[linux.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"

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;
	/* 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;

#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 */
	int nobarriers;
	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 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 {
	atomic_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;
	unsigned long flags;
	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 sector_size;
	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 mirror_num,
		     int need_raid_map);
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);
int 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,
					 char *device_path,
					 struct btrfs_device **device);
int btrfs_find_device_by_devspec(struct btrfs_fs_info *fs_info, u64 devid,
					 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,
		    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_info *fs_info, u64 devid,
				       u8 *uuid, u8 *fsid);
int btrfs_shrink_device(struct btrfs_device *device, u64 new_size);
int btrfs_init_new_device(struct btrfs_fs_info *fs_info, char *path);
int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
				  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, char *device_path);
int btrfs_is_parity_mirror(struct btrfs_mapping_tree *map_tree,
			   u64 logical, u64 len, int mirror_num);
unsigned long btrfs_full_stripe_len(struct btrfs_fs_info *fs_info,
				    struct btrfs_mapping_tree *map_tree,
				    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);
	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);
	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);

#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
67a2c45e MX	27	extern struct mutex uuid_mutex;
67a2c45e MX	28
ee22184b	29	#define BTRFS_STRIPE_LEN SZ_64K
b2117a39	30
f2984462	31	struct buffer_head;
ffbd517d CM	32	struct btrfs_pending_bios {
	33	struct bio *head;
	34	struct bio *tail;
	35	};
	36
7cc8e58d MX	37	/*
	38	* Use sequence counter to get consistent device stat data on
	39	* 32-bit processors.
	40	*/
	41	#if BITS_PER_LONG==32 && defined(CONFIG_SMP)
	42	#include <linux/seqlock.h>
	43	#define __BTRFS_NEED_DEVICE_DATA_ORDERED
	44	#define btrfs_device_data_ordered_init(device) \
	45	seqcount_init(&device->data_seqcount)
	46	#else
	47	#define btrfs_device_data_ordered_init(device) do { } while (0)
	48	#endif
	49
0b86a832 CM	50	struct btrfs_device {
0b86a832 CM	51	struct list_head dev_list;
b3075717	52	struct list_head dev_alloc_list;
2b82032c	53	struct btrfs_fs_devices *fs_devices;
fb456252	54	struct btrfs_fs_info *fs_info;
ffbd517d	55
d5ee37bc MX	56	struct rcu_string *name;
	57
	58	u64 generation;
	59
	60	spinlock_t io_lock ____cacheline_aligned;
	61	int running_pending;
ffbd517d CM	62	/* regular prio bios */
ffbd517d CM	63	struct btrfs_pending_bios pending_bios;
70fd7614	64	/* sync bios */
ffbd517d CM	65	struct btrfs_pending_bios pending_sync_bios;
ffbd517d CM	66
d5ee37bc MX	67	struct block_device *bdev;
	68
	69	/* the mode sent to blkdev_get */
	70	fmode_t mode;
	71
2b82032c	72	int writeable;
dfe25020	73	int in_fs_metadata;
cd02dca5	74	int missing;
d5e2003c	75	int can_discard;
63a212ab	76	int is_tgtdev_for_dev_replace;
b3075717	77
7cc8e58d MX	78	#ifdef __BTRFS_NEED_DEVICE_DATA_ORDERED
	79	seqcount_t data_seqcount;
	80	#endif
	81
0b86a832 CM	82	/* the internal btrfs device id */
	83	u64 devid;
	84
6ba40b61	85	/* size of the device in memory */
0b86a832 CM	86	u64 total_bytes;
0b86a832 CM	87
6ba40b61	88	/* size of the device on disk */
d6397bae CB	89	u64 disk_total_bytes;
d6397bae CB	90
0b86a832 CM	91	/* bytes used */
	92	u64 bytes_used;
	93
	94	/* optimal io alignment for this device */
	95	u32 io_align;
	96
	97	/* optimal io width for this device */
	98	u32 io_width;
3c45bfc1 DG	99	/* type and info about this device */
3c45bfc1 DG	100	u64 type;
0b86a832 CM	101
	102	/* minimal io size for this device */
	103	u32 sector_size;
	104
0b86a832	105	/* physical drive uuid (or lvm uuid) */
e17cade2	106	u8 uuid[BTRFS_UUID_SIZE];
8b712842	107
935e5cc9 MX	108	/*
	109	* size of the device on the current transaction
	110	*
	111	* This variant is update when committing the transaction,
	112	* and protected by device_list_mutex
	113	*/
	114	u64 commit_total_bytes;
	115
ce7213c7 MX	116	/* bytes used on the current transaction */
ce7213c7 MX	117	u64 commit_bytes_used;
935e5cc9 MX	118	/*
	119	* used to manage the device which is resized
	120	*
	121	* It is protected by chunk_lock.
	122	*/
	123	struct list_head resized_list;
	124
3c45bfc1 DG	125	/* for sending down flush barriers */
	126	int nobarriers;
	127	struct bio *flush_bio;
	128	struct completion flush_wait;
	129
a2de733c	130	/* per-device scrub information */
d9d181c1	131	struct scrub_ctx *scrub_device;
a2de733c	132
d458b054	133	struct btrfs_work work;
1f78160c XG	134	struct rcu_head rcu;
1f78160c XG	135	struct work_struct rcu_work;
90519d66 AJ	136
	137	/* readahead state */
	138	spinlock_t reada_lock;
	139	atomic_t reada_in_flight;
	140	u64 reada_next;
	141	struct reada_zone *reada_curr_zone;
	142	struct radix_tree_root reada_zones;
	143	struct radix_tree_root reada_extents;
387125fc	144
442a4f63 SB	145	/* disk I/O failure stats. For detailed description refer to
442a4f63 SB	146	* enum btrfs_dev_stat_values in ioctl.h */
733f4fbb	147	int dev_stats_valid;
addc3fa7 MX	148
	149	/* Counter to record the change of device stats */
	150	atomic_t dev_stats_ccnt;
442a4f63	151	atomic_t dev_stat_values[BTRFS_DEV_STAT_VALUES_MAX];
0b86a832 CM	152	};
0b86a832 CM	153
7cc8e58d MX	154	/*
	155	* If we read those variants at the context of their own lock, we needn't
	156	* use the following helpers, reading them directly is safe.
	157	*/
	158	#if BITS_PER_LONG==32 && defined(CONFIG_SMP)
	159	#define BTRFS_DEVICE_GETSET_FUNCS(name) \
	160	static inline u64 \
	161	btrfs_device_get_##name(const struct btrfs_device *dev) \
	162	{ \
	163	u64 size; \
	164	unsigned int seq; \
	165	\
	166	do { \
	167	seq = read_seqcount_begin(&dev->data_seqcount); \
	168	size = dev->name; \
	169	} while (read_seqcount_retry(&dev->data_seqcount, seq)); \
	170	return size; \
	171	} \
	172	\
	173	static inline void \
	174	btrfs_device_set_##name(struct btrfs_device *dev, u64 size) \
	175	{ \
	176	preempt_disable(); \
	177	write_seqcount_begin(&dev->data_seqcount); \
	178	dev->name = size; \
	179	write_seqcount_end(&dev->data_seqcount); \
	180	preempt_enable(); \
	181	}
	182	#elif BITS_PER_LONG==32 && defined(CONFIG_PREEMPT)
	183	#define BTRFS_DEVICE_GETSET_FUNCS(name) \
	184	static inline u64 \
	185	btrfs_device_get_##name(const struct btrfs_device *dev) \
	186	{ \
	187	u64 size; \
	188	\
	189	preempt_disable(); \
	190	size = dev->name; \
	191	preempt_enable(); \
	192	return size; \
	193	} \
	194	\
	195	static inline void \
	196	btrfs_device_set_##name(struct btrfs_device *dev, u64 size) \
	197	{ \
	198	preempt_disable(); \
	199	dev->name = size; \
	200	preempt_enable(); \
	201	}
	202	#else
	203	#define BTRFS_DEVICE_GETSET_FUNCS(name) \
	204	static inline u64 \
	205	btrfs_device_get_##name(const struct btrfs_device *dev) \
	206	{ \
	207	return dev->name; \
	208	} \
	209	\
	210	static inline void \
	211	btrfs_device_set_##name(struct btrfs_device *dev, u64 size) \
	212	{ \
	213	dev->name = size; \
	214	}
	215	#endif
	216
	217	BTRFS_DEVICE_GETSET_FUNCS(total_bytes);
218	BTRFS_DEVICE_GETSET_FUNCS(disk_total_bytes);
219	BTRFS_DEVICE_GETSET_FUNCS(bytes_used);
220
8a4b83cc CM	221	struct btrfs_fs_devices {
	222	u8 fsid[BTRFS_FSID_SIZE]; /* FS specific uuid */
	223
8a4b83cc	224	u64 num_devices;
a0af469b	225	u64 open_devices;
2b82032c	226	u64 rw_devices;
cd02dca5	227	u64 missing_devices;
2b82032c	228	u64 total_rw_bytes;
02db0844	229	u64 total_devices;
8a4b83cc	230	struct block_device *latest_bdev;
e5e9a520 CM	231
	232	/* all of the devices in the FS, protected by a mutex
	233	* so we can safely walk it to write out the supers without
9b011adf WS	234	* worrying about add/remove by the multi-device code.
	235	* Scrubbing super can kick off supers writing by holding
	236	* this mutex lock.
e5e9a520 CM	237	*/
e5e9a520 CM	238	struct mutex device_list_mutex;
8a4b83cc	239	struct list_head devices;
b3075717	240
935e5cc9	241	struct list_head resized_devices;
b3075717 CM	242	/* devices not currently being allocated */
b3075717 CM	243	struct list_head alloc_list;
8a4b83cc	244	struct list_head list;
2b82032c YZ	245
	246	struct btrfs_fs_devices *seed;
	247	int seeding;
2b82032c YZ	248
2b82032c YZ	249	int opened;
c289811c CM	250
	251	/* set when we find or add a device that doesn't have the
	252	* nonrot flag set
	253	*/
	254	int rotating;
2e7910d6	255
5a13f430	256	struct btrfs_fs_info *fs_info;
2e7910d6	257	/* sysfs kobjects */
c1b7e474	258	struct kobject fsid_kobj;
2e7910d6 AJ	259	struct kobject *device_dir_kobj;
2e7910d6 AJ	260	struct completion kobj_unregister;
8a4b83cc CM	261	};
8a4b83cc CM	262
facc8a22 MX	263	#define BTRFS_BIO_INLINE_CSUM_SIZE 64
facc8a22 MX	264
9be3395b CM	265	/*
	266	* we need the mirror number and stripe index to be passed around
	267	* the call chain while we are processing end_io (especially errors).
	268	* Really, what we need is a btrfs_bio structure that has this info
	269	* and is properly sized with its stripe array, but we're not there
	270	* quite yet. We have our own btrfs bioset, and all of the bios
	271	* we allocate are actually btrfs_io_bios. We'll cram as much of
	272	* struct btrfs_bio as we can into this over time.
	273	*/
facc8a22	274	typedef void (btrfs_io_bio_end_io_t) (struct btrfs_io_bio *bio, int err);
9be3395b	275	struct btrfs_io_bio {
c1dc0896 MX	276	unsigned int mirror_num;
	277	unsigned int stripe_index;
	278	u64 logical;
facc8a22 MX	279	u8 *csum;
	280	u8 csum_inline[BTRFS_BIO_INLINE_CSUM_SIZE];
	281	u8 *csum_allocated;
	282	btrfs_io_bio_end_io_t *end_io;
9be3395b CM	283	struct bio bio;
	284	};
	285
	286	static inline struct btrfs_io_bio btrfs_io_bio(struct bio bio)
	287	{
	288	return container_of(bio, struct btrfs_io_bio, bio);
	289	}
	290
cea9e445 CM	291	struct btrfs_bio_stripe {
	292	struct btrfs_device *dev;
	293	u64 physical;
fce3bb9a	294	u64 length; /* only used for discard mappings */
cea9e445 CM	295	};
cea9e445 CM	296
a1d3c478 JS	297	struct btrfs_bio;
	298	typedef void (btrfs_bio_end_io_t) (struct btrfs_bio *bio, int err);
	299
	300	struct btrfs_bio {
6e9606d2	301	atomic_t refs;
cea9e445	302	atomic_t stripes_pending;
c404e0dc	303	struct btrfs_fs_info *fs_info;
10f11900	304	u64 map_type; /* get from map_lookup->type */
cea9e445	305	bio_end_io_t *end_io;
7d2b4daa	306	struct bio *orig_bio;
c55f1396	307	unsigned long flags;
cea9e445	308	void *private;
a236aed1 CM	309	atomic_t error;
a236aed1 CM	310	int max_errors;
cea9e445	311	int num_stripes;
a1d3c478	312	int mirror_num;
2c8cdd6e MX	313	int num_tgtdevs;
2c8cdd6e MX	314	int *tgtdev_map;
8e5cfb55 ZL	315	/*
	316	* logical block numbers for the start of each stripe
	317	* The last one or two are p/q. These are sorted,
	318	* so raid_map[0] is the start of our full stripe
	319	*/
	320	u64 *raid_map;
cea9e445 CM	321	struct btrfs_bio_stripe stripes[];
	322	};
	323
b2117a39 MX	324	struct btrfs_device_info {
	325	struct btrfs_device *dev;
	326	u64 dev_offset;
	327	u64 max_avail;
73c5de00	328	u64 total_avail;
b2117a39 MX	329	};
b2117a39 MX	330
31e50229 LB	331	struct btrfs_raid_attr {
	332	int sub_stripes; /* sub_stripes info for map */
	333	int dev_stripes; /* stripes per dev */
	334	int devs_max; /* max devs to use */
	335	int devs_min; /* min devs needed */
8789f4fe	336	int tolerated_failures; /* max tolerated fail devs */
31e50229 LB	337	int devs_increment; /* ndevs has to be a multiple of this */
	338	int ncopies; /* how many copies to data has */
	339	};
	340
af902047	341	extern const struct btrfs_raid_attr btrfs_raid_array[BTRFS_NR_RAID_TYPES];
621292ba	342	extern const int btrfs_raid_mindev_error[BTRFS_NR_RAID_TYPES];
af902047 ZL	343	extern const u64 btrfs_raid_group[BTRFS_NR_RAID_TYPES];
af902047 ZL	344
1abe9b8a	345	struct map_lookup {
	346	u64 type;
	347	int io_align;
	348	int io_width;
3d8da678	349	u64 stripe_len;
1abe9b8a	350	int sector_size;
	351	int num_stripes;
	352	int sub_stripes;
	353	struct btrfs_bio_stripe stripes[];
	354	};
	355
a2de733c AJ	356	#define map_lookup_size(n) (sizeof(struct map_lookup) + \
	357	(sizeof(struct btrfs_bio_stripe) * (n)))
	358
c9e9f97b	359	struct btrfs_balance_args;
19a39dce	360	struct btrfs_balance_progress;
c9e9f97b ID	361	struct btrfs_balance_control {
	362	struct btrfs_fs_info *fs_info;
	363
	364	struct btrfs_balance_args data;
	365	struct btrfs_balance_args meta;
	366	struct btrfs_balance_args sys;
	367
	368	u64 flags;
19a39dce ID	369
19a39dce ID	370	struct btrfs_balance_progress stat;
c9e9f97b ID	371	};
c9e9f97b ID	372
cf8cddd3 CH	373	enum btrfs_map_op {
	374	BTRFS_MAP_READ,
	375	BTRFS_MAP_WRITE,
	376	BTRFS_MAP_DISCARD,
	377	BTRFS_MAP_GET_READ_MIRRORS,
	378	};
	379
	380	static inline enum btrfs_map_op btrfs_op(struct bio *bio)
	381	{
	382	switch (bio_op(bio)) {
	383	case REQ_OP_DISCARD:
	384	return BTRFS_MAP_DISCARD;
	385	case REQ_OP_WRITE:
	386	return BTRFS_MAP_WRITE;
	387	default:
	388	WARN_ON_ONCE(1);
	389	case REQ_OP_READ:
	390	return BTRFS_MAP_READ;
	391	}
	392	}
	393
6d07bcec MX	394	int btrfs_account_dev_extents_size(struct btrfs_device *device, u64 start,
6d07bcec MX	395	u64 end, u64 *length);
6e9606d2 ZL	396	void btrfs_get_bbio(struct btrfs_bio *bbio);
6e9606d2 ZL	397	void btrfs_put_bbio(struct btrfs_bio *bbio);
cf8cddd3	398	int btrfs_map_block(struct btrfs_fs_info *fs_info, enum btrfs_map_op op,
cea9e445	399	u64 logical, u64 *length,
a1d3c478	400	struct btrfs_bio **bbio_ret, int mirror_num);
cf8cddd3	401	int btrfs_map_sblock(struct btrfs_fs_info *fs_info, enum btrfs_map_op op,
af8e2d1d MX	402	u64 logical, u64 *length,
af8e2d1d MX	403	struct btrfs_bio **bbio_ret, int mirror_num,
8e5cfb55	404	int need_raid_map);
ab8d0fc4	405	int btrfs_rmap_block(struct btrfs_fs_info *fs_info,
a512bbf8 YZ	406	u64 chunk_start, u64 physical, u64 devid,
a512bbf8 YZ	407	u64 *logical, int naddrs, int *stripe_len);
6bccf3ab	408	int btrfs_read_sys_array(struct btrfs_fs_info *fs_info);
5b4aacef	409	int btrfs_read_chunk_tree(struct btrfs_fs_info *fs_info);
0b86a832	410	int btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
2ff7e61e	411	struct btrfs_fs_info *fs_info, u64 type);
0b86a832 CM	412	void btrfs_mapping_init(struct btrfs_mapping_tree *tree);
0b86a832 CM	413	void btrfs_mapping_tree_free(struct btrfs_mapping_tree *tree);
2ff7e61e	414	int btrfs_map_bio(struct btrfs_fs_info fs_info, struct bio bio,
8b712842	415	int mirror_num, int async_submit);
8a4b83cc	416	int btrfs_open_devices(struct btrfs_fs_devices *fs_devices,
97288f2c CH	417	fmode_t flags, void *holder);
97288f2c CH	418	int btrfs_scan_one_device(const char path, fmode_t flags, void holder,
8a4b83cc CM	419	struct btrfs_fs_devices **fs_devices_ret);
8a4b83cc CM	420	int btrfs_close_devices(struct btrfs_fs_devices *fs_devices);
9eaed21e	421	void btrfs_close_extra_devices(struct btrfs_fs_devices *fs_devices, int step);
88acff64 AJ	422	void btrfs_assign_next_active_device(struct btrfs_fs_info *fs_info,
88acff64 AJ	423	struct btrfs_device device, struct btrfs_device this_dev);
2ff7e61e	424	int btrfs_find_device_missing_or_by_path(struct btrfs_fs_info *fs_info,
7ba15b7d SB	425	char *device_path,
7ba15b7d SB	426	struct btrfs_device **device);
2ff7e61e	427	int btrfs_find_device_by_devspec(struct btrfs_fs_info *fs_info, u64 devid,
5c5c0df0	428	char *devpath,
24e0474b	429	struct btrfs_device **device);
12bd2fc0 ID	430	struct btrfs_device btrfs_alloc_device(struct btrfs_fs_info fs_info,
	431	const u64 *devid,
	432	const u8 *uuid);
2ff7e61e JM	433	int btrfs_rm_device(struct btrfs_fs_info *fs_info,
2ff7e61e JM	434	char *device_path, u64 devid);
143bede5	435	void btrfs_cleanup_fs_uuids(void);
5d964051	436	int btrfs_num_copies(struct btrfs_fs_info *fs_info, u64 logical, u64 len);
8f18cf13 CM	437	int btrfs_grow_device(struct btrfs_trans_handle *trans,
8f18cf13 CM	438	struct btrfs_device *device, u64 new_size);
aa1b8cd4	439	struct btrfs_device btrfs_find_device(struct btrfs_fs_info fs_info, u64 devid,
2b82032c	440	u8 uuid, u8 fsid);
8f18cf13	441	int btrfs_shrink_device(struct btrfs_device *device, u64 new_size);
5112febb	442	int btrfs_init_new_device(struct btrfs_fs_info fs_info, char path);
2ff7e61e JM	443	int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
2ff7e61e JM	444	char *device_path,
1c43366d	445	struct btrfs_device *srcdev,
e93c89c1	446	struct btrfs_device **device_out);
c9e9f97b ID	447	int btrfs_balance(struct btrfs_balance_control *bctl,
c9e9f97b ID	448	struct btrfs_ioctl_balance_args *bargs);
2b6ba629	449	int btrfs_resume_balance_async(struct btrfs_fs_info *fs_info);
68310a5e	450	int btrfs_recover_balance(struct btrfs_fs_info *fs_info);
837d5b6e	451	int btrfs_pause_balance(struct btrfs_fs_info *fs_info);
a7e99c69	452	int btrfs_cancel_balance(struct btrfs_fs_info *fs_info);
f7a81ea4	453	int btrfs_create_uuid_tree(struct btrfs_fs_info *fs_info);
70f80175	454	int btrfs_check_uuid_tree(struct btrfs_fs_info *fs_info);
2ff7e61e	455	int btrfs_chunk_readonly(struct btrfs_fs_info *fs_info, u64 chunk_offset);
499f377f JM	456	int find_free_dev_extent_start(struct btrfs_transaction *transaction,
	457	struct btrfs_device *device, u64 num_bytes,
	458	u64 search_start, u64 start, u64 max_avail);
6df9a95e JB	459	int find_free_dev_extent(struct btrfs_trans_handle *trans,
6df9a95e JB	460	struct btrfs_device *device, u64 num_bytes,
ba1bf481	461	u64 start, u64 max_avail);
442a4f63	462	void btrfs_dev_stat_inc_and_print(struct btrfs_device *dev, int index);
2ff7e61e	463	int btrfs_get_dev_stats(struct btrfs_fs_info *fs_info,
b27f7c0c	464	struct btrfs_ioctl_get_dev_stats *stats);
cb517eab	465	void btrfs_init_devices_late(struct btrfs_fs_info *fs_info);
733f4fbb SB	466	int btrfs_init_dev_stats(struct btrfs_fs_info *fs_info);
	467	int btrfs_run_dev_stats(struct btrfs_trans_handle *trans,
	468	struct btrfs_fs_info *fs_info);
084b6e7c QW	469	void btrfs_rm_dev_replace_remove_srcdev(struct btrfs_fs_info *fs_info,
	470	struct btrfs_device *srcdev);
	471	void btrfs_rm_dev_replace_free_srcdev(struct btrfs_fs_info *fs_info,
	472	struct btrfs_device *srcdev);
e93c89c1 SB	473	void btrfs_destroy_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
	474	struct btrfs_device *tgtdev);
	475	void btrfs_init_dev_replace_tgtdev_for_resume(struct btrfs_fs_info *fs_info,
	476	struct btrfs_device *tgtdev);
12b1c263	477	void btrfs_scratch_superblocks(struct block_device bdev, char device_path);
53b381b3 DW	478	int btrfs_is_parity_mirror(struct btrfs_mapping_tree *map_tree,
53b381b3 DW	479	u64 logical, u64 len, int mirror_num);
2ff7e61e	480	unsigned long btrfs_full_stripe_len(struct btrfs_fs_info *fs_info,
53b381b3 DW	481	struct btrfs_mapping_tree *map_tree,
53b381b3 DW	482	u64 logical);
6df9a95e	483	int btrfs_finish_chunk_alloc(struct btrfs_trans_handle *trans,
6bccf3ab	484	struct btrfs_fs_info *fs_info,
6df9a95e	485	u64 chunk_offset, u64 chunk_size);
47ab2a6c	486	int btrfs_remove_chunk(struct btrfs_trans_handle *trans,
5b4aacef	487	struct btrfs_fs_info *fs_info, u64 chunk_offset);
addc3fa7 MX	488
	489	static inline int btrfs_dev_stats_dirty(struct btrfs_device *dev)
	490	{
	491	return atomic_read(&dev->dev_stats_ccnt);
	492	}
	493
442a4f63 SB	494	static inline void btrfs_dev_stat_inc(struct btrfs_device *dev,
	495	int index)
	496	{
	497	atomic_inc(dev->dev_stat_values + index);
addc3fa7 MX	498	smp_mb__before_atomic();
addc3fa7 MX	499	atomic_inc(&dev->dev_stats_ccnt);
442a4f63 SB	500	}
	501
	502	static inline int btrfs_dev_stat_read(struct btrfs_device *dev,
	503	int index)
	504	{
	505	return atomic_read(dev->dev_stat_values + index);
	506	}
	507
	508	static inline int btrfs_dev_stat_read_and_reset(struct btrfs_device *dev,
	509	int index)
	510	{
	511	int ret;
	512
	513	ret = atomic_xchg(dev->dev_stat_values + index, 0);
addc3fa7 MX	514	smp_mb__before_atomic();
addc3fa7 MX	515	atomic_inc(&dev->dev_stats_ccnt);
442a4f63 SB	516	return ret;
	517	}
	518
	519	static inline void btrfs_dev_stat_set(struct btrfs_device *dev,
	520	int index, unsigned long val)
	521	{
	522	atomic_set(dev->dev_stat_values + index, val);
addc3fa7 MX	523	smp_mb__before_atomic();
addc3fa7 MX	524	atomic_inc(&dev->dev_stats_ccnt);
442a4f63 SB	525	}
	526
	527	static inline void btrfs_dev_stat_reset(struct btrfs_device *dev,
	528	int index)
	529	{
	530	btrfs_dev_stat_set(dev, index, 0);
	531	}
935e5cc9 MX	532
935e5cc9 MX	533	void btrfs_update_commit_device_size(struct btrfs_fs_info *fs_info);
2ff7e61e	534	void btrfs_update_commit_device_bytes_used(struct btrfs_fs_info *fs_info,
ce7213c7	535	struct btrfs_transaction *transaction);
04216820	536
c73eccf7	537	struct list_head *btrfs_get_fs_uuids(void);
5a13f430 AJ	538	void btrfs_set_fs_info_ptr(struct btrfs_fs_info *fs_info);
5a13f430 AJ	539	void btrfs_reset_fs_info_ptr(struct btrfs_fs_info *fs_info);
04216820	540
0b86a832	541	#endif