volumes.c

// SPDX-License-Identifier: GPL-2.0
/*
 * Copyright (C) 2007 Oracle.  All rights reserved.
 */

#include <linux/sched.h>
#include <linux/bio.h>
#include <linux/slab.h>
#include <linux/buffer_head.h>
#include <linux/blkdev.h>
#include <linux/ratelimit.h>
#include <linux/kthread.h>
#include <linux/raid/pq.h>
#include <linux/semaphore.h>
#include <linux/uuid.h>
#include <linux/list_sort.h>
#include "misc.h"
#include "ctree.h"
#include "extent_map.h"
#include "disk-io.h"
#include "transaction.h"
#include "print-tree.h"
#include "volumes.h"
#include "raid56.h"
#include "async-thread.h"
#include "check-integrity.h"
#include "rcu-string.h"
#include "dev-replace.h"
#include "sysfs.h"
#include "tree-checker.h"
#include "space-info.h"
#include "block-group.h"

const struct btrfs_raid_attr btrfs_raid_array[BTRFS_NR_RAID_TYPES] = {
	[BTRFS_RAID_RAID10] = {
		.sub_stripes	= 2,
		.dev_stripes	= 1,
		.devs_max	= 0,	/* 0 == as many as possible */
		.devs_min	= 4,
		.tolerated_failures = 1,
		.devs_increment	= 2,
		.ncopies	= 2,
		.nparity        = 0,
		.raid_name	= "raid10",
		.bg_flag	= BTRFS_BLOCK_GROUP_RAID10,
		.mindev_error	= BTRFS_ERROR_DEV_RAID10_MIN_NOT_MET,
	},
	[BTRFS_RAID_RAID1] = {
		.sub_stripes	= 1,
		.dev_stripes	= 1,
		.devs_max	= 2,
		.devs_min	= 2,
		.tolerated_failures = 1,
		.devs_increment	= 2,
		.ncopies	= 2,
		.nparity        = 0,
		.raid_name	= "raid1",
		.bg_flag	= BTRFS_BLOCK_GROUP_RAID1,
		.mindev_error	= BTRFS_ERROR_DEV_RAID1_MIN_NOT_MET,
	},
	[BTRFS_RAID_RAID1C3] = {
		.sub_stripes	= 1,
		.dev_stripes	= 1,
		.devs_max	= 0,
		.devs_min	= 3,
		.tolerated_failures = 2,
		.devs_increment	= 3,
		.ncopies	= 3,
		.raid_name	= "raid1c3",
		.bg_flag	= BTRFS_BLOCK_GROUP_RAID1C3,
		.mindev_error	= BTRFS_ERROR_DEV_RAID1C3_MIN_NOT_MET,
	},
	[BTRFS_RAID_DUP] = {
		.sub_stripes	= 1,
		.dev_stripes	= 2,
		.devs_max	= 1,
		.devs_min	= 1,
		.tolerated_failures = 0,
		.devs_increment	= 1,
		.ncopies	= 2,
		.nparity        = 0,
		.raid_name	= "dup",
		.bg_flag	= BTRFS_BLOCK_GROUP_DUP,
		.mindev_error	= 0,
	},
	[BTRFS_RAID_RAID0] = {
		.sub_stripes	= 1,
		.dev_stripes	= 1,
		.devs_max	= 0,
		.devs_min	= 2,
		.tolerated_failures = 0,
		.devs_increment	= 1,
		.ncopies	= 1,
		.nparity        = 0,
		.raid_name	= "raid0",
		.bg_flag	= BTRFS_BLOCK_GROUP_RAID0,
		.mindev_error	= 0,
	},
	[BTRFS_RAID_SINGLE] = {
		.sub_stripes	= 1,
		.dev_stripes	= 1,
		.devs_max	= 1,
		.devs_min	= 1,
		.tolerated_failures = 0,
		.devs_increment	= 1,
		.ncopies	= 1,
		.nparity        = 0,
		.raid_name	= "single",
		.bg_flag	= 0,
		.mindev_error	= 0,
	},
	[BTRFS_RAID_RAID5] = {
		.sub_stripes	= 1,
		.dev_stripes	= 1,
		.devs_max	= 0,
		.devs_min	= 2,
		.tolerated_failures = 1,
		.devs_increment	= 1,
		.ncopies	= 1,
		.nparity        = 1,
		.raid_name	= "raid5",
		.bg_flag	= BTRFS_BLOCK_GROUP_RAID5,
		.mindev_error	= BTRFS_ERROR_DEV_RAID5_MIN_NOT_MET,
	},
	[BTRFS_RAID_RAID6] = {
		.sub_stripes	= 1,
		.dev_stripes	= 1,
		.devs_max	= 0,
		.devs_min	= 3,
		.tolerated_failures = 2,
		.devs_increment	= 1,
		.ncopies	= 1,
		.nparity        = 2,
		.raid_name	= "raid6",
		.bg_flag	= BTRFS_BLOCK_GROUP_RAID6,
		.mindev_error	= BTRFS_ERROR_DEV_RAID6_MIN_NOT_MET,
	},
};

const char *btrfs_bg_type_to_raid_name(u64 flags)
{
	const int index = btrfs_bg_flags_to_raid_index(flags);

	if (index >= BTRFS_NR_RAID_TYPES)
		return NULL;

	return btrfs_raid_array[index].raid_name;
}

/*
 * Fill @buf with textual description of @bg_flags, no more than @size_buf
 * bytes including terminating null byte.
 */
void btrfs_describe_block_groups(u64 bg_flags, char *buf, u32 size_buf)
{
	int i;
	int ret;
	char *bp = buf;
	u64 flags = bg_flags;
	u32 size_bp = size_buf;

	if (!flags) {
		strcpy(bp, "NONE");
		return;
	}

#define DESCRIBE_FLAG(flag, desc)						\
	do {								\
		if (flags & (flag)) {					\
			ret = snprintf(bp, size_bp, "%s|", (desc));	\
			if (ret < 0 || ret >= size_bp)			\
				goto out_overflow;			\
			size_bp -= ret;					\
			bp += ret;					\
			flags &= ~(flag);				\
		}							\
	} while (0)

	DESCRIBE_FLAG(BTRFS_BLOCK_GROUP_DATA, "data");
	DESCRIBE_FLAG(BTRFS_BLOCK_GROUP_SYSTEM, "system");
	DESCRIBE_FLAG(BTRFS_BLOCK_GROUP_METADATA, "metadata");

	DESCRIBE_FLAG(BTRFS_AVAIL_ALLOC_BIT_SINGLE, "single");
	for (i = 0; i < BTRFS_NR_RAID_TYPES; i++)
		DESCRIBE_FLAG(btrfs_raid_array[i].bg_flag,
			      btrfs_raid_array[i].raid_name);
#undef DESCRIBE_FLAG

	if (flags) {
		ret = snprintf(bp, size_bp, "0x%llx|", flags);
		size_bp -= ret;
	}

	if (size_bp < size_buf)
		buf[size_buf - size_bp - 1] = '\0'; /* remove last | */

	/*
	 * The text is trimmed, it's up to the caller to provide sufficiently
	 * large buffer
	 */
out_overflow:;
}

static int init_first_rw_device(struct btrfs_trans_handle *trans);
static int btrfs_relocate_sys_chunks(struct btrfs_fs_info *fs_info);
static void btrfs_dev_stat_print_on_error(struct btrfs_device *dev);
static void btrfs_dev_stat_print_on_load(struct btrfs_device *device);
static int __btrfs_map_block(struct btrfs_fs_info *fs_info,
			     enum btrfs_map_op op,
			     u64 logical, u64 *length,
			     struct btrfs_bio **bbio_ret,
			     int mirror_num, int need_raid_map);

/*
 * Device locking
 * ==============
 *
 * There are several mutexes that protect manipulation of devices and low-level
 * structures like chunks but not block groups, extents or files
 *
 * uuid_mutex (global lock)
 * ------------------------
 * protects the fs_uuids list that tracks all per-fs fs_devices, resulting from
 * the SCAN_DEV ioctl registration or from mount either implicitly (the first
 * device) or requested by the device= mount option
 *
 * the mutex can be very coarse and can cover long-running operations
 *
 * protects: updates to fs_devices counters like missing devices, rw devices,
 * seeding, structure cloning, opening/closing devices at mount/umount time
 *
 * global::fs_devs - add, remove, updates to the global list
 *
 * does not protect: manipulation of the fs_devices::devices list!
 *
 * btrfs_device::name - renames (write side), read is RCU
 *
 * fs_devices::device_list_mutex (per-fs, with RCU)
 * ------------------------------------------------
 * protects updates to fs_devices::devices, ie. adding and deleting
 *
 * simple list traversal with read-only actions can be done with RCU protection
 *
 * may be used to exclude some operations from running concurrently without any
 * modifications to the list (see write_all_supers)
 *
 * balance_mutex
 * -------------
 * protects balance structures (status, state) and context accessed from
 * several places (internally, ioctl)
 *
 * chunk_mutex
 * -----------
 * protects chunks, adding or removing during allocation, trim or when a new
 * device is added/removed. Additionally it also protects post_commit_list of
 * individual devices, since they can be added to the transaction's
 * post_commit_list only with chunk_mutex held.
 *
 * cleaner_mutex
 * -------------
 * a big lock that is held by the cleaner thread and prevents running subvolume
 * cleaning together with relocation or delayed iputs
 *
 *
 * Lock nesting
 * ============
 *
 * uuid_mutex
 *   volume_mutex
 *     device_list_mutex
 *       chunk_mutex
 *     balance_mutex
 *
 *
 * Exclusive operations, BTRFS_FS_EXCL_OP
 * ======================================
 *
 * Maintains the exclusivity of the following operations that apply to the
 * whole filesystem and cannot run in parallel.
 *
 * - Balance (*)
 * - Device add
 * - Device remove
 * - Device replace (*)
 * - Resize
 *
 * The device operations (as above) can be in one of the following states:
 *
 * - Running state
 * - Paused state
 * - Completed state
 *
 * Only device operations marked with (*) can go into the Paused state for the
 * following reasons:
 *
 * - ioctl (only Balance can be Paused through ioctl)
 * - filesystem remounted as read-only
 * - filesystem unmounted and mounted as read-only
 * - system power-cycle and filesystem mounted as read-only
 * - filesystem or device errors leading to forced read-only
 *
 * BTRFS_FS_EXCL_OP flag is set and cleared using atomic operations.
 * During the course of Paused state, the BTRFS_FS_EXCL_OP remains set.
 * A device operation in Paused or Running state can be canceled or resumed
 * either by ioctl (Balance only) or when remounted as read-write.
 * BTRFS_FS_EXCL_OP flag is cleared when the device operation is canceled or
 * completed.
 */

DEFINE_MUTEX(uuid_mutex);
static LIST_HEAD(fs_uuids);
struct list_head * __attribute_const__ btrfs_get_fs_uuids(void)
{
	return &fs_uuids;
}

/*
 * alloc_fs_devices - allocate struct btrfs_fs_devices
 * @fsid:		if not NULL, copy the UUID to fs_devices::fsid
 * @metadata_fsid:	if not NULL, copy the UUID to fs_devices::metadata_fsid
 *
 * Return a pointer to a new struct btrfs_fs_devices on success, or ERR_PTR().
 * The returned struct is not linked onto any lists and can be destroyed with
 * kfree() right away.
 */
static struct btrfs_fs_devices *alloc_fs_devices(const u8 *fsid,
						 const u8 *metadata_fsid)
{
	struct btrfs_fs_devices *fs_devs;

	fs_devs = kzalloc(sizeof(*fs_devs), GFP_KERNEL);
	if (!fs_devs)
		return ERR_PTR(-ENOMEM);

	mutex_init(&fs_devs->device_list_mutex);

	INIT_LIST_HEAD(&fs_devs->devices);
	INIT_LIST_HEAD(&fs_devs->alloc_list);
	INIT_LIST_HEAD(&fs_devs->fs_list);
	if (fsid)
		memcpy(fs_devs->fsid, fsid, BTRFS_FSID_SIZE);

	if (metadata_fsid)
		memcpy(fs_devs->metadata_uuid, metadata_fsid, BTRFS_FSID_SIZE);
	else if (fsid)
		memcpy(fs_devs->metadata_uuid, fsid, BTRFS_FSID_SIZE);

	return fs_devs;
}

void btrfs_free_device(struct btrfs_device *device)
{
	WARN_ON(!list_empty(&device->post_commit_list));
	rcu_string_free(device->name);
	extent_io_tree_release(&device->alloc_state);
	bio_put(device->flush_bio);
	kfree(device);
}

static void free_fs_devices(struct btrfs_fs_devices *fs_devices)
{
	struct btrfs_device *device;
	WARN_ON(fs_devices->opened);
	while (!list_empty(&fs_devices->devices)) {
		device = list_entry(fs_devices->devices.next,
				    struct btrfs_device, dev_list);
		list_del(&device->dev_list);
		btrfs_free_device(device);
	}
	kfree(fs_devices);
}

void __exit btrfs_cleanup_fs_uuids(void)
{
	struct btrfs_fs_devices *fs_devices;

	while (!list_empty(&fs_uuids)) {
		fs_devices = list_entry(fs_uuids.next,
					struct btrfs_fs_devices, fs_list);
		list_del(&fs_devices->fs_list);
		free_fs_devices(fs_devices);
	}
}

/*
 * Returns a pointer to a new btrfs_device on success; ERR_PTR() on error.
 * Returned struct is not linked onto any lists and must be destroyed using
 * btrfs_free_device.
 */
static struct btrfs_device *__alloc_device(void)
{
	struct btrfs_device *dev;

	dev = kzalloc(sizeof(*dev), GFP_KERNEL);
	if (!dev)
		return ERR_PTR(-ENOMEM);

	/*
	 * Preallocate a bio that's always going to be used for flushing device
	 * barriers and matches the device lifespan
	 */
	dev->flush_bio = bio_alloc_bioset(GFP_KERNEL, 0, NULL);
	if (!dev->flush_bio) {
		kfree(dev);
		return ERR_PTR(-ENOMEM);
	}

	INIT_LIST_HEAD(&dev->dev_list);
	INIT_LIST_HEAD(&dev->dev_alloc_list);
	INIT_LIST_HEAD(&dev->post_commit_list);

	atomic_set(&dev->reada_in_flight, 0);
	atomic_set(&dev->dev_stats_ccnt, 0);
	btrfs_device_data_ordered_init(dev);
	INIT_RADIX_TREE(&dev->reada_zones, GFP_NOFS & ~__GFP_DIRECT_RECLAIM);
	INIT_RADIX_TREE(&dev->reada_extents, GFP_NOFS & ~__GFP_DIRECT_RECLAIM);
	extent_io_tree_init(NULL, &dev->alloc_state, 0, NULL);

	return dev;
}

static noinline struct btrfs_fs_devices *find_fsid(
		const u8 *fsid, const u8 *metadata_fsid)
{
	struct btrfs_fs_devices *fs_devices;

	ASSERT(fsid);

	if (metadata_fsid) {
		/*
		 * Handle scanned device having completed its fsid change but
		 * belonging to a fs_devices that was created by first scanning
		 * a device which didn't have its fsid/metadata_uuid changed
		 * at all and the CHANGING_FSID_V2 flag set.
		 */
		list_for_each_entry(fs_devices, &fs_uuids, fs_list) {
			if (fs_devices->fsid_change &&
			    memcmp(metadata_fsid, fs_devices->fsid,
				   BTRFS_FSID_SIZE) == 0 &&
			    memcmp(fs_devices->fsid, fs_devices->metadata_uuid,
				   BTRFS_FSID_SIZE) == 0) {
				return fs_devices;
			}
		}
		/*
		 * Handle scanned device having completed its fsid change but
		 * belonging to a fs_devices that was created by a device that
		 * has an outdated pair of fsid/metadata_uuid and
		 * CHANGING_FSID_V2 flag set.
		 */
		list_for_each_entry(fs_devices, &fs_uuids, fs_list) {
			if (fs_devices->fsid_change &&
			    memcmp(fs_devices->metadata_uuid,
				   fs_devices->fsid, BTRFS_FSID_SIZE) != 0 &&
			    memcmp(metadata_fsid, fs_devices->metadata_uuid,
				   BTRFS_FSID_SIZE) == 0) {
				return fs_devices;
			}
		}
	}

	/* Handle non-split brain cases */
	list_for_each_entry(fs_devices, &fs_uuids, fs_list) {
		if (metadata_fsid) {
			if (memcmp(fsid, fs_devices->fsid, BTRFS_FSID_SIZE) == 0
			    && memcmp(metadata_fsid, fs_devices->metadata_uuid,
				      BTRFS_FSID_SIZE) == 0)
				return fs_devices;
		} else {
			if (memcmp(fsid, fs_devices->fsid, BTRFS_FSID_SIZE) == 0)
				return fs_devices;
		}
	}
	return NULL;
}

static int
btrfs_get_bdev_and_sb(const char *device_path, fmode_t flags, void *holder,
		      int flush, struct block_device **bdev,
		      struct buffer_head **bh)
{
	int ret;

	*bdev = blkdev_get_by_path(device_path, flags, holder);

	if (IS_ERR(*bdev)) {
		ret = PTR_ERR(*bdev);
		goto error;
	}

	if (flush)
		filemap_write_and_wait((*bdev)->bd_inode->i_mapping);
	ret = set_blocksize(*bdev, BTRFS_BDEV_BLOCKSIZE);
	if (ret) {
		blkdev_put(*bdev, flags);
		goto error;
	}
	invalidate_bdev(*bdev);
	*bh = btrfs_read_dev_super(*bdev);
	if (IS_ERR(*bh)) {
		ret = PTR_ERR(*bh);
		blkdev_put(*bdev, flags);
		goto error;
	}

	return 0;

error:
	*bdev = NULL;
	*bh = NULL;
	return ret;
}

static bool device_path_matched(const char *path, struct btrfs_device *device)
{
	int found;

	rcu_read_lock();
	found = strcmp(rcu_str_deref(device->name), path);
	rcu_read_unlock();

	return found == 0;
}

/*
 *  Search and remove all stale (devices which are not mounted) devices.
 *  When both inputs are NULL, it will search and release all stale devices.
 *  path:	Optional. When provided will it release all unmounted devices
 *		matching this path only.
 *  skip_dev:	Optional. Will skip this device when searching for the stale
 *		devices.
 *  Return:	0 for success or if @path is NULL.
 * 		-EBUSY if @path is a mounted device.
 * 		-ENOENT if @path does not match any device in the list.
 */
static int btrfs_free_stale_devices(const char *path,
				     struct btrfs_device *skip_device)
{
	struct btrfs_fs_devices *fs_devices, *tmp_fs_devices;
	struct btrfs_device *device, *tmp_device;
	int ret = 0;

	if (path)
		ret = -ENOENT;

	list_for_each_entry_safe(fs_devices, tmp_fs_devices, &fs_uuids, fs_list) {

		mutex_lock(&fs_devices->device_list_mutex);
		list_for_each_entry_safe(device, tmp_device,
					 &fs_devices->devices, dev_list) {
			if (skip_device && skip_device == device)
				continue;
			if (path && !device->name)
				continue;
			if (path && !device_path_matched(path, device))
				continue;
			if (fs_devices->opened) {
				/* for an already deleted device return 0 */
				if (path && ret != 0)
					ret = -EBUSY;
				break;
			}

			/* delete the stale device */
			fs_devices->num_devices--;
			list_del(&device->dev_list);
			btrfs_free_device(device);

			ret = 0;
			if (fs_devices->num_devices == 0)
				break;
		}
		mutex_unlock(&fs_devices->device_list_mutex);

		if (fs_devices->num_devices == 0) {
			btrfs_sysfs_remove_fsid(fs_devices);
			list_del(&fs_devices->fs_list);
			free_fs_devices(fs_devices);
		}
	}

	return ret;
}

static int btrfs_open_one_device(struct btrfs_fs_devices *fs_devices,
			struct btrfs_device *device, fmode_t flags,
			void *holder)
{
	struct request_queue *q;
	struct block_device *bdev;
	struct buffer_head *bh;
	struct btrfs_super_block *disk_super;
	u64 devid;
	int ret;

	if (device->bdev)
		return -EINVAL;
	if (!device->name)
		return -EINVAL;

	ret = btrfs_get_bdev_and_sb(device->name->str, flags, holder, 1,
				    &bdev, &bh);
	if (ret)
		return ret;

	disk_super = (struct btrfs_super_block *)bh->b_data;
	devid = btrfs_stack_device_id(&disk_super->dev_item);
	if (devid != device->devid)
		goto error_brelse;

	if (memcmp(device->uuid, disk_super->dev_item.uuid, BTRFS_UUID_SIZE))
		goto error_brelse;

	device->generation = btrfs_super_generation(disk_super);

	if (btrfs_super_flags(disk_super) & BTRFS_SUPER_FLAG_SEEDING) {
		if (btrfs_super_incompat_flags(disk_super) &
		    BTRFS_FEATURE_INCOMPAT_METADATA_UUID) {
			pr_err(
		"BTRFS: Invalid seeding and uuid-changed device detected\n");
			goto error_brelse;
		}

		clear_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state);
		fs_devices->seeding = 1;
	} else {
		if (bdev_read_only(bdev))
			clear_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state);
		else
			set_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state);
	}

	q = bdev_get_queue(bdev);
	if (!blk_queue_nonrot(q))
		fs_devices->rotating = 1;

	device->bdev = bdev;
	clear_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &device->dev_state);
	device->mode = flags;

	fs_devices->open_devices++;
	if (test_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state) &&
	    device->devid != BTRFS_DEV_REPLACE_DEVID) {
		fs_devices->rw_devices++;
		list_add_tail(&device->dev_alloc_list, &fs_devices->alloc_list);
	}
	brelse(bh);

	return 0;

error_brelse:
	brelse(bh);
	blkdev_put(bdev, flags);

	return -EINVAL;
}

/*
 * Handle scanned device having its CHANGING_FSID_V2 flag set and the fs_devices
 * being created with a disk that has already completed its fsid change.
 */
static struct btrfs_fs_devices *find_fsid_inprogress(
					struct btrfs_super_block *disk_super)
{
	struct btrfs_fs_devices *fs_devices;

	list_for_each_entry(fs_devices, &fs_uuids, fs_list) {
		if (memcmp(fs_devices->metadata_uuid, fs_devices->fsid,
			   BTRFS_FSID_SIZE) != 0 &&
		    memcmp(fs_devices->metadata_uuid, disk_super->fsid,
			   BTRFS_FSID_SIZE) == 0 && !fs_devices->fsid_change) {
			return fs_devices;
		}
	}

	return NULL;
}


static struct btrfs_fs_devices *find_fsid_changed(
					struct btrfs_super_block *disk_super)
{
	struct btrfs_fs_devices *fs_devices;

	/*
	 * Handles the case where scanned device is part of an fs that had
	 * multiple successful changes of FSID but curently device didn't
	 * observe it. Meaning our fsid will be different than theirs.
	 */
	list_for_each_entry(fs_devices, &fs_uuids, fs_list) {
		if (memcmp(fs_devices->metadata_uuid, fs_devices->fsid,
			   BTRFS_FSID_SIZE) != 0 &&
		    memcmp(fs_devices->metadata_uuid, disk_super->metadata_uuid,
			   BTRFS_FSID_SIZE) == 0 &&
		    memcmp(fs_devices->fsid, disk_super->fsid,
			   BTRFS_FSID_SIZE) != 0) {
			return fs_devices;
		}
	}

	return NULL;
}
/*
 * Add new device to list of registered devices
 *
 * Returns:
 * device pointer which was just added or updated when successful
 * error pointer when failed
 */
static noinline struct btrfs_device *device_list_add(const char *path,
			   struct btrfs_super_block *disk_super,
			   bool *new_device_added)
{
	struct btrfs_device *device;
	struct btrfs_fs_devices *fs_devices = NULL;
	struct rcu_string *name;
	u64 found_transid = btrfs_super_generation(disk_super);
	u64 devid = btrfs_stack_device_id(&disk_super->dev_item);
	bool has_metadata_uuid = (btrfs_super_incompat_flags(disk_super) &
		BTRFS_FEATURE_INCOMPAT_METADATA_UUID);
	bool fsid_change_in_progress = (btrfs_super_flags(disk_super) &
					BTRFS_SUPER_FLAG_CHANGING_FSID_V2);

	if (fsid_change_in_progress) {
		if (!has_metadata_uuid) {
			/*
			 * When we have an image which has CHANGING_FSID_V2 set
			 * it might belong to either a filesystem which has
			 * disks with completed fsid change or it might belong
			 * to fs with no UUID changes in effect, handle both.
			 */
			fs_devices = find_fsid_inprogress(disk_super);
			if (!fs_devices)
				fs_devices = find_fsid(disk_super->fsid, NULL);
		} else {
			fs_devices = find_fsid_changed(disk_super);
		}
	} else if (has_metadata_uuid) {
		fs_devices = find_fsid(disk_super->fsid,
				       disk_super->metadata_uuid);
	} else {
		fs_devices = find_fsid(disk_super->fsid, NULL);
	}


	if (!fs_devices) {
		if (has_metadata_uuid)
			fs_devices = alloc_fs_devices(disk_super->fsid,
						      disk_super->metadata_uuid);
		else
			fs_devices = alloc_fs_devices(disk_super->fsid, NULL);

		if (IS_ERR(fs_devices))
			return ERR_CAST(fs_devices);

		fs_devices->fsid_change = fsid_change_in_progress;

		mutex_lock(&fs_devices->device_list_mutex);
		list_add(&fs_devices->fs_list, &fs_uuids);

		device = NULL;
	} else {
		mutex_lock(&fs_devices->device_list_mutex);
		device = btrfs_find_device(fs_devices, devid,
				disk_super->dev_item.uuid, NULL, false);

		/*
		 * If this disk has been pulled into an fs devices created by
		 * a device which had the CHANGING_FSID_V2 flag then replace the
		 * metadata_uuid/fsid values of the fs_devices.
		 */
		if (has_metadata_uuid && fs_devices->fsid_change &&
		    found_transid > fs_devices->latest_generation) {
			memcpy(fs_devices->fsid, disk_super->fsid,
					BTRFS_FSID_SIZE);
			memcpy(fs_devices->metadata_uuid,
					disk_super->metadata_uuid, BTRFS_FSID_SIZE);

			fs_devices->fsid_change = false;
		}
	}

	if (!device) {
		if (fs_devices->opened) {
			mutex_unlock(&fs_devices->device_list_mutex);
			return ERR_PTR(-EBUSY);
		}

		device = btrfs_alloc_device(NULL, &devid,
					    disk_super->dev_item.uuid);
		if (IS_ERR(device)) {
			mutex_unlock(&fs_devices->device_list_mutex);
			/* we can safely leave the fs_devices entry around */
			return device;
		}

		name = rcu_string_strdup(path, GFP_NOFS);
		if (!name) {
			btrfs_free_device(device);
			mutex_unlock(&fs_devices->device_list_mutex);
			return ERR_PTR(-ENOMEM);
		}
		rcu_assign_pointer(device->name, name);

		list_add_rcu(&device->dev_list, &fs_devices->devices);
		fs_devices->num_devices++;

		device->fs_devices = fs_devices;
		*new_device_added = true;

		if (disk_super->label[0])
			pr_info(
	"BTRFS: device label %s devid %llu transid %llu %s scanned by %s (%d)\n",
				disk_super->label, devid, found_transid, path,
				current->comm, task_pid_nr(current));
		else
			pr_info(
	"BTRFS: device fsid %pU devid %llu transid %llu %s scanned by %s (%d)\n",
				disk_super->fsid, devid, found_transid, path,
				current->comm, task_pid_nr(current));

	} else if (!device->name || strcmp(device->name->str, path)) {
		/*
		 * When FS is already mounted.
		 * 1. If you are here and if the device->name is NULL that
		 *    means this device was missing at time of FS mount.
		 * 2. If you are here and if the device->name is different
		 *    from 'path' that means either
		 *      a. The same device disappeared and reappeared with
		 *         different name. or
		 *      b. The missing-disk-which-was-replaced, has
		 *         reappeared now.
		 *
		 * We must allow 1 and 2a above. But 2b would be a spurious
		 * and unintentional.
		 *
		 * Further in case of 1 and 2a above, the disk at 'path'
		 * would have missed some transaction when it was away and
		 * in case of 2a the stale bdev has to be updated as well.
		 * 2b must not be allowed at all time.
		 */

		/*
		 * For now, we do allow update to btrfs_fs_device through the
		 * btrfs dev scan cli after FS has been mounted.  We're still
		 * tracking a problem where systems fail mount by subvolume id
		 * when we reject replacement on a mounted FS.
		 */
		if (!fs_devices->opened && found_transid < device->generation) {
			/*
			 * That is if the FS is _not_ mounted and if you
			 * are here, that means there is more than one
			 * disk with same uuid and devid.We keep the one
			 * with larger generation number or the last-in if
			 * generation are equal.
			 */
			mutex_unlock(&fs_devices->device_list_mutex);
			return ERR_PTR(-EEXIST);
		}

		/*
		 * We are going to replace the device path for a given devid,
		 * make sure it's the same device if the device is mounted
		 */
		if (device->bdev) {
			struct block_device *path_bdev;

			path_bdev = lookup_bdev(path);
			if (IS_ERR(path_bdev)) {
				mutex_unlock(&fs_devices->device_list_mutex);
				return ERR_CAST(path_bdev);
			}

			if (device->bdev != path_bdev) {
				bdput(path_bdev);
				mutex_unlock(&fs_devices->device_list_mutex);
				btrfs_warn_in_rcu(device->fs_info,
			"duplicate device fsid:devid for %pU:%llu old:%s new:%s",
					disk_super->fsid, devid,
					rcu_str_deref(device->name), path);
				return ERR_PTR(-EEXIST);
			}
			bdput(path_bdev);
			btrfs_info_in_rcu(device->fs_info,
				"device fsid %pU devid %llu moved old:%s new:%s",
				disk_super->fsid, devid,
				rcu_str_deref(device->name), path);
		}

		name = rcu_string_strdup(path, GFP_NOFS);
		if (!name) {
			mutex_unlock(&fs_devices->device_list_mutex);
			return ERR_PTR(-ENOMEM);
		}
		rcu_string_free(device->name);
		rcu_assign_pointer(device->name, name);
		if (test_bit(BTRFS_DEV_STATE_MISSING, &device->dev_state)) {
			fs_devices->missing_devices--;
			clear_bit(BTRFS_DEV_STATE_MISSING, &device->dev_state);
		}
	}

	/*
	 * Unmount does not free the btrfs_device struct but would zero
	 * generation along with most of the other members. So just update
	 * it back. We need it to pick the disk with largest generation
	 * (as above).
	 */
	if (!fs_devices->opened) {
		device->generation = found_transid;
		fs_devices->latest_generation = max_t(u64, found_transid,
						fs_devices->latest_generation);
	}

	fs_devices->total_devices = btrfs_super_num_devices(disk_super);

	mutex_unlock(&fs_devices->device_list_mutex);
	return device;
}

static struct btrfs_fs_devices *clone_fs_devices(struct btrfs_fs_devices *orig)
{
	struct btrfs_fs_devices *fs_devices;
	struct btrfs_device *device;
	struct btrfs_device *orig_dev;
	int ret = 0;

	fs_devices = alloc_fs_devices(orig->fsid, NULL);
	if (IS_ERR(fs_devices))
		return fs_devices;

	mutex_lock(&orig->device_list_mutex);
	fs_devices->total_devices = orig->total_devices;

	list_for_each_entry(orig_dev, &orig->devices, dev_list) {
		struct rcu_string *name;

		device = btrfs_alloc_device(NULL, &orig_dev->devid,
					    orig_dev->uuid);
		if (IS_ERR(device)) {
			ret = PTR_ERR(device);
			goto error;
		}

		/*
		 * This is ok to do without rcu read locked because we hold the
		 * uuid mutex so nothing we touch in here is going to disappear.
		 */
		if (orig_dev->name) {
			name = rcu_string_strdup(orig_dev->name->str,
					GFP_KERNEL);
			if (!name) {
				btrfs_free_device(device);
				ret = -ENOMEM;
				goto error;
			}
			rcu_assign_pointer(device->name, name);
		}

		list_add(&device->dev_list, &fs_devices->devices);
		device->fs_devices = fs_devices;
		fs_devices->num_devices++;
	}
	mutex_unlock(&orig->device_list_mutex);
	return fs_devices;
error:
	mutex_unlock(&orig->device_list_mutex);
	free_fs_devices(fs_devices);
	return ERR_PTR(ret);
}

/*
 * After we have read the system tree and know devids belonging to
 * this filesystem, remove the device which does not belong there.
 */
void btrfs_free_extra_devids(struct btrfs_fs_devices *fs_devices, int step)
{
	struct btrfs_device *device, *next;
	struct btrfs_device *latest_dev = NULL;

	mutex_lock(&uuid_mutex);
again:
	/* This is the initialized path, it is safe to release the devices. */
	list_for_each_entry_safe(device, next, &fs_devices->devices, dev_list) {
		if (test_bit(BTRFS_DEV_STATE_IN_FS_METADATA,
							&device->dev_state)) {
			if (!test_bit(BTRFS_DEV_STATE_REPLACE_TGT,
			     &device->dev_state) &&
			     (!latest_dev ||
			      device->generation > latest_dev->generation)) {
				latest_dev = device;
			}
			continue;
		}

		if (device->devid == BTRFS_DEV_REPLACE_DEVID) {
			/*
			 * In the first step, keep the device which has
			 * the correct fsid and the devid that is used
			 * for the dev_replace procedure.