_compat);
	osb->s_feature_ro_compat =
		le32_to_cpu(OCFS2_RAW_SB(di)->s_feature_ro_compat);
	osb->s_feature_incompat =
		le32_to_cpu(OCFS2_RAW_SB(di)->s_feature_incompat);

	if ((i = OCFS2_HAS_INCOMPAT_FEATURE(osb->sb, ~OCFS2_FEATURE_INCOMPAT_SUPP))) {
		mlog(ML_ERROR, "couldn't mount because of unsupported "
		     "optional features (%x).\n", i);
		status = -EINVAL;
		goto bail;
	}
	if (!(osb->sb->s_flags & MS_RDONLY) &&
	    (i = OCFS2_HAS_RO_COMPAT_FEATURE(osb->sb, ~OCFS2_FEATURE_RO_COMPAT_SUPP))) {
		mlog(ML_ERROR, "couldn't mount RDWR because of "
		     "unsupported optional features (%x).\n", i);
		status = -EINVAL;
		goto bail;
	}

	get_random_bytes(&osb->s_next_generation, sizeof(u32));

	/* FIXME
	 * This should be done in ocfs2_journal_init(), but unknown
	 * ordering issues will cause the filesystem to crash.
	 * If anyone wants to figure out what part of the code
	 * refers to osb->journal before ocfs2_journal_init() is run,
	 * be my guest.
	 */
	/* initialize our journal structure */

	journal = kzalloc(sizeof(struct ocfs2_journal), GFP_KERNEL);
	if (!journal) {
		mlog(ML_ERROR, "unable to alloc journal\n");
		status = -ENOMEM;
		goto bail;
	}
	osb->journal = journal;
	journal->j_osb = osb;

	atomic_set(&journal->j_num_trans, 0);
	init_rwsem(&journal->j_trans_barrier);
	init_waitqueue_head(&journal->j_checkpointed);
	spin_lock_init(&journal->j_lock);
	journal->j_trans_id = (unsigned long) 1;
	INIT_LIST_HEAD(&journal->j_la_cleanups);
	KAPI_INIT_WORK(&journal->j_recovery_work, ocfs2_complete_recovery, journal);
	journal->j_state = OCFS2_JOURNAL_FREE;

	/* get some pseudo constants for clustersize bits */
	osb->s_clustersize_bits =
		le32_to_cpu(di->id2.i_super.s_clustersize_bits);
	osb->s_clustersize = 1 << osb->s_clustersize_bits;
	mlog(0, "clusterbits=%d\n", osb->s_clustersize_bits);

	if (osb->s_clustersize < OCFS2_MIN_CLUSTERSIZE ||
	    osb->s_clustersize > OCFS2_MAX_CLUSTERSIZE) {
		mlog(ML_ERROR, "Volume has invalid cluster size (%d)\n",
		     osb->s_clustersize);
		status = -EINVAL;
		goto bail;
	}

	if (ocfs2_clusters_to_blocks(osb->sb, le32_to_cpu(di->i_clusters) - 1)
	    > (u32)~0UL) {
		mlog(ML_ERROR, "Volume might try to write to blocks beyond "
		     "what jbd can address in 32 bits.\n");
		status = -EINVAL;
		goto bail;
	}

	if (ocfs2_setup_osb_uuid(osb, di->id2.i_super.s_uuid,
				 sizeof(di->id2.i_super.s_uuid))) {
		mlog(ML_ERROR, "Out of memory trying to setup our uuid.\n");
		status = -ENOMEM;
		goto bail;
	}

	memcpy(&uuid_net_key, di->id2.i_super.s_uuid, sizeof(uuid_net_key));

	strncpy(osb->vol_label, di->id2.i_super.s_label, 63);
	osb->vol_label[63] = '\0';
	osb->root_blkno = le64_to_cpu(di->id2.i_super.s_root_blkno);
	osb->system_dir_blkno = le64_to_cpu(di->id2.i_super.s_system_dir_blkno);
	osb->first_cluster_group_blkno =
		le64_to_cpu(di->id2.i_super.s_first_cluster_group);
	osb->fs_generation = le32_to_cpu(di->i_fs_generation);
	mlog(0, "vol_label: %s\n", osb->vol_label);
	mlog(0, "uuid: %s\n", osb->uuid_str);
	mlog(0, "root_blkno=%llu, system_dir_blkno=%llu\n",
	     (unsigned long long)osb->root_blkno,
	     (unsigned long long)osb->system_dir_blkno);

	osb->osb_dlm_debug = ocfs2_new_dlm_debug();
	if (!osb->osb_dlm_debug) {
		status = -ENOMEM;
		mlog_errno(status);
		goto bail;
	}

	atomic_set(&osb->vol_state, VOLUME_INIT);

	/* load root, system_dir, and all global system inodes */
	status = ocfs2_init_global_system_inodes(osb);
	if (status < 0) {
		mlog_errno(status);
		goto bail;
	}

	/*
	 * global bitmap
	 */
	inode = ocfs2_get_system_file_inode(osb, GLOBAL_BITMAP_SYSTEM_INODE,
					    OCFS2_INVALID_SLOT);
	if (!inode) {
		status = -EINVAL;
		mlog_errno(status);
		goto bail;
	}

	osb->bitmap_blkno = OCFS2_I(inode)->ip_blkno;
	iput(inode);

	osb->bitmap_cpg = ocfs2_group_bitmap_size(sb) * 8;

	status = ocfs2_init_slot_info(osb);
	if (status < 0) {
		mlog_errno(status);
		goto bail;
	}

bail:
	mlog_exit(status);
	return status;
}

/*
 * will return: -EAGAIN if it is ok to keep searching for superblocks
 *              -EINVAL if there is a bad superblock
 *              0 on success
 */
static int ocfs2_verify_volume(struct ocfs2_dinode *di,
			       struct buffer_head *bh,
			       u32 blksz)
{
	int status = -EAGAIN;

	mlog_entry_void();

	if (memcmp(di->i_signature, OCFS2_SUPER_BLOCK_SIGNATURE,
		   strlen(OCFS2_SUPER_BLOCK_SIGNATURE)) == 0) {
		status = -EINVAL;
		if ((1 << le32_to_cpu(di->id2.i_super.s_blocksize_bits)) != blksz) {
			mlog(ML_ERROR, "found superblock with incorrect block "
			     "size: found %u, should be %u\n",
			     1 << le32_to_cpu(di->id2.i_super.s_blocksize_bits),
			       blksz);
		} else if (le16_to_cpu(di->id2.i_super.s_major_rev_level) !=
			   OCFS2_MAJOR_REV_LEVEL ||
			   le16_to_cpu(di->id2.i_super.s_minor_rev_level) !=
			   OCFS2_MINOR_REV_LEVEL) {
			mlog(ML_ERROR, "found superblock with bad version: "
			     "found %u.%u, should be %u.%u\n",
			     le16_to_cpu(di->id2.i_super.s_major_rev_level),
			     le16_to_cpu(di->id2.i_super.s_minor_rev_level),
			     OCFS2_MAJOR_REV_LEVEL,
			     OCFS2_MINOR_REV_LEVEL);
		} else if (bh->b_blocknr != le64_to_cpu(di->i_blkno)) {
			mlog(ML_ERROR, "bad block number on superblock: "
			     "found %llu, should be %llu\n",
			     (unsigned long long)le64_to_cpu(di->i_blkno),
			     (unsigned long long)bh->b_blocknr);
		} else if (le32_to_cpu(di->id2.i_super.s_clustersize_bits) < 12 ||
			    le32_to_cpu(di->id2.i_super.s_clustersize_bits) > 20) {
			mlog(ML_ERROR, "bad cluster size found: %u\n",
			     1 << le32_to_cpu(di->id2.i_super.s_clustersize_bits));
		} else if (!le64_to_cpu(di->id2.i_super.s_root_blkno)) {
			mlog(ML_ERROR, "bad root_blkno: 0\n");
		} else if (!le64_to_cpu(di->id2.i_super.s_system_dir_blkno)) {
			mlog(ML_ERROR, "bad system_dir_blkno: 0\n");
		} else if (le16_to_cpu(di->id2.i_super.s_max_slots) > OCFS2_MAX_SLOTS) {
			mlog(ML_ERROR,
			     "Superblock slots found greater than file system "
			     "maximum: found %u, max %u\n",
			     le16_to_cpu(di->id2.i_super.s_max_slots),
			     OCFS2_MAX_SLOTS);
		} else {
			/* found it! */
			status = 0;
		}
	}

	mlog_exit(status);
	return status;
}

static int ocfs2_check_volume(struct ocfs2_super *osb)
{
	int status;
	int dirty;
	int local;
	struct ocfs2_dinode *local_alloc = NULL; /* only used if we
						  * recover
						  * ourselves. */

	mlog_entry_void();

	/* Init our journal object. */
	status = ocfs2_journal_init(osb->journal, &dirty);
	if (status < 0) {
		mlog(ML_ERROR, "Could not initialize journal!\n");
		goto finally;
	}

	/* If the journal was unmounted cleanly then we don't want to
	 * recover anything. Otherwise, journal_load will do that
	 * dirty work for us :) */
	if (!dirty) {
		status = ocfs2_journal_wipe(osb->journal, 0);
		if (status < 0) {
			mlog_errno(status);
			goto finally;
		}
	} else {
		mlog(ML_NOTICE, "File system was not unmounted cleanly, "
		     "recovering volume.\n");
	}

	local = ocfs2_mount_local(osb);

	/* will play back anything left in the journal. */
	status = ocfs2_journal_load(osb->journal, local, dirty);
	if (status < 0) {
		mlog(ML_ERROR, "ocfs2 journal load failed! %d\n", status);
		goto finally;
	}

	if (dirty) {
		/* recover my local alloc if we didn't unmount cleanly. */
		status = ocfs2_begin_local_alloc_recovery(osb,
							  osb->slot_num,
							  &local_alloc);
		if (status < 0) {
			mlog_errno(status);
			goto finally;
		}
		/* we complete the recovery process after we've marked
		 * ourselves as mounted. */
	}

	mlog(0, "Journal loaded.\n");

	status = ocfs2_load_local_alloc(osb);
	if (status < 0) {
		mlog_errno(status);
		goto finally;
	}

	if (dirty) {
		/* Recovery will be completed after we've mounted the
		 * rest of the volume. */
		osb->dirty = 1;
		osb->local_alloc_copy = local_alloc;
		local_alloc = NULL;
	}

	/* go through each journal, trylock it and if you get the
	 * lock, and it's marked as dirty, set the bit in the recover
	 * map and launch a recovery thread for it. */
	status = ocfs2_mark_dead_nodes(osb);
	if (status < 0)
		mlog_errno(status);

finally:
	if (local_alloc)
		kfree(local_alloc);

	mlog_exit(status);
	return status;
}

/*
 * The routine gets called from dismount or close whenever a dismount on
 * volume is requested and the osb open count becomes 1.
 * It will remove the osb from the global list and also free up all the
 * initialized resources and fileobject.
 */
static void ocfs2_delete_osb(struct ocfs2_super *osb)
{
	mlog_entry_void();

	/* This function assumes that the caller has the main osb resource */

	if (osb->slot_info)
		ocfs2_free_slot_info(osb->slot_info);

	kfree(osb->osb_orphan_wipes);
	kfree(osb->slot_recovery_generations);
	/* FIXME
	 * This belongs in journal shutdown, but because we have to
	 * allocate osb->journal at the start of ocfs2_initalize_osb(),
	 * we free it here.
	 */
	kfree(osb->journal);
	if (osb->local_alloc_copy)
		kfree(osb->local_alloc_copy);
	kfree(osb->uuid_str);
	ocfs2_put_dlm_debug(osb->osb_dlm_debug);
	memset(osb, 0, sizeof(struct ocfs2_super));

	mlog_exit_void();
}

/* Put OCFS2 into a readonly state, or (if the user specifies it),
 * panic(). We do not support continue-on-error operation. */
static void ocfs2_handle_error(struct super_block *sb)
{
	struct ocfs2_super *osb = OCFS2_SB(sb);

	if (osb->s_mount_opt & OCFS2_MOUNT_ERRORS_PANIC)
		panic("OCFS2: (device %s): panic forced after error\n",
		      sb->s_id);

	ocfs2_set_osb_flag(osb, OCFS2_OSB_ERROR_FS);

	if (sb->s_flags & MS_RDONLY &&
	    (ocfs2_is_soft_readonly(osb) ||
	     ocfs2_is_hard_readonly(osb)))
		return;

	printk(KERN_CRIT "File system is now read-only due to the potential "
	       "of on-disk corruption. Please run fsck.ocfs2 once the file "
	       "system is unmounted.\n");
	sb->s_flags |= MS_RDONLY;
	ocfs2_set_ro_flag(osb, 0);
}

static char error_buf[1024];

void __ocfs2_error(struct super_block *sb,
		   const char *function,
		   const char *fmt, ...)
{
	va_list args;

	va_start(args, fmt);
	vsnprintf(error_buf, sizeof(error_buf), fmt, args);
	va_end(args);

	/* Not using mlog here because we want to show the actual
	 * function the error came from. */
	printk(KERN_CRIT "OCFS2: ERROR (device %s): %s: %s\n",
	       sb->s_id, function, error_buf);

	ocfs2_handle_error(sb);
}

/* Handle critical errors. This is intentionally more drastic than
 * ocfs2_handle_error, so we only use for things like journal errors,
 * etc. */
void __ocfs2_abort(struct super_block* sb,
		   const char *function,
		   const char *fmt, ...)
{
	va_list args;

	va_start(args, fmt);
	vsnprintf(error_buf, sizeof(error_buf), fmt, args);
	va_end(args);

	printk(KERN_CRIT "OCFS2: abort (device %s): %s: %s\n",
	       sb->s_id, function, error_buf);

	/* We don't have the cluster support yet to go straight to
	 * hard readonly in here. Until then, we want to keep
	 * ocfs2_abort() so that we can at least mark critical
	 * errors.
	 *
	 * TODO: This should abort the journal and alert other nodes
	 * that our slot needs recovery. */

	/* Force a panic(). This stinks, but it's better than letting
	 * things continue without having a proper hard readonly
	 * here. */
	OCFS2_SB(sb)->s_mount_opt |= OCFS2_MOUNT_ERRORS_PANIC;
	ocfs2_handle_error(sb);
}

module_init(ocfs2_init);
module_exit(ocfs2_exit);