[mirror_ubuntu-artful-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"

extern struct mutex uuid_mutex;

#define BTRFS_STRIPE_LEN	(64 * 1024)

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_root *dev_root;

	struct rcu_string *name;

	u64 generation;

	spinlock_t io_lock ____cacheline_aligned;
	int running_pending;
	/* regular prio bios */
	struct btrfs_pending_bios pending_bios;
	/* WRITE_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 u64 btrfs_raid_group[BTRFS_NR_RAID_TYPES];

struct map_lookup {
	u64 type;
	int io_align;
	int io_width;
	int 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)))

/*
 * Restriper's general type filter
 */
#define BTRFS_BALANCE_DATA		(1ULL << 0)
#define BTRFS_BALANCE_SYSTEM		(1ULL << 1)
#define BTRFS_BALANCE_METADATA		(1ULL << 2)

#define BTRFS_BALANCE_TYPE_MASK		(BTRFS_BALANCE_DATA |	    \
					 BTRFS_BALANCE_SYSTEM |	    \
					 BTRFS_BALANCE_METADATA)

#define BTRFS_BALANCE_FORCE		(1ULL << 3)
#define BTRFS_BALANCE_RESUME		(1ULL << 4)

/*
 * Balance filters
 */
#define BTRFS_BALANCE_ARGS_PROFILES	(1ULL << 0)
#define BTRFS_BALANCE_ARGS_USAGE	(1ULL << 1)
#define BTRFS_BALANCE_ARGS_DEVID	(1ULL << 2)
#define BTRFS_BALANCE_ARGS_DRANGE	(1ULL << 3)
#define BTRFS_BALANCE_ARGS_VRANGE	(1ULL << 4)
#define BTRFS_BALANCE_ARGS_LIMIT	(1ULL << 5)
#define BTRFS_BALANCE_ARGS_LIMIT_RANGE	(1ULL << 6)
#define BTRFS_BALANCE_ARGS_STRIPES_RANGE (1ULL << 7)
#define BTRFS_BALANCE_ARGS_USAGE_RANGE	(1ULL << 10)

#define BTRFS_BALANCE_ARGS_MASK			\
	(BTRFS_BALANCE_ARGS_PROFILES |		\
	 BTRFS_BALANCE_ARGS_USAGE |		\
	 BTRFS_BALANCE_ARGS_DEVID | 		\
	 BTRFS_BALANCE_ARGS_DRANGE |		\
	 BTRFS_BALANCE_ARGS_VRANGE |		\
	 BTRFS_BALANCE_ARGS_LIMIT |		\
	 BTRFS_BALANCE_ARGS_LIMIT_RANGE |	\
	 BTRFS_BALANCE_ARGS_STRIPES_RANGE |	\
	 BTRFS_BALANCE_ARGS_USAGE_RANGE)

/*
 * Profile changing flags.  When SOFT is set we won't relocate chunk if
 * it already has the target profile (even though it may be
 * half-filled).
 */
#define BTRFS_BALANCE_ARGS_CONVERT	(1ULL << 8)
#define BTRFS_BALANCE_ARGS_SOFT		(1ULL << 9)

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;
};

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, int rw,
		    u64 logical, u64 *length,
		    struct btrfs_bio **bbio_ret, int mirror_num);
int btrfs_map_sblock(struct btrfs_fs_info *fs_info, int rw,
		     u64 logical, u64 *length,
		     struct btrfs_bio **bbio_ret, int mirror_num,
		     int need_raid_map);
int btrfs_rmap_block(struct btrfs_mapping_tree *map_tree,
		     u64 chunk_start, u64 physical, u64 devid,
		     u64 **logical, int *naddrs, int *stripe_len);
int btrfs_read_sys_array(struct btrfs_root *root);
int btrfs_read_chunk_tree(struct btrfs_root *root);
int btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
		      struct btrfs_root *extent_root, 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_root *root, int rw, 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);
int btrfs_find_device_missing_or_by_path(struct btrfs_root *root,
					 char *device_path,
					 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_root *root, char *device_path);
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_root *root, char *path);
int btrfs_init_dev_replace_tgtdev(struct btrfs_root *root, 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_root *root, 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_root *root,
			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_root *root,
				    struct btrfs_mapping_tree *map_tree,
				    u64 logical);
int btrfs_finish_chunk_alloc(struct btrfs_trans_handle *trans,
				struct btrfs_root *extent_root,
				u64 chunk_offset, u64 chunk_size);
int btrfs_remove_chunk(struct btrfs_trans_handle *trans,
		       struct btrfs_root *root, 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_root *root,
					struct btrfs_transaction *transaction);

static inline void lock_chunks(struct btrfs_root *root)
{
	mutex_lock(&root->fs_info->chunk_mutex);
}

static inline void unlock_chunks(struct btrfs_root *root)
{
	mutex_unlock(&root->fs_info->chunk_mutex);
}

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);
void btrfs_close_one_device(struct btrfs_device *device);

#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
b2117a39 MX	29	#define BTRFS_STRIPE_LEN (64 * 1024)
b2117a39 MX	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;
d5ee37bc	54
0b86a832	55	struct btrfs_root *dev_root;
ffbd517d	56
d5ee37bc MX	57	struct rcu_string *name;
	58
	59	u64 generation;
	60
	61	spinlock_t io_lock ____cacheline_aligned;
	62	int running_pending;
ffbd517d CM	63	/* regular prio bios */
	64	struct btrfs_pending_bios pending_bios;
	65	/* WRITE_SYNC bios */
	66	struct btrfs_pending_bios pending_sync_bios;
	67
d5ee37bc MX	68	struct block_device *bdev;
	69
	70	/* the mode sent to blkdev_get */
	71	fmode_t mode;
	72
2b82032c	73	int writeable;
dfe25020	74	int in_fs_metadata;
cd02dca5	75	int missing;
d5e2003c	76	int can_discard;
63a212ab	77	int is_tgtdev_for_dev_replace;
b3075717	78
7cc8e58d MX	79	#ifdef __BTRFS_NEED_DEVICE_DATA_ORDERED
	80	seqcount_t data_seqcount;
	81	#endif
	82
0b86a832 CM	83	/* the internal btrfs device id */
	84	u64 devid;
	85
6ba40b61	86	/* size of the device in memory */
0b86a832 CM	87	u64 total_bytes;
0b86a832 CM	88
6ba40b61	89	/* size of the device on disk */
d6397bae CB	90	u64 disk_total_bytes;
d6397bae CB	91
0b86a832 CM	92	/* bytes used */
	93	u64 bytes_used;
	94
	95	/* optimal io alignment for this device */
	96	u32 io_align;
	97
	98	/* optimal io width for this device */
	99	u32 io_width;
3c45bfc1 DG	100	/* type and info about this device */
3c45bfc1 DG	101	u64 type;
0b86a832 CM	102
	103	/* minimal io size for this device */
	104	u32 sector_size;
	105
0b86a832	106	/* physical drive uuid (or lvm uuid) */
e17cade2	107	u8 uuid[BTRFS_UUID_SIZE];
8b712842	108
935e5cc9 MX	109	/*
	110	* size of the device on the current transaction
	111	*
	112	* This variant is update when committing the transaction,
	113	* and protected by device_list_mutex
	114	*/
	115	u64 commit_total_bytes;
	116
ce7213c7 MX	117	/* bytes used on the current transaction */
ce7213c7 MX	118	u64 commit_bytes_used;
935e5cc9 MX	119	/*
	120	* used to manage the device which is resized
	121	*
	122	* It is protected by chunk_lock.
	123	*/
	124	struct list_head resized_list;
	125
3c45bfc1 DG	126	/* for sending down flush barriers */
	127	int nobarriers;
	128	struct bio *flush_bio;
	129	struct completion flush_wait;
	130
a2de733c	131	/* per-device scrub information */
d9d181c1	132	struct scrub_ctx *scrub_device;
a2de733c	133
d458b054	134	struct btrfs_work work;
1f78160c XG	135	struct rcu_head rcu;
1f78160c XG	136	struct work_struct rcu_work;
90519d66 AJ	137
	138	/* readahead state */
	139	spinlock_t reada_lock;
	140	atomic_t reada_in_flight;
	141	u64 reada_next;
	142	struct reada_zone *reada_curr_zone;
	143	struct radix_tree_root reada_zones;
	144	struct radix_tree_root reada_extents;
387125fc	145
442a4f63 SB	146	/* disk I/O failure stats. For detailed description refer to
442a4f63 SB	147	* enum btrfs_dev_stat_values in ioctl.h */
733f4fbb	148	int dev_stats_valid;
addc3fa7 MX	149
	150	/* Counter to record the change of device stats */
	151	atomic_t dev_stats_ccnt;
442a4f63	152	atomic_t dev_stat_values[BTRFS_DEV_STAT_VALUES_MAX];
0b86a832 CM	153	};
0b86a832 CM	154
7cc8e58d MX	155	/*
	156	* If we read those variants at the context of their own lock, we needn't
	157	* use the following helpers, reading them directly is safe.
	158	*/
	159	#if BITS_PER_LONG==32 && defined(CONFIG_SMP)
	160	#define BTRFS_DEVICE_GETSET_FUNCS(name) \
	161	static inline u64 \
	162	btrfs_device_get_##name(const struct btrfs_device *dev) \
	163	{ \
	164	u64 size; \
	165	unsigned int seq; \
	166	\
	167	do { \
	168	seq = read_seqcount_begin(&dev->data_seqcount); \
	169	size = dev->name; \
	170	} while (read_seqcount_retry(&dev->data_seqcount, seq)); \
	171	return size; \
	172	} \
	173	\
	174	static inline void \
	175	btrfs_device_set_##name(struct btrfs_device *dev, u64 size) \
	176	{ \
	177	preempt_disable(); \
	178	write_seqcount_begin(&dev->data_seqcount); \
	179	dev->name = size; \
	180	write_seqcount_end(&dev->data_seqcount); \
	181	preempt_enable(); \
	182	}
	183	#elif BITS_PER_LONG==32 && defined(CONFIG_PREEMPT)
	184	#define BTRFS_DEVICE_GETSET_FUNCS(name) \
	185	static inline u64 \
	186	btrfs_device_get_##name(const struct btrfs_device *dev) \
	187	{ \
	188	u64 size; \
	189	\
	190	preempt_disable(); \
	191	size = dev->name; \
	192	preempt_enable(); \
	193	return size; \
	194	} \
	195	\
	196	static inline void \
	197	btrfs_device_set_##name(struct btrfs_device *dev, u64 size) \
	198	{ \
	199	preempt_disable(); \
	200	dev->name = size; \
	201	preempt_enable(); \
	202	}
	203	#else
	204	#define BTRFS_DEVICE_GETSET_FUNCS(name) \
	205	static inline u64 \
	206	btrfs_device_get_##name(const struct btrfs_device *dev) \
	207	{ \
	208	return dev->name; \
	209	} \
	210	\
	211	static inline void \
	212	btrfs_device_set_##name(struct btrfs_device *dev, u64 size) \
	213	{ \
	214	dev->name = size; \
	215	}
	216	#endif
	217
	218	BTRFS_DEVICE_GETSET_FUNCS(total_bytes);
219	BTRFS_DEVICE_GETSET_FUNCS(disk_total_bytes);
220	BTRFS_DEVICE_GETSET_FUNCS(bytes_used);
221
8a4b83cc CM	222	struct btrfs_fs_devices {
	223	u8 fsid[BTRFS_FSID_SIZE]; /* FS specific uuid */
	224
8a4b83cc	225	u64 num_devices;
a0af469b	226	u64 open_devices;
2b82032c	227	u64 rw_devices;
cd02dca5	228	u64 missing_devices;
2b82032c	229	u64 total_rw_bytes;
02db0844	230	u64 total_devices;
8a4b83cc	231	struct block_device *latest_bdev;
e5e9a520 CM	232
	233	/* all of the devices in the FS, protected by a mutex
	234	* so we can safely walk it to write out the supers without
9b011adf WS	235	* worrying about add/remove by the multi-device code.
	236	* Scrubbing super can kick off supers writing by holding
	237	* this mutex lock.
e5e9a520 CM	238	*/
e5e9a520 CM	239	struct mutex device_list_mutex;
8a4b83cc	240	struct list_head devices;
b3075717	241
935e5cc9	242	struct list_head resized_devices;
b3075717 CM	243	/* devices not currently being allocated */
b3075717 CM	244	struct list_head alloc_list;
8a4b83cc	245	struct list_head list;
2b82032c YZ	246
	247	struct btrfs_fs_devices *seed;
	248	int seeding;
2b82032c YZ	249
2b82032c YZ	250	int opened;
c289811c CM	251
	252	/* set when we find or add a device that doesn't have the
	253	* nonrot flag set
	254	*/
	255	int rotating;
2e7910d6	256
5a13f430	257	struct btrfs_fs_info *fs_info;
2e7910d6	258	/* sysfs kobjects */
c1b7e474	259	struct kobject fsid_kobj;
2e7910d6 AJ	260	struct kobject *device_dir_kobj;
2e7910d6 AJ	261	struct completion kobj_unregister;
8a4b83cc CM	262	};
8a4b83cc CM	263
facc8a22 MX	264	#define BTRFS_BIO_INLINE_CSUM_SIZE 64
facc8a22 MX	265
9be3395b CM	266	/*
	267	* we need the mirror number and stripe index to be passed around
	268	* the call chain while we are processing end_io (especially errors).
	269	* Really, what we need is a btrfs_bio structure that has this info
	270	* and is properly sized with its stripe array, but we're not there
	271	* quite yet. We have our own btrfs bioset, and all of the bios
	272	* we allocate are actually btrfs_io_bios. We'll cram as much of
	273	* struct btrfs_bio as we can into this over time.
	274	*/
facc8a22	275	typedef void (btrfs_io_bio_end_io_t) (struct btrfs_io_bio *bio, int err);
9be3395b	276	struct btrfs_io_bio {
c1dc0896 MX	277	unsigned int mirror_num;
	278	unsigned int stripe_index;
	279	u64 logical;
facc8a22 MX	280	u8 *csum;
	281	u8 csum_inline[BTRFS_BIO_INLINE_CSUM_SIZE];
	282	u8 *csum_allocated;
	283	btrfs_io_bio_end_io_t *end_io;
9be3395b CM	284	struct bio bio;
	285	};
	286
	287	static inline struct btrfs_io_bio btrfs_io_bio(struct bio bio)
	288	{
	289	return container_of(bio, struct btrfs_io_bio, bio);
	290	}
	291
cea9e445 CM	292	struct btrfs_bio_stripe {
	293	struct btrfs_device *dev;
	294	u64 physical;
fce3bb9a	295	u64 length; /* only used for discard mappings */
cea9e445 CM	296	};
cea9e445 CM	297
a1d3c478 JS	298	struct btrfs_bio;
	299	typedef void (btrfs_bio_end_io_t) (struct btrfs_bio *bio, int err);
	300
	301	struct btrfs_bio {
6e9606d2	302	atomic_t refs;
cea9e445	303	atomic_t stripes_pending;
c404e0dc	304	struct btrfs_fs_info *fs_info;
10f11900	305	u64 map_type; /* get from map_lookup->type */
cea9e445	306	bio_end_io_t *end_io;
7d2b4daa	307	struct bio *orig_bio;
c55f1396	308	unsigned long flags;
cea9e445	309	void *private;
a236aed1 CM	310	atomic_t error;
a236aed1 CM	311	int max_errors;
cea9e445	312	int num_stripes;
a1d3c478	313	int mirror_num;
2c8cdd6e MX	314	int num_tgtdevs;
2c8cdd6e MX	315	int *tgtdev_map;
8e5cfb55 ZL	316	/*
	317	* logical block numbers for the start of each stripe
	318	* The last one or two are p/q. These are sorted,
	319	* so raid_map[0] is the start of our full stripe
	320	*/
	321	u64 *raid_map;
cea9e445 CM	322	struct btrfs_bio_stripe stripes[];
	323	};
	324
b2117a39 MX	325	struct btrfs_device_info {
	326	struct btrfs_device *dev;
	327	u64 dev_offset;
	328	u64 max_avail;
73c5de00	329	u64 total_avail;
b2117a39 MX	330	};
b2117a39 MX	331
31e50229 LB	332	struct btrfs_raid_attr {
	333	int sub_stripes; /* sub_stripes info for map */
	334	int dev_stripes; /* stripes per dev */
	335	int devs_max; /* max devs to use */
	336	int devs_min; /* min devs needed */
8789f4fe	337	int tolerated_failures; /* max tolerated fail devs */
31e50229 LB	338	int devs_increment; /* ndevs has to be a multiple of this */
	339	int ncopies; /* how many copies to data has */
	340	};
	341
af902047 ZL	342	extern const struct btrfs_raid_attr btrfs_raid_array[BTRFS_NR_RAID_TYPES];
	343
	344	extern const u64 btrfs_raid_group[BTRFS_NR_RAID_TYPES];
	345
1abe9b8a	346	struct map_lookup {
	347	u64 type;
	348	int io_align;
	349	int io_width;
	350	int stripe_len;
	351	int sector_size;
	352	int num_stripes;
	353	int sub_stripes;
	354	struct btrfs_bio_stripe stripes[];
	355	};
	356
a2de733c AJ	357	#define map_lookup_size(n) (sizeof(struct map_lookup) + \
	358	(sizeof(struct btrfs_bio_stripe) * (n)))
	359
f43ffb60 ID	360	/*
	361	* Restriper's general type filter
	362	*/
	363	#define BTRFS_BALANCE_DATA (1ULL << 0)
	364	#define BTRFS_BALANCE_SYSTEM (1ULL << 1)
	365	#define BTRFS_BALANCE_METADATA (1ULL << 2)
	366
	367	#define BTRFS_BALANCE_TYPE_MASK (BTRFS_BALANCE_DATA \| \
	368	BTRFS_BALANCE_SYSTEM \| \
	369	BTRFS_BALANCE_METADATA)
e4d8ec0f ID	370
e4d8ec0f ID	371	#define BTRFS_BALANCE_FORCE (1ULL << 3)
59641015	372	#define BTRFS_BALANCE_RESUME (1ULL << 4)
e4d8ec0f	373
ed25e9b2 ID	374	/*
	375	* Balance filters
	376	*/
	377	#define BTRFS_BALANCE_ARGS_PROFILES (1ULL << 0)
5ce5b3c0	378	#define BTRFS_BALANCE_ARGS_USAGE (1ULL << 1)
409d404b	379	#define BTRFS_BALANCE_ARGS_DEVID (1ULL << 2)
94e60d5a	380	#define BTRFS_BALANCE_ARGS_DRANGE (1ULL << 3)
ea67176a	381	#define BTRFS_BALANCE_ARGS_VRANGE (1ULL << 4)
7d824b6f	382	#define BTRFS_BALANCE_ARGS_LIMIT (1ULL << 5)
12907fc7	383	#define BTRFS_BALANCE_ARGS_LIMIT_RANGE (1ULL << 6)
dee32d0a	384	#define BTRFS_BALANCE_ARGS_STRIPES_RANGE (1ULL << 7)
da02c689	385	#define BTRFS_BALANCE_ARGS_USAGE_RANGE (1ULL << 10)
f43ffb60	386
849ef928 DS	387	#define BTRFS_BALANCE_ARGS_MASK \
	388	(BTRFS_BALANCE_ARGS_PROFILES \| \
	389	BTRFS_BALANCE_ARGS_USAGE \| \
	390	BTRFS_BALANCE_ARGS_DEVID \| \
	391	BTRFS_BALANCE_ARGS_DRANGE \| \
	392	BTRFS_BALANCE_ARGS_VRANGE \| \
b66d62ba DS	393	BTRFS_BALANCE_ARGS_LIMIT \| \
	394	BTRFS_BALANCE_ARGS_LIMIT_RANGE \| \
	395	BTRFS_BALANCE_ARGS_STRIPES_RANGE \| \
	396	BTRFS_BALANCE_ARGS_USAGE_RANGE)
849ef928	397
cfa4c961 ID	398	/*
	399	* Profile changing flags. When SOFT is set we won't relocate chunk if
	400	* it already has the target profile (even though it may be
	401	* half-filled).
	402	*/
e4d8ec0f	403	#define BTRFS_BALANCE_ARGS_CONVERT (1ULL << 8)
cfa4c961	404	#define BTRFS_BALANCE_ARGS_SOFT (1ULL << 9)
e4d8ec0f	405
c9e9f97b	406	struct btrfs_balance_args;
19a39dce	407	struct btrfs_balance_progress;
c9e9f97b ID	408	struct btrfs_balance_control {
	409	struct btrfs_fs_info *fs_info;
	410
	411	struct btrfs_balance_args data;
	412	struct btrfs_balance_args meta;
	413	struct btrfs_balance_args sys;
	414
	415	u64 flags;
19a39dce ID	416
19a39dce ID	417	struct btrfs_balance_progress stat;
c9e9f97b ID	418	};
c9e9f97b ID	419
6d07bcec MX	420	int btrfs_account_dev_extents_size(struct btrfs_device *device, u64 start,
6d07bcec MX	421	u64 end, u64 *length);
6e9606d2 ZL	422	void btrfs_get_bbio(struct btrfs_bio *bbio);
6e9606d2 ZL	423	void btrfs_put_bbio(struct btrfs_bio *bbio);
3ec706c8	424	int btrfs_map_block(struct btrfs_fs_info *fs_info, int rw,
cea9e445	425	u64 logical, u64 *length,
a1d3c478	426	struct btrfs_bio **bbio_ret, int mirror_num);
af8e2d1d MX	427	int btrfs_map_sblock(struct btrfs_fs_info *fs_info, int rw,
	428	u64 logical, u64 *length,
	429	struct btrfs_bio **bbio_ret, int mirror_num,
8e5cfb55	430	int need_raid_map);
a512bbf8 YZ	431	int btrfs_rmap_block(struct btrfs_mapping_tree *map_tree,
	432	u64 chunk_start, u64 physical, u64 devid,
	433	u64 *logical, int naddrs, int *stripe_len);
e4404d6e	434	int btrfs_read_sys_array(struct btrfs_root *root);
0b86a832 CM	435	int btrfs_read_chunk_tree(struct btrfs_root *root);
0b86a832 CM	436	int btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
2b82032c	437	struct btrfs_root *extent_root, u64 type);
0b86a832 CM	438	void btrfs_mapping_init(struct btrfs_mapping_tree *tree);
0b86a832 CM	439	void btrfs_mapping_tree_free(struct btrfs_mapping_tree *tree);
f188591e	440	int btrfs_map_bio(struct btrfs_root root, int rw, struct bio bio,
8b712842	441	int mirror_num, int async_submit);
8a4b83cc	442	int btrfs_open_devices(struct btrfs_fs_devices *fs_devices,
97288f2c CH	443	fmode_t flags, void *holder);
97288f2c CH	444	int btrfs_scan_one_device(const char path, fmode_t flags, void holder,
8a4b83cc CM	445	struct btrfs_fs_devices **fs_devices_ret);
8a4b83cc CM	446	int btrfs_close_devices(struct btrfs_fs_devices *fs_devices);
9eaed21e	447	void btrfs_close_extra_devices(struct btrfs_fs_devices *fs_devices, int step);
7ba15b7d SB	448	int btrfs_find_device_missing_or_by_path(struct btrfs_root *root,
	449	char *device_path,
	450	struct btrfs_device **device);
12bd2fc0 ID	451	struct btrfs_device btrfs_alloc_device(struct btrfs_fs_info fs_info,
	452	const u64 *devid,
	453	const u8 *uuid);
a061fc8d	454	int btrfs_rm_device(struct btrfs_root root, char device_path);
143bede5	455	void btrfs_cleanup_fs_uuids(void);
5d964051	456	int btrfs_num_copies(struct btrfs_fs_info *fs_info, u64 logical, u64 len);
8f18cf13 CM	457	int btrfs_grow_device(struct btrfs_trans_handle *trans,
8f18cf13 CM	458	struct btrfs_device *device, u64 new_size);
aa1b8cd4	459	struct btrfs_device btrfs_find_device(struct btrfs_fs_info fs_info, u64 devid,
2b82032c	460	u8 uuid, u8 fsid);
8f18cf13	461	int btrfs_shrink_device(struct btrfs_device *device, u64 new_size);
788f20eb	462	int btrfs_init_new_device(struct btrfs_root root, char path);
e93c89c1	463	int btrfs_init_dev_replace_tgtdev(struct btrfs_root root, char device_path,
1c43366d	464	struct btrfs_device *srcdev,
e93c89c1	465	struct btrfs_device **device_out);
c9e9f97b ID	466	int btrfs_balance(struct btrfs_balance_control *bctl,
c9e9f97b ID	467	struct btrfs_ioctl_balance_args *bargs);
2b6ba629	468	int btrfs_resume_balance_async(struct btrfs_fs_info *fs_info);
68310a5e	469	int btrfs_recover_balance(struct btrfs_fs_info *fs_info);
837d5b6e	470	int btrfs_pause_balance(struct btrfs_fs_info *fs_info);
a7e99c69	471	int btrfs_cancel_balance(struct btrfs_fs_info *fs_info);
f7a81ea4	472	int btrfs_create_uuid_tree(struct btrfs_fs_info *fs_info);
70f80175	473	int btrfs_check_uuid_tree(struct btrfs_fs_info *fs_info);
2b82032c	474	int btrfs_chunk_readonly(struct btrfs_root *root, u64 chunk_offset);
499f377f JM	475	int find_free_dev_extent_start(struct btrfs_transaction *transaction,
	476	struct btrfs_device *device, u64 num_bytes,
	477	u64 search_start, u64 start, u64 max_avail);
6df9a95e JB	478	int find_free_dev_extent(struct btrfs_trans_handle *trans,
6df9a95e JB	479	struct btrfs_device *device, u64 num_bytes,
ba1bf481	480	u64 start, u64 max_avail);
442a4f63	481	void btrfs_dev_stat_inc_and_print(struct btrfs_device *dev, int index);
c11d2c23	482	int btrfs_get_dev_stats(struct btrfs_root *root,
b27f7c0c	483	struct btrfs_ioctl_get_dev_stats *stats);
cb517eab	484	void btrfs_init_devices_late(struct btrfs_fs_info *fs_info);
733f4fbb SB	485	int btrfs_init_dev_stats(struct btrfs_fs_info *fs_info);
	486	int btrfs_run_dev_stats(struct btrfs_trans_handle *trans,
	487	struct btrfs_fs_info *fs_info);
084b6e7c QW	488	void btrfs_rm_dev_replace_remove_srcdev(struct btrfs_fs_info *fs_info,
	489	struct btrfs_device *srcdev);
	490	void btrfs_rm_dev_replace_free_srcdev(struct btrfs_fs_info *fs_info,
	491	struct btrfs_device *srcdev);
e93c89c1 SB	492	void btrfs_destroy_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
	493	struct btrfs_device *tgtdev);
	494	void btrfs_init_dev_replace_tgtdev_for_resume(struct btrfs_fs_info *fs_info,
	495	struct btrfs_device *tgtdev);
12b1c263	496	void btrfs_scratch_superblocks(struct block_device bdev, char device_path);
53b381b3 DW	497	int btrfs_is_parity_mirror(struct btrfs_mapping_tree *map_tree,
	498	u64 logical, u64 len, int mirror_num);
	499	unsigned long btrfs_full_stripe_len(struct btrfs_root *root,
	500	struct btrfs_mapping_tree *map_tree,
	501	u64 logical);
6df9a95e JB	502	int btrfs_finish_chunk_alloc(struct btrfs_trans_handle *trans,
	503	struct btrfs_root *extent_root,
	504	u64 chunk_offset, u64 chunk_size);
47ab2a6c JB	505	int btrfs_remove_chunk(struct btrfs_trans_handle *trans,
47ab2a6c JB	506	struct btrfs_root *root, u64 chunk_offset);
addc3fa7 MX	507
	508	static inline int btrfs_dev_stats_dirty(struct btrfs_device *dev)
	509	{
	510	return atomic_read(&dev->dev_stats_ccnt);
	511	}
	512
442a4f63 SB	513	static inline void btrfs_dev_stat_inc(struct btrfs_device *dev,
	514	int index)
	515	{
	516	atomic_inc(dev->dev_stat_values + index);
addc3fa7 MX	517	smp_mb__before_atomic();
addc3fa7 MX	518	atomic_inc(&dev->dev_stats_ccnt);
442a4f63 SB	519	}
	520
	521	static inline int btrfs_dev_stat_read(struct btrfs_device *dev,
	522	int index)
	523	{
	524	return atomic_read(dev->dev_stat_values + index);
	525	}
	526
	527	static inline int btrfs_dev_stat_read_and_reset(struct btrfs_device *dev,
	528	int index)
	529	{
	530	int ret;
	531
	532	ret = atomic_xchg(dev->dev_stat_values + index, 0);
addc3fa7 MX	533	smp_mb__before_atomic();
addc3fa7 MX	534	atomic_inc(&dev->dev_stats_ccnt);
442a4f63 SB	535	return ret;
	536	}
	537
	538	static inline void btrfs_dev_stat_set(struct btrfs_device *dev,
	539	int index, unsigned long val)
	540	{
	541	atomic_set(dev->dev_stat_values + index, val);
addc3fa7 MX	542	smp_mb__before_atomic();
addc3fa7 MX	543	atomic_inc(&dev->dev_stats_ccnt);
442a4f63 SB	544	}
	545
	546	static inline void btrfs_dev_stat_reset(struct btrfs_device *dev,
	547	int index)
	548	{
	549	btrfs_dev_stat_set(dev, index, 0);
	550	}
935e5cc9 MX	551
935e5cc9 MX	552	void btrfs_update_commit_device_size(struct btrfs_fs_info *fs_info);
ce7213c7 MX	553	void btrfs_update_commit_device_bytes_used(struct btrfs_root *root,
ce7213c7 MX	554	struct btrfs_transaction *transaction);
04216820 FM	555
	556	static inline void lock_chunks(struct btrfs_root *root)
	557	{
	558	mutex_lock(&root->fs_info->chunk_mutex);
	559	}
	560
	561	static inline void unlock_chunks(struct btrfs_root *root)
	562	{
	563	mutex_unlock(&root->fs_info->chunk_mutex);
	564	}
	565
c73eccf7	566	struct list_head *btrfs_get_fs_uuids(void);
5a13f430 AJ	567	void btrfs_set_fs_info_ptr(struct btrfs_fs_info *fs_info);
5a13f430 AJ	568	void btrfs_reset_fs_info_ptr(struct btrfs_fs_info *fs_info);
f190aa47	569	void btrfs_close_one_device(struct btrfs_device *device);
04216820	570
0b86a832	571	#endif