[mirror_ubuntu-bionic-kernel.git] / fs / btrfs / volumes.h

/*
 * Copyright (C) 2007 Oracle.  All rights reserved.
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of the GNU General Public
 * License v2 as published by the Free Software Foundation.
 *
 * 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., 59 Temple Place - Suite 330,
 * Boston, MA 021110-1307, USA.
 */

#ifndef __BTRFS_VOLUMES_
#define __BTRFS_VOLUMES_

#include <linux/bio.h>
#include <linux/sort.h>
#include <linux/btrfs.h>
#include "async-thread.h"

#define BTRFS_MAX_DATA_CHUNK_SIZE	(10ULL * SZ_1G)

extern struct mutex uuid_mutex;

#define BTRFS_STRIPE_LEN	SZ_64K

struct buffer_head;
struct btrfs_pending_bios {
	struct bio *head;
	struct bio *tail;
};

/*
 * Use sequence counter to get consistent device stat data on
 * 32-bit processors.
 */
#if BITS_PER_LONG==32 && defined(CONFIG_SMP)
#include <linux/seqlock.h>
#define __BTRFS_NEED_DEVICE_DATA_ORDERED
#define btrfs_device_data_ordered_init(device)	\
	seqcount_init(&device->data_seqcount)
#else
#define btrfs_device_data_ordered_init(device) do { } while (0)
#endif

struct btrfs_device {
	struct list_head dev_list;
	struct list_head dev_alloc_list;
	struct btrfs_fs_devices *fs_devices;
	struct btrfs_fs_info *fs_info;

	struct rcu_string *name;

	u64 generation;

	spinlock_t io_lock ____cacheline_aligned;
	int running_pending;
	/* When true means this device has pending chunk alloc in
	 * current transaction. Protected by chunk_mutex.
	 */
	bool has_pending_chunks;

	/* regular prio bios */
	struct btrfs_pending_bios pending_bios;
	/* sync bios */
	struct btrfs_pending_bios pending_sync_bios;

	struct block_device *bdev;

	/* the mode sent to blkdev_get */
	fmode_t mode;

	int writeable;
	int in_fs_metadata;
	int missing;
	int can_discard;
	int is_tgtdev_for_dev_replace;
	blk_status_t last_flush_error;
	int flush_bio_sent;

#ifdef __BTRFS_NEED_DEVICE_DATA_ORDERED
	seqcount_t data_seqcount;
#endif

	/* the internal btrfs device id */
	u64 devid;

	/* size of the device in memory */
	u64 total_bytes;

	/* size of the device on disk */
	u64 disk_total_bytes;

	/* bytes used */
	u64 bytes_used;

	/* optimal io alignment for this device */
	u32 io_align;

	/* optimal io width for this device */
	u32 io_width;
	/* type and info about this device */
	u64 type;

	/* minimal io size for this device */
	u32 sector_size;

	/* physical drive uuid (or lvm uuid) */
	u8 uuid[BTRFS_UUID_SIZE];

	/*
	 * size of the device on the current transaction
	 *
	 * This variant is update when committing the transaction,
	 * and protected by device_list_mutex
	 */
	u64 commit_total_bytes;

	/* bytes used on the current transaction */
	u64 commit_bytes_used;
	/*
	 * used to manage the device which is resized
	 *
	 * It is protected by chunk_lock.
	 */
	struct list_head resized_list;

	/* for sending down flush barriers */
	struct bio *flush_bio;
	struct completion flush_wait;

	/* per-device scrub information */
	struct scrub_ctx *scrub_device;

	struct btrfs_work work;
	struct rcu_head rcu;
	struct work_struct rcu_work;

	/* readahead state */
	spinlock_t reada_lock;
	atomic_t reada_in_flight;
	u64 reada_next;
	struct reada_zone *reada_curr_zone;
	struct radix_tree_root reada_zones;
	struct radix_tree_root reada_extents;

	/* disk I/O failure stats. For detailed description refer to
	 * enum btrfs_dev_stat_values in ioctl.h */
	int dev_stats_valid;

	/* Counter to record the change of device stats */
	atomic_t dev_stats_ccnt;
	atomic_t dev_stat_values[BTRFS_DEV_STAT_VALUES_MAX];
};

/*
 * If we read those variants at the context of their own lock, we needn't
 * use the following helpers, reading them directly is safe.
 */
#if BITS_PER_LONG==32 && defined(CONFIG_SMP)
#define BTRFS_DEVICE_GETSET_FUNCS(name)					\
static inline u64							\
btrfs_device_get_##name(const struct btrfs_device *dev)			\
{									\
	u64 size;							\
	unsigned int seq;						\
									\
	do {								\
		seq = read_seqcount_begin(&dev->data_seqcount);		\
		size = dev->name;					\
	} while (read_seqcount_retry(&dev->data_seqcount, seq));	\
	return size;							\
}									\
									\
static inline void							\
btrfs_device_set_##name(struct btrfs_device *dev, u64 size)		\
{									\
	preempt_disable();						\
	write_seqcount_begin(&dev->data_seqcount);			\
	dev->name = size;						\
	write_seqcount_end(&dev->data_seqcount);			\
	preempt_enable();						\
}
#elif BITS_PER_LONG==32 && defined(CONFIG_PREEMPT)
#define BTRFS_DEVICE_GETSET_FUNCS(name)					\
static inline u64							\
btrfs_device_get_##name(const struct btrfs_device *dev)			\
{									\
	u64 size;							\
									\
	preempt_disable();						\
	size = dev->name;						\
	preempt_enable();						\
	return size;							\
}									\
									\
static inline void							\
btrfs_device_set_##name(struct btrfs_device *dev, u64 size)		\
{									\
	preempt_disable();						\
	dev->name = size;						\
	preempt_enable();						\
}
#else
#define BTRFS_DEVICE_GETSET_FUNCS(name)					\
static inline u64							\
btrfs_device_get_##name(const struct btrfs_device *dev)			\
{									\
	return dev->name;						\
}									\
									\
static inline void							\
btrfs_device_set_##name(struct btrfs_device *dev, u64 size)		\
{									\
	dev->name = size;						\
}
#endif

BTRFS_DEVICE_GETSET_FUNCS(total_bytes);
BTRFS_DEVICE_GETSET_FUNCS(disk_total_bytes);
BTRFS_DEVICE_GETSET_FUNCS(bytes_used);

struct btrfs_fs_devices {
	u8 fsid[BTRFS_FSID_SIZE]; /* FS specific uuid */

	u64 num_devices;
	u64 open_devices;
	u64 rw_devices;
	u64 missing_devices;
	u64 total_rw_bytes;
	u64 total_devices;
	struct block_device *latest_bdev;

	/* all of the devices in the FS, protected by a mutex
	 * so we can safely walk it to write out the supers without
	 * worrying about add/remove by the multi-device code.
	 * Scrubbing super can kick off supers writing by holding
	 * this mutex lock.
	 */
	struct mutex device_list_mutex;
	struct list_head devices;

	struct list_head resized_devices;
	/* devices not currently being allocated */
	struct list_head alloc_list;
	struct list_head list;

	struct btrfs_fs_devices *seed;
	int seeding;

	int opened;

	/* set when we find or add a device that doesn't have the
	 * nonrot flag set
	 */
	int rotating;

	struct btrfs_fs_info *fs_info;
	/* sysfs kobjects */
	struct kobject fsid_kobj;
	struct kobject *device_dir_kobj;
	struct completion kobj_unregister;
};

#define BTRFS_BIO_INLINE_CSUM_SIZE	64

/*
 * we need the mirror number and stripe index to be passed around
 * the call chain while we are processing end_io (especially errors).
 * Really, what we need is a btrfs_bio structure that has this info
 * and is properly sized with its stripe array, but we're not there
 * quite yet.  We have our own btrfs bioset, and all of the bios
 * we allocate are actually btrfs_io_bios.  We'll cram as much of
 * struct btrfs_bio as we can into this over time.
 */
typedef void (btrfs_io_bio_end_io_t) (struct btrfs_io_bio *bio, int err);
struct btrfs_io_bio {
	unsigned int mirror_num;
	unsigned int stripe_index;
	u64 logical;
	u8 *csum;
	u8 csum_inline[BTRFS_BIO_INLINE_CSUM_SIZE];
	u8 *csum_allocated;
	btrfs_io_bio_end_io_t *end_io;
	struct bvec_iter iter;
	/*
	 * This member must come last, bio_alloc_bioset will allocate enough
	 * bytes for entire btrfs_io_bio but relies on bio being last.
	 */
	struct bio bio;
};

static inline struct btrfs_io_bio *btrfs_io_bio(struct bio *bio)
{
	return container_of(bio, struct btrfs_io_bio, bio);
}

struct btrfs_bio_stripe {
	struct btrfs_device *dev;
	u64 physical;
	u64 length; /* only used for discard mappings */
};

struct btrfs_bio;
typedef void (btrfs_bio_end_io_t) (struct btrfs_bio *bio, int err);

struct btrfs_bio {
	refcount_t refs;
	atomic_t stripes_pending;
	struct btrfs_fs_info *fs_info;
	u64 map_type; /* get from map_lookup->type */
	bio_end_io_t *end_io;
	struct bio *orig_bio;
	void *private;
	atomic_t error;
	int max_errors;
	int num_stripes;
	int mirror_num;
	int num_tgtdevs;
	int *tgtdev_map;
	/*
	 * logical block numbers for the start of each stripe
	 * The last one or two are p/q.  These are sorted,
	 * so raid_map[0] is the start of our full stripe
	 */
	u64 *raid_map;
	struct btrfs_bio_stripe stripes[];
};

struct btrfs_device_info {
	struct btrfs_device *dev;
	u64 dev_offset;
	u64 max_avail;
	u64 total_avail;
};

struct btrfs_raid_attr {
	int sub_stripes;	/* sub_stripes info for map */
	int dev_stripes;	/* stripes per dev */
	int devs_max;		/* max devs to use */
	int devs_min;		/* min devs needed */
	int tolerated_failures; /* max tolerated fail devs */
	int devs_increment;	/* ndevs has to be a multiple of this */
	int ncopies;		/* how many copies to data has */
};

extern const struct btrfs_raid_attr btrfs_raid_array[BTRFS_NR_RAID_TYPES];
extern const int btrfs_raid_mindev_error[BTRFS_NR_RAID_TYPES];
extern const u64 btrfs_raid_group[BTRFS_NR_RAID_TYPES];

struct map_lookup {
	u64 type;
	int io_align;
	int io_width;
	u64 stripe_len;
	int num_stripes;
	int sub_stripes;
	struct btrfs_bio_stripe stripes[];
};

#define map_lookup_size(n) (sizeof(struct map_lookup) + \
			    (sizeof(struct btrfs_bio_stripe) * (n)))

struct btrfs_balance_args;
struct btrfs_balance_progress;
struct btrfs_balance_control {
	struct btrfs_fs_info *fs_info;

	struct btrfs_balance_args data;
	struct btrfs_balance_args meta;
	struct btrfs_balance_args sys;

	u64 flags;

	struct btrfs_balance_progress stat;
};

enum btrfs_map_op {
	BTRFS_MAP_READ,
	BTRFS_MAP_WRITE,
	BTRFS_MAP_DISCARD,
	BTRFS_MAP_GET_READ_MIRRORS,
};

static inline enum btrfs_map_op btrfs_op(struct bio *bio)
{
	switch (bio_op(bio)) {
	case REQ_OP_DISCARD:
		return BTRFS_MAP_DISCARD;
	case REQ_OP_WRITE:
		return BTRFS_MAP_WRITE;
	default:
		WARN_ON_ONCE(1);
	case REQ_OP_READ:
		return BTRFS_MAP_READ;
	}
}

int btrfs_account_dev_extents_size(struct btrfs_device *device, u64 start,
				   u64 end, u64 *length);
void btrfs_get_bbio(struct btrfs_bio *bbio);
void btrfs_put_bbio(struct btrfs_bio *bbio);
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 btrfs_map_sblock(struct btrfs_fs_info *fs_info, enum btrfs_map_op op,
		     u64 logical, u64 *length,
		     struct btrfs_bio **bbio_ret);
int btrfs_rmap_block(struct btrfs_fs_info *fs_info,
		     u64 chunk_start, u64 physical, u64 devid,
		     u64 **logical, int *naddrs, int *stripe_len);
int btrfs_read_sys_array(struct btrfs_fs_info *fs_info);
int btrfs_read_chunk_tree(struct btrfs_fs_info *fs_info);
int btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
		      struct btrfs_fs_info *fs_info, u64 type);
void btrfs_mapping_init(struct btrfs_mapping_tree *tree);
void btrfs_mapping_tree_free(struct btrfs_mapping_tree *tree);
blk_status_t btrfs_map_bio(struct btrfs_fs_info *fs_info, struct bio *bio,
			   int mirror_num, int async_submit);
int btrfs_open_devices(struct btrfs_fs_devices *fs_devices,
		       fmode_t flags, void *holder);
int btrfs_scan_one_device(const char *path, fmode_t flags, void *holder,
			  struct btrfs_fs_devices **fs_devices_ret);
int btrfs_close_devices(struct btrfs_fs_devices *fs_devices);
void btrfs_close_extra_devices(struct btrfs_fs_devices *fs_devices, int step);
void btrfs_assign_next_active_device(struct btrfs_fs_info *fs_info,
		struct btrfs_device *device, struct btrfs_device *this_dev);
int btrfs_find_device_missing_or_by_path(struct btrfs_fs_info *fs_info,
					 const char *device_path,
					 struct btrfs_device **device);
int btrfs_find_device_by_devspec(struct btrfs_fs_info *fs_info, u64 devid,
					 const char *devpath,
					 struct btrfs_device **device);
struct btrfs_device *btrfs_alloc_device(struct btrfs_fs_info *fs_info,
					const u64 *devid,
					const u8 *uuid);
int btrfs_rm_device(struct btrfs_fs_info *fs_info,
		    const char *device_path, u64 devid);
void btrfs_cleanup_fs_uuids(void);
int btrfs_num_copies(struct btrfs_fs_info *fs_info, u64 logical, u64 len);
int btrfs_grow_device(struct btrfs_trans_handle *trans,
		      struct btrfs_device *device, u64 new_size);
struct btrfs_device *btrfs_find_device(struct btrfs_fs_devices *fs_devices,
				       u64 devid, u8 *uuid, u8 *fsid, bool seed);
int btrfs_shrink_device(struct btrfs_device *device, u64 new_size);
int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *path);
int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
				  const char *device_path,
				  struct btrfs_device *srcdev,
				  struct btrfs_device **device_out);
int btrfs_balance(struct btrfs_balance_control *bctl,
		  struct btrfs_ioctl_balance_args *bargs);
int btrfs_resume_balance_async(struct btrfs_fs_info *fs_info);
int btrfs_recover_balance(struct btrfs_fs_info *fs_info);
int btrfs_pause_balance(struct btrfs_fs_info *fs_info);
int btrfs_cancel_balance(struct btrfs_fs_info *fs_info);
int btrfs_create_uuid_tree(struct btrfs_fs_info *fs_info);
int btrfs_check_uuid_tree(struct btrfs_fs_info *fs_info);
int btrfs_chunk_readonly(struct btrfs_fs_info *fs_info, u64 chunk_offset);
int find_free_dev_extent_start(struct btrfs_transaction *transaction,
			 struct btrfs_device *device, u64 num_bytes,
			 u64 search_start, u64 *start, u64 *max_avail);
int find_free_dev_extent(struct btrfs_trans_handle *trans,
			 struct btrfs_device *device, u64 num_bytes,
			 u64 *start, u64 *max_avail);
void btrfs_dev_stat_inc_and_print(struct btrfs_device *dev, int index);
int btrfs_get_dev_stats(struct btrfs_fs_info *fs_info,
			struct btrfs_ioctl_get_dev_stats *stats);
void btrfs_init_devices_late(struct btrfs_fs_info *fs_info);
int btrfs_init_dev_stats(struct btrfs_fs_info *fs_info);
int btrfs_run_dev_stats(struct btrfs_trans_handle *trans,
			struct btrfs_fs_info *fs_info);
void btrfs_rm_dev_replace_remove_srcdev(struct btrfs_fs_info *fs_info,
					struct btrfs_device *srcdev);
void btrfs_rm_dev_replace_free_srcdev(struct btrfs_fs_info *fs_info,
				      struct btrfs_device *srcdev);
void btrfs_destroy_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
				      struct btrfs_device *tgtdev);
void btrfs_init_dev_replace_tgtdev_for_resume(struct btrfs_fs_info *fs_info,
					      struct btrfs_device *tgtdev);
void btrfs_scratch_superblocks(struct block_device *bdev, const char *device_path);
int btrfs_is_parity_mirror(struct btrfs_fs_info *fs_info,
			   u64 logical, u64 len);
unsigned long btrfs_full_stripe_len(struct btrfs_fs_info *fs_info,
				    u64 logical);
int btrfs_finish_chunk_alloc(struct btrfs_trans_handle *trans,
				struct btrfs_fs_info *fs_info,
				u64 chunk_offset, u64 chunk_size);
int btrfs_remove_chunk(struct btrfs_trans_handle *trans,
		       struct btrfs_fs_info *fs_info, u64 chunk_offset);

static inline int btrfs_dev_stats_dirty(struct btrfs_device *dev)
{
	return atomic_read(&dev->dev_stats_ccnt);
}

static inline void btrfs_dev_stat_inc(struct btrfs_device *dev,
				      int index)
{
	atomic_inc(dev->dev_stat_values + index);
	/*
	 * This memory barrier orders stores updating statistics before stores
	 * updating dev_stats_ccnt.
	 *
	 * It pairs with smp_rmb() in btrfs_run_dev_stats().
	 */
	smp_mb__before_atomic();
	atomic_inc(&dev->dev_stats_ccnt);
}

static inline int btrfs_dev_stat_read(struct btrfs_device *dev,
				      int index)
{
	return atomic_read(dev->dev_stat_values + index);
}

static inline int btrfs_dev_stat_read_and_reset(struct btrfs_device *dev,
						int index)
{
	int ret;

	ret = atomic_xchg(dev->dev_stat_values + index, 0);
	smp_mb__before_atomic();
	atomic_inc(&dev->dev_stats_ccnt);
	return ret;
}

static inline void btrfs_dev_stat_set(struct btrfs_device *dev,
				      int index, unsigned long val)
{
	atomic_set(dev->dev_stat_values + index, val);
	/*
	 * This memory barrier orders stores updating statistics before stores
	 * updating dev_stats_ccnt.
	 *
	 * It pairs with smp_rmb() in btrfs_run_dev_stats().
	 */
	smp_mb__before_atomic();
	atomic_inc(&dev->dev_stats_ccnt);
}

static inline void btrfs_dev_stat_reset(struct btrfs_device *dev,
					int index)
{
	btrfs_dev_stat_set(dev, index, 0);
}

void btrfs_update_commit_device_size(struct btrfs_fs_info *fs_info);
void btrfs_update_commit_device_bytes_used(struct btrfs_fs_info *fs_info,
					struct btrfs_transaction *transaction);

struct list_head *btrfs_get_fs_uuids(void);
void btrfs_set_fs_info_ptr(struct btrfs_fs_info *fs_info);
void btrfs_reset_fs_info_ptr(struct btrfs_fs_info *fs_info);

bool btrfs_check_rw_degradable(struct btrfs_fs_info *fs_info);

#endif
Commit	Line	Data
0b86a832 CM	1	/*
	2	* Copyright (C) 2007 Oracle. All rights reserved.
	3	*
	4	* This program is free software; you can redistribute it and/or
	5	* modify it under the terms of the GNU General Public
	6	* License v2 as published by the Free Software Foundation.
	7	*
	8	* This program is distributed in the hope that it will be useful,
	9	* but WITHOUT ANY WARRANTY; without even the implied warranty of
	10	* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
	11	* General Public License for more details.
	12	*
	13	* You should have received a copy of the GNU General Public
	14	* License along with this program; if not, write to the
	15	* Free Software Foundation, Inc., 59 Temple Place - Suite 330,
	16	* Boston, MA 021110-1307, USA.
	17	*/
	18
	19	#ifndef __BTRFS_VOLUMES_
	20	#define __BTRFS_VOLUMES_
8790d502	21
cea9e445	22	#include <linux/bio.h>
b2117a39	23	#include <linux/sort.h>
55e301fd	24	#include <linux/btrfs.h>
8b712842	25	#include "async-thread.h"
cea9e445	26
404f17e9 QW	27	#define BTRFS_MAX_DATA_CHUNK_SIZE (10ULL * SZ_1G)
404f17e9 QW	28
67a2c45e MX	29	extern struct mutex uuid_mutex;
67a2c45e MX	30
ee22184b	31	#define BTRFS_STRIPE_LEN SZ_64K
b2117a39	32
f2984462	33	struct buffer_head;
ffbd517d CM	34	struct btrfs_pending_bios {
	35	struct bio *head;
	36	struct bio *tail;
	37	};
	38
7cc8e58d MX	39	/*
	40	* Use sequence counter to get consistent device stat data on
	41	* 32-bit processors.
	42	*/
	43	#if BITS_PER_LONG==32 && defined(CONFIG_SMP)
	44	#include <linux/seqlock.h>
	45	#define __BTRFS_NEED_DEVICE_DATA_ORDERED
	46	#define btrfs_device_data_ordered_init(device) \
	47	seqcount_init(&device->data_seqcount)
	48	#else
	49	#define btrfs_device_data_ordered_init(device) do { } while (0)
	50	#endif
	51
0b86a832 CM	52	struct btrfs_device {
0b86a832 CM	53	struct list_head dev_list;
b3075717	54	struct list_head dev_alloc_list;
2b82032c	55	struct btrfs_fs_devices *fs_devices;
fb456252	56	struct btrfs_fs_info *fs_info;
ffbd517d	57
d5ee37bc MX	58	struct rcu_string *name;
	59
	60	u64 generation;
	61
	62	spinlock_t io_lock ____cacheline_aligned;
	63	int running_pending;
8217e0f8 NB	64	/* When true means this device has pending chunk alloc in
	65	* current transaction. Protected by chunk_mutex.
	66	*/
	67	bool has_pending_chunks;
	68
ffbd517d CM	69	/* regular prio bios */
ffbd517d CM	70	struct btrfs_pending_bios pending_bios;
70fd7614	71	/* sync bios */
ffbd517d CM	72	struct btrfs_pending_bios pending_sync_bios;
ffbd517d CM	73
d5ee37bc MX	74	struct block_device *bdev;
	75
	76	/* the mode sent to blkdev_get */
	77	fmode_t mode;
	78
2b82032c	79	int writeable;
dfe25020	80	int in_fs_metadata;
cd02dca5	81	int missing;
d5e2003c	82	int can_discard;
63a212ab	83	int is_tgtdev_for_dev_replace;
58efbc9f	84	blk_status_t last_flush_error;
e0ae9994	85	int flush_bio_sent;
b3075717	86
7cc8e58d MX	87	#ifdef __BTRFS_NEED_DEVICE_DATA_ORDERED
	88	seqcount_t data_seqcount;
	89	#endif
	90
0b86a832 CM	91	/* the internal btrfs device id */
	92	u64 devid;
	93
6ba40b61	94	/* size of the device in memory */
0b86a832 CM	95	u64 total_bytes;
0b86a832 CM	96
6ba40b61	97	/* size of the device on disk */
d6397bae CB	98	u64 disk_total_bytes;
d6397bae CB	99
0b86a832 CM	100	/* bytes used */
	101	u64 bytes_used;
	102
	103	/* optimal io alignment for this device */
	104	u32 io_align;
	105
	106	/* optimal io width for this device */
	107	u32 io_width;
3c45bfc1 DG	108	/* type and info about this device */
3c45bfc1 DG	109	u64 type;
0b86a832 CM	110
	111	/* minimal io size for this device */
	112	u32 sector_size;
	113
0b86a832	114	/* physical drive uuid (or lvm uuid) */
e17cade2	115	u8 uuid[BTRFS_UUID_SIZE];
8b712842	116
935e5cc9 MX	117	/*
	118	* size of the device on the current transaction
	119	*
	120	* This variant is update when committing the transaction,
	121	* and protected by device_list_mutex
	122	*/
	123	u64 commit_total_bytes;
	124
ce7213c7 MX	125	/* bytes used on the current transaction */
ce7213c7 MX	126	u64 commit_bytes_used;
935e5cc9 MX	127	/*
	128	* used to manage the device which is resized
	129	*
	130	* It is protected by chunk_lock.
	131	*/
	132	struct list_head resized_list;
	133
3c45bfc1	134	/* for sending down flush barriers */
3c45bfc1 DG	135	struct bio *flush_bio;
	136	struct completion flush_wait;
	137
a2de733c	138	/* per-device scrub information */
d9d181c1	139	struct scrub_ctx *scrub_device;
a2de733c	140
d458b054	141	struct btrfs_work work;
1f78160c XG	142	struct rcu_head rcu;
1f78160c XG	143	struct work_struct rcu_work;
90519d66 AJ	144
	145	/* readahead state */
	146	spinlock_t reada_lock;
	147	atomic_t reada_in_flight;
	148	u64 reada_next;
	149	struct reada_zone *reada_curr_zone;
	150	struct radix_tree_root reada_zones;
	151	struct radix_tree_root reada_extents;
387125fc	152
442a4f63 SB	153	/* disk I/O failure stats. For detailed description refer to
442a4f63 SB	154	* enum btrfs_dev_stat_values in ioctl.h */
733f4fbb	155	int dev_stats_valid;
addc3fa7 MX	156
	157	/* Counter to record the change of device stats */
	158	atomic_t dev_stats_ccnt;
442a4f63	159	atomic_t dev_stat_values[BTRFS_DEV_STAT_VALUES_MAX];
0b86a832 CM	160	};
0b86a832 CM	161
7cc8e58d MX	162	/*
	163	* If we read those variants at the context of their own lock, we needn't
	164	* use the following helpers, reading them directly is safe.
	165	*/
	166	#if BITS_PER_LONG==32 && defined(CONFIG_SMP)
	167	#define BTRFS_DEVICE_GETSET_FUNCS(name) \
	168	static inline u64 \
	169	btrfs_device_get_##name(const struct btrfs_device *dev) \
	170	{ \
	171	u64 size; \
	172	unsigned int seq; \
	173	\
	174	do { \
	175	seq = read_seqcount_begin(&dev->data_seqcount); \
	176	size = dev->name; \
	177	} while (read_seqcount_retry(&dev->data_seqcount, seq)); \
	178	return size; \
	179	} \
	180	\
	181	static inline void \
	182	btrfs_device_set_##name(struct btrfs_device *dev, u64 size) \
	183	{ \
	184	preempt_disable(); \
	185	write_seqcount_begin(&dev->data_seqcount); \
	186	dev->name = size; \
	187	write_seqcount_end(&dev->data_seqcount); \
	188	preempt_enable(); \
	189	}
	190	#elif BITS_PER_LONG==32 && defined(CONFIG_PREEMPT)
	191	#define BTRFS_DEVICE_GETSET_FUNCS(name) \
	192	static inline u64 \
	193	btrfs_device_get_##name(const struct btrfs_device *dev) \
	194	{ \
	195	u64 size; \
	196	\
	197	preempt_disable(); \
	198	size = dev->name; \
	199	preempt_enable(); \
	200	return size; \
	201	} \
	202	\
	203	static inline void \
	204	btrfs_device_set_##name(struct btrfs_device *dev, u64 size) \
	205	{ \
	206	preempt_disable(); \
	207	dev->name = size; \
	208	preempt_enable(); \
	209	}
	210	#else
	211	#define BTRFS_DEVICE_GETSET_FUNCS(name) \
	212	static inline u64 \
	213	btrfs_device_get_##name(const struct btrfs_device *dev) \
	214	{ \
	215	return dev->name; \
	216	} \
	217	\
	218	static inline void \
	219	btrfs_device_set_##name(struct btrfs_device *dev, u64 size) \
	220	{ \
	221	dev->name = size; \
	222	}
	223	#endif
	224
	225	BTRFS_DEVICE_GETSET_FUNCS(total_bytes);
226	BTRFS_DEVICE_GETSET_FUNCS(disk_total_bytes);
227	BTRFS_DEVICE_GETSET_FUNCS(bytes_used);
228
8a4b83cc CM	229	struct btrfs_fs_devices {
	230	u8 fsid[BTRFS_FSID_SIZE]; /* FS specific uuid */
	231
8a4b83cc	232	u64 num_devices;
a0af469b	233	u64 open_devices;
2b82032c	234	u64 rw_devices;
cd02dca5	235	u64 missing_devices;
2b82032c	236	u64 total_rw_bytes;
02db0844	237	u64 total_devices;
8a4b83cc	238	struct block_device *latest_bdev;
e5e9a520 CM	239
	240	/* all of the devices in the FS, protected by a mutex
	241	* so we can safely walk it to write out the supers without
9b011adf WS	242	* worrying about add/remove by the multi-device code.
	243	* Scrubbing super can kick off supers writing by holding
	244	* this mutex lock.
e5e9a520 CM	245	*/
e5e9a520 CM	246	struct mutex device_list_mutex;
8a4b83cc	247	struct list_head devices;
b3075717	248
935e5cc9	249	struct list_head resized_devices;
b3075717 CM	250	/* devices not currently being allocated */
b3075717 CM	251	struct list_head alloc_list;
8a4b83cc	252	struct list_head list;
2b82032c YZ	253
	254	struct btrfs_fs_devices *seed;
	255	int seeding;
2b82032c YZ	256
2b82032c YZ	257	int opened;
c289811c CM	258
	259	/* set when we find or add a device that doesn't have the
	260	* nonrot flag set
	261	*/
	262	int rotating;
2e7910d6	263
5a13f430	264	struct btrfs_fs_info *fs_info;
2e7910d6	265	/* sysfs kobjects */
c1b7e474	266	struct kobject fsid_kobj;
2e7910d6 AJ	267	struct kobject *device_dir_kobj;
2e7910d6 AJ	268	struct completion kobj_unregister;
8a4b83cc CM	269	};
8a4b83cc CM	270
facc8a22 MX	271	#define BTRFS_BIO_INLINE_CSUM_SIZE 64
facc8a22 MX	272
9be3395b CM	273	/*
	274	* we need the mirror number and stripe index to be passed around
	275	* the call chain while we are processing end_io (especially errors).
	276	* Really, what we need is a btrfs_bio structure that has this info
	277	* and is properly sized with its stripe array, but we're not there
	278	* quite yet. We have our own btrfs bioset, and all of the bios
	279	* we allocate are actually btrfs_io_bios. We'll cram as much of
	280	* struct btrfs_bio as we can into this over time.
	281	*/
facc8a22	282	typedef void (btrfs_io_bio_end_io_t) (struct btrfs_io_bio *bio, int err);
9be3395b	283	struct btrfs_io_bio {
c1dc0896 MX	284	unsigned int mirror_num;
	285	unsigned int stripe_index;
	286	u64 logical;
facc8a22 MX	287	u8 *csum;
	288	u8 csum_inline[BTRFS_BIO_INLINE_CSUM_SIZE];
	289	u8 *csum_allocated;
	290	btrfs_io_bio_end_io_t *end_io;
17347cec	291	struct bvec_iter iter;
fa1bcbe0 DS	292	/*
	293	* This member must come last, bio_alloc_bioset will allocate enough
	294	* bytes for entire btrfs_io_bio but relies on bio being last.
	295	*/
9be3395b CM	296	struct bio bio;
	297	};
	298
	299	static inline struct btrfs_io_bio btrfs_io_bio(struct bio bio)
	300	{
	301	return container_of(bio, struct btrfs_io_bio, bio);
	302	}
	303
cea9e445 CM	304	struct btrfs_bio_stripe {
	305	struct btrfs_device *dev;
	306	u64 physical;
fce3bb9a	307	u64 length; /* only used for discard mappings */
cea9e445 CM	308	};
cea9e445 CM	309
a1d3c478 JS	310	struct btrfs_bio;
	311	typedef void (btrfs_bio_end_io_t) (struct btrfs_bio *bio, int err);
	312
	313	struct btrfs_bio {
140475ae	314	refcount_t refs;
cea9e445	315	atomic_t stripes_pending;
c404e0dc	316	struct btrfs_fs_info *fs_info;
10f11900	317	u64 map_type; /* get from map_lookup->type */
cea9e445	318	bio_end_io_t *end_io;
7d2b4daa	319	struct bio *orig_bio;
cea9e445	320	void *private;
a236aed1 CM	321	atomic_t error;
a236aed1 CM	322	int max_errors;
cea9e445	323	int num_stripes;
a1d3c478	324	int mirror_num;
2c8cdd6e MX	325	int num_tgtdevs;
2c8cdd6e MX	326	int *tgtdev_map;
8e5cfb55 ZL	327	/*
	328	* logical block numbers for the start of each stripe
	329	* The last one or two are p/q. These are sorted,
	330	* so raid_map[0] is the start of our full stripe
	331	*/
	332	u64 *raid_map;
cea9e445 CM	333	struct btrfs_bio_stripe stripes[];
	334	};
	335
b2117a39 MX	336	struct btrfs_device_info {
	337	struct btrfs_device *dev;
	338	u64 dev_offset;
	339	u64 max_avail;
73c5de00	340	u64 total_avail;
b2117a39 MX	341	};
b2117a39 MX	342
31e50229 LB	343	struct btrfs_raid_attr {
	344	int sub_stripes; /* sub_stripes info for map */
	345	int dev_stripes; /* stripes per dev */
	346	int devs_max; /* max devs to use */
	347	int devs_min; /* min devs needed */
8789f4fe	348	int tolerated_failures; /* max tolerated fail devs */
31e50229 LB	349	int devs_increment; /* ndevs has to be a multiple of this */
	350	int ncopies; /* how many copies to data has */
	351	};
	352
af902047	353	extern const struct btrfs_raid_attr btrfs_raid_array[BTRFS_NR_RAID_TYPES];
621292ba	354	extern const int btrfs_raid_mindev_error[BTRFS_NR_RAID_TYPES];
af902047 ZL	355	extern const u64 btrfs_raid_group[BTRFS_NR_RAID_TYPES];
af902047 ZL	356
1abe9b8a	357	struct map_lookup {
	358	u64 type;
	359	int io_align;
	360	int io_width;
3d8da678	361	u64 stripe_len;
1abe9b8a	362	int num_stripes;
	363	int sub_stripes;
	364	struct btrfs_bio_stripe stripes[];
	365	};
	366
a2de733c AJ	367	#define map_lookup_size(n) (sizeof(struct map_lookup) + \
	368	(sizeof(struct btrfs_bio_stripe) * (n)))
	369
c9e9f97b	370	struct btrfs_balance_args;
19a39dce	371	struct btrfs_balance_progress;
c9e9f97b ID	372	struct btrfs_balance_control {
	373	struct btrfs_fs_info *fs_info;
	374
	375	struct btrfs_balance_args data;
	376	struct btrfs_balance_args meta;
	377	struct btrfs_balance_args sys;
	378
	379	u64 flags;
19a39dce ID	380
19a39dce ID	381	struct btrfs_balance_progress stat;
c9e9f97b ID	382	};
c9e9f97b ID	383
cf8cddd3 CH	384	enum btrfs_map_op {
	385	BTRFS_MAP_READ,
	386	BTRFS_MAP_WRITE,
	387	BTRFS_MAP_DISCARD,
	388	BTRFS_MAP_GET_READ_MIRRORS,
	389	};
	390
	391	static inline enum btrfs_map_op btrfs_op(struct bio *bio)
	392	{
	393	switch (bio_op(bio)) {
	394	case REQ_OP_DISCARD:
	395	return BTRFS_MAP_DISCARD;
	396	case REQ_OP_WRITE:
	397	return BTRFS_MAP_WRITE;
	398	default:
	399	WARN_ON_ONCE(1);
	400	case REQ_OP_READ:
	401	return BTRFS_MAP_READ;
	402	}
	403	}
	404
6d07bcec MX	405	int btrfs_account_dev_extents_size(struct btrfs_device *device, u64 start,
6d07bcec MX	406	u64 end, u64 *length);
6e9606d2 ZL	407	void btrfs_get_bbio(struct btrfs_bio *bbio);
6e9606d2 ZL	408	void btrfs_put_bbio(struct btrfs_bio *bbio);
cf8cddd3	409	int btrfs_map_block(struct btrfs_fs_info *fs_info, enum btrfs_map_op op,
cea9e445	410	u64 logical, u64 *length,
a1d3c478	411	struct btrfs_bio **bbio_ret, int mirror_num);
cf8cddd3	412	int btrfs_map_sblock(struct btrfs_fs_info *fs_info, enum btrfs_map_op op,
af8e2d1d	413	u64 logical, u64 *length,
825ad4c9	414	struct btrfs_bio **bbio_ret);
ab8d0fc4	415	int btrfs_rmap_block(struct btrfs_fs_info *fs_info,
a512bbf8 YZ	416	u64 chunk_start, u64 physical, u64 devid,
a512bbf8 YZ	417	u64 *logical, int naddrs, int *stripe_len);
6bccf3ab	418	int btrfs_read_sys_array(struct btrfs_fs_info *fs_info);
5b4aacef	419	int btrfs_read_chunk_tree(struct btrfs_fs_info *fs_info);
0b86a832	420	int btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
2ff7e61e	421	struct btrfs_fs_info *fs_info, u64 type);
0b86a832 CM	422	void btrfs_mapping_init(struct btrfs_mapping_tree *tree);
0b86a832 CM	423	void btrfs_mapping_tree_free(struct btrfs_mapping_tree *tree);
58efbc9f OS	424	blk_status_t btrfs_map_bio(struct btrfs_fs_info fs_info, struct bio bio,
58efbc9f OS	425	int mirror_num, int async_submit);
8a4b83cc	426	int btrfs_open_devices(struct btrfs_fs_devices *fs_devices,
97288f2c CH	427	fmode_t flags, void *holder);
97288f2c CH	428	int btrfs_scan_one_device(const char path, fmode_t flags, void holder,
8a4b83cc CM	429	struct btrfs_fs_devices **fs_devices_ret);
8a4b83cc CM	430	int btrfs_close_devices(struct btrfs_fs_devices *fs_devices);
9eaed21e	431	void btrfs_close_extra_devices(struct btrfs_fs_devices *fs_devices, int step);
88acff64 AJ	432	void btrfs_assign_next_active_device(struct btrfs_fs_info *fs_info,
88acff64 AJ	433	struct btrfs_device device, struct btrfs_device this_dev);
2ff7e61e	434	int btrfs_find_device_missing_or_by_path(struct btrfs_fs_info *fs_info,
da353f6b	435	const char *device_path,
7ba15b7d	436	struct btrfs_device **device);
2ff7e61e	437	int btrfs_find_device_by_devspec(struct btrfs_fs_info *fs_info, u64 devid,
da353f6b	438	const char *devpath,
24e0474b	439	struct btrfs_device **device);
12bd2fc0 ID	440	struct btrfs_device btrfs_alloc_device(struct btrfs_fs_info fs_info,
	441	const u64 *devid,
	442	const u8 *uuid);
2ff7e61e	443	int btrfs_rm_device(struct btrfs_fs_info *fs_info,
da353f6b	444	const char *device_path, u64 devid);
143bede5	445	void btrfs_cleanup_fs_uuids(void);
5d964051	446	int btrfs_num_copies(struct btrfs_fs_info *fs_info, u64 logical, u64 len);
8f18cf13 CM	447	int btrfs_grow_device(struct btrfs_trans_handle *trans,
8f18cf13 CM	448	struct btrfs_device *device, u64 new_size);
2f05c56f	449	struct btrfs_device btrfs_find_device(struct btrfs_fs_devices fs_devices,
34857e15	450	u64 devid, u8 uuid, u8 fsid, bool seed);
8f18cf13	451	int btrfs_shrink_device(struct btrfs_device *device, u64 new_size);
da353f6b	452	int btrfs_init_new_device(struct btrfs_fs_info fs_info, const char path);
2ff7e61e	453	int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
da353f6b	454	const char *device_path,
1c43366d	455	struct btrfs_device *srcdev,
e93c89c1	456	struct btrfs_device **device_out);
c9e9f97b ID	457	int btrfs_balance(struct btrfs_balance_control *bctl,
c9e9f97b ID	458	struct btrfs_ioctl_balance_args *bargs);
2b6ba629	459	int btrfs_resume_balance_async(struct btrfs_fs_info *fs_info);
68310a5e	460	int btrfs_recover_balance(struct btrfs_fs_info *fs_info);
837d5b6e	461	int btrfs_pause_balance(struct btrfs_fs_info *fs_info);
a7e99c69	462	int btrfs_cancel_balance(struct btrfs_fs_info *fs_info);
f7a81ea4	463	int btrfs_create_uuid_tree(struct btrfs_fs_info *fs_info);
70f80175	464	int btrfs_check_uuid_tree(struct btrfs_fs_info *fs_info);
2ff7e61e	465	int btrfs_chunk_readonly(struct btrfs_fs_info *fs_info, u64 chunk_offset);
499f377f JM	466	int find_free_dev_extent_start(struct btrfs_transaction *transaction,
	467	struct btrfs_device *device, u64 num_bytes,
	468	u64 search_start, u64 start, u64 max_avail);
6df9a95e JB	469	int find_free_dev_extent(struct btrfs_trans_handle *trans,
6df9a95e JB	470	struct btrfs_device *device, u64 num_bytes,
ba1bf481	471	u64 start, u64 max_avail);
442a4f63	472	void btrfs_dev_stat_inc_and_print(struct btrfs_device *dev, int index);
2ff7e61e	473	int btrfs_get_dev_stats(struct btrfs_fs_info *fs_info,
b27f7c0c	474	struct btrfs_ioctl_get_dev_stats *stats);
cb517eab	475	void btrfs_init_devices_late(struct btrfs_fs_info *fs_info);
733f4fbb SB	476	int btrfs_init_dev_stats(struct btrfs_fs_info *fs_info);
	477	int btrfs_run_dev_stats(struct btrfs_trans_handle *trans,
	478	struct btrfs_fs_info *fs_info);
084b6e7c QW	479	void btrfs_rm_dev_replace_remove_srcdev(struct btrfs_fs_info *fs_info,
	480	struct btrfs_device *srcdev);
	481	void btrfs_rm_dev_replace_free_srcdev(struct btrfs_fs_info *fs_info,
	482	struct btrfs_device *srcdev);
e93c89c1 SB	483	void btrfs_destroy_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
	484	struct btrfs_device *tgtdev);
	485	void btrfs_init_dev_replace_tgtdev_for_resume(struct btrfs_fs_info *fs_info,
	486	struct btrfs_device *tgtdev);
da353f6b	487	void btrfs_scratch_superblocks(struct block_device bdev, const char device_path);
592d92ee	488	int btrfs_is_parity_mirror(struct btrfs_fs_info *fs_info,
e4ff5fb5	489	u64 logical, u64 len);
2ff7e61e	490	unsigned long btrfs_full_stripe_len(struct btrfs_fs_info *fs_info,
53b381b3	491	u64 logical);
6df9a95e	492	int btrfs_finish_chunk_alloc(struct btrfs_trans_handle *trans,
6bccf3ab	493	struct btrfs_fs_info *fs_info,
6df9a95e	494	u64 chunk_offset, u64 chunk_size);
47ab2a6c	495	int btrfs_remove_chunk(struct btrfs_trans_handle *trans,
5b4aacef	496	struct btrfs_fs_info *fs_info, u64 chunk_offset);
addc3fa7 MX	497
	498	static inline int btrfs_dev_stats_dirty(struct btrfs_device *dev)
	499	{
	500	return atomic_read(&dev->dev_stats_ccnt);
	501	}
	502
442a4f63 SB	503	static inline void btrfs_dev_stat_inc(struct btrfs_device *dev,
	504	int index)
	505	{
	506	atomic_inc(dev->dev_stat_values + index);
9ee59c3a NB	507	/*
	508	* This memory barrier orders stores updating statistics before stores
	509	* updating dev_stats_ccnt.
	510	*
	511	* It pairs with smp_rmb() in btrfs_run_dev_stats().
	512	*/
addc3fa7 MX	513	smp_mb__before_atomic();
addc3fa7 MX	514	atomic_inc(&dev->dev_stats_ccnt);
442a4f63 SB	515	}
	516
	517	static inline int btrfs_dev_stat_read(struct btrfs_device *dev,
	518	int index)
	519	{
	520	return atomic_read(dev->dev_stat_values + index);
	521	}
	522
	523	static inline int btrfs_dev_stat_read_and_reset(struct btrfs_device *dev,
	524	int index)
	525	{
	526	int ret;
	527
	528	ret = atomic_xchg(dev->dev_stat_values + index, 0);
addc3fa7 MX	529	smp_mb__before_atomic();
addc3fa7 MX	530	atomic_inc(&dev->dev_stats_ccnt);
442a4f63 SB	531	return ret;
	532	}
	533
	534	static inline void btrfs_dev_stat_set(struct btrfs_device *dev,
	535	int index, unsigned long val)
	536	{
	537	atomic_set(dev->dev_stat_values + index, val);
9ee59c3a NB	538	/*
	539	* This memory barrier orders stores updating statistics before stores
	540	* updating dev_stats_ccnt.
	541	*
	542	* It pairs with smp_rmb() in btrfs_run_dev_stats().
	543	*/
addc3fa7 MX	544	smp_mb__before_atomic();
addc3fa7 MX	545	atomic_inc(&dev->dev_stats_ccnt);
442a4f63 SB	546	}
	547
	548	static inline void btrfs_dev_stat_reset(struct btrfs_device *dev,
	549	int index)
	550	{
	551	btrfs_dev_stat_set(dev, index, 0);
	552	}
935e5cc9 MX	553
935e5cc9 MX	554	void btrfs_update_commit_device_size(struct btrfs_fs_info *fs_info);
2ff7e61e	555	void btrfs_update_commit_device_bytes_used(struct btrfs_fs_info *fs_info,
ce7213c7	556	struct btrfs_transaction *transaction);
04216820	557
c73eccf7	558	struct list_head *btrfs_get_fs_uuids(void);
5a13f430 AJ	559	void btrfs_set_fs_info_ptr(struct btrfs_fs_info *fs_info);
5a13f430 AJ	560	void btrfs_reset_fs_info_ptr(struct btrfs_fs_info *fs_info);
04216820	561
21634a19 QW	562	bool btrfs_check_rw_degradable(struct btrfs_fs_info *fs_info);
21634a19 QW	563
0b86a832	564	#endif