[linux.git] / drivers / md / raid1.h

#ifndef _RAID1_H
#define _RAID1_H

struct mirror_info {
	struct md_rdev	*rdev;
	sector_t	head_position;
};

/*
 * memory pools need a pointer to the mddev, so they can force an unplug
 * when memory is tight, and a count of the number of drives that the
 * pool was allocated for, so they know how much to allocate and free.
 * mddev->raid_disks cannot be used, as it can change while a pool is active
 * These two datums are stored in a kmalloced struct.
 * The 'raid_disks' here is twice the raid_disks in r1conf.
 * This allows space for each 'real' device can have a replacement in the
 * second half of the array.
 */

struct pool_info {
	struct mddev *mddev;
	int	raid_disks;
};

struct r1conf {
	struct mddev		*mddev;
	struct mirror_info	*mirrors;	/* twice 'raid_disks' to
						 * allow for replacements.
						 */
	int			raid_disks;

	/* When choose the best device for a read (read_balance())
	 * we try to keep sequential reads one the same device
	 * using 'last_used' and 'next_seq_sect'
	 */
	int			last_used;
	sector_t		next_seq_sect;
	/* During resync, read_balancing is only allowed on the part
	 * of the array that has been resynced.  'next_resync' tells us
	 * where that is.
	 */
	sector_t		next_resync;

	spinlock_t		device_lock;

	/* list of 'struct r1bio' that need to be processed by raid1d,
	 * whether to retry a read, writeout a resync or recovery
	 * block, or anything else.
	 */
	struct list_head	retry_list;

	/* queue pending writes to be submitted on unplug */
	struct bio_list		pending_bio_list;
	int			pending_count;

	/* for use when syncing mirrors:
	 * We don't allow both normal IO and resync/recovery IO at
	 * the same time - resync/recovery can only happen when there
	 * is no other IO.  So when either is active, the other has to wait.
	 * See more details description in raid1.c near raise_barrier().
	 */
	wait_queue_head_t	wait_barrier;
	spinlock_t		resync_lock;
	int			nr_pending;
	int			nr_waiting;
	int			nr_queued;
	int			barrier;

	/* Set to 1 if a full sync is needed, (fresh device added).
	 * Cleared when a sync completes.
	 */
	int			fullsync;

	/* When the same as mddev->recovery_disabled we don't allow
	 * recovery to be attempted as we expect a read error.
	 */
	int			recovery_disabled;


	/* poolinfo contains information about the content of the
	 * mempools - it changes when the array grows or shrinks
	 */
	struct pool_info	*poolinfo;
	mempool_t		*r1bio_pool;
	mempool_t		*r1buf_pool;

	/* temporary buffer to synchronous IO when attempting to repair
	 * a read error.
	 */
	struct page		*tmppage;


	/* When taking over an array from a different personality, we store
	 * the new thread here until we fully activate the array.
	 */
	struct md_thread	*thread;
};

/*
 * this is our 'private' RAID1 bio.
 *
 * it contains information about what kind of IO operations were started
 * for this RAID1 operation, and about their status:
 */

struct r1bio {
	atomic_t		remaining; /* 'have we finished' count,
					    * used from IRQ handlers
					    */
	atomic_t		behind_remaining; /* number of write-behind ios remaining
						 * in this BehindIO request
						 */
	sector_t		sector;
	int			sectors;
	unsigned long		state;
	struct mddev		*mddev;
	/*
	 * original bio going to /dev/mdx
	 */
	struct bio		*master_bio;
	/*
	 * if the IO is in READ direction, then this is where we read
	 */
	int			read_disk;

	struct list_head	retry_list;
	/* Next two are only valid when R1BIO_BehindIO is set */
	struct bio_vec		*behind_bvecs;
	int			behind_page_count;
	/*
	 * if the IO is in WRITE direction, then multiple bios are used.
	 * We choose the number when they are allocated.
	 */
	struct bio		*bios[0];
	/* DO NOT PUT ANY NEW FIELDS HERE - bios array is contiguously alloced*/
};

/* when we get a read error on a read-only array, we redirect to another
 * device without failing the first device, or trying to over-write to
 * correct the read error.  To keep track of bad blocks on a per-bio
 * level, we store IO_BLOCKED in the appropriate 'bios' pointer
 */
#define IO_BLOCKED ((struct bio *)1)
/* When we successfully write to a known bad-block, we need to remove the
 * bad-block marking which must be done from process context.  So we record
 * the success by setting bios[n] to IO_MADE_GOOD
 */
#define IO_MADE_GOOD ((struct bio *)2)

#define BIO_SPECIAL(bio) ((unsigned long)bio <= 2)

/* bits for r1bio.state */
#define	R1BIO_Uptodate	0
#define	R1BIO_IsSync	1
#define	R1BIO_Degraded	2
#define	R1BIO_BehindIO	3
/* Set ReadError on bios that experience a readerror so that
 * raid1d knows what to do with them.
 */
#define R1BIO_ReadError 4
/* For write-behind requests, we call bi_end_io when
 * the last non-write-behind device completes, providing
 * any write was successful.  Otherwise we call when
 * any write-behind write succeeds, otherwise we call
 * with failure when last write completes (and all failed).
 * Record that bi_end_io was called with this flag...
 */
#define	R1BIO_Returned 6
/* If a write for this request means we can clear some
 * known-bad-block records, we set this flag
 */
#define	R1BIO_MadeGood 7
#define	R1BIO_WriteError 8

extern int md_raid1_congested(struct mddev *mddev, int bits);

#endif
Commit	Line	Data
1da177e4 LT	1	#ifndef _RAID1_H
	2	#define _RAID1_H
	3
1da177e4	4	struct mirror_info {
3cb03002	5	struct md_rdev *rdev;
1da177e4 LT	6	sector_t head_position;
	7	};
	8
	9	/*
	10	* memory pools need a pointer to the mddev, so they can force an unplug
	11	* when memory is tight, and a count of the number of drives that the
	12	* pool was allocated for, so they know how much to allocate and free.
	13	* mddev->raid_disks cannot be used, as it can change while a pool is active
	14	* These two datums are stored in a kmalloced struct.
8f19ccb2 N	15	* The 'raid_disks' here is twice the raid_disks in r1conf.
	16	* This allows space for each 'real' device can have a replacement in the
	17	* second half of the array.
1da177e4 LT	18	*/
	19
	20	struct pool_info {
fd01b88c	21	struct mddev *mddev;
1da177e4 LT	22	int raid_disks;
	23	};
	24
e8096360	25	struct r1conf {
fd01b88c	26	struct mddev *mddev;
8f19ccb2 N	27	struct mirror_info mirrors; / twice 'raid_disks' to
	28	* allow for replacements.
	29	*/
1da177e4	30	int raid_disks;
ce550c20 N	31
	32	/* When choose the best device for a read (read_balance())
	33	* we try to keep sequential reads one the same device
	34	* using 'last_used' and 'next_seq_sect'
	35	*/
1da177e4 LT	36	int last_used;
1da177e4 LT	37	sector_t next_seq_sect;
ce550c20 N	38	/* During resync, read_balancing is only allowed on the part
	39	* of the array that has been resynced. 'next_resync' tells us
	40	* where that is.
	41	*/
	42	sector_t next_resync;
	43
1da177e4 LT	44	spinlock_t device_lock;
1da177e4 LT	45
9f2c9d12 N	46	/* list of 'struct r1bio' that need to be processed by raid1d,
	47	* whether to retry a read, writeout a resync or recovery
	48	* block, or anything else.
ce550c20	49	*/
1da177e4	50	struct list_head retry_list;
191ea9b2	51
ce550c20 N	52	/* queue pending writes to be submitted on unplug */
ce550c20 N	53	struct bio_list pending_bio_list;
34db0cd6	54	int pending_count;
1da177e4	55
ce550c20 N	56	/* for use when syncing mirrors:
	57	* We don't allow both normal IO and resync/recovery IO at
	58	* the same time - resync/recovery can only happen when there
	59	* is no other IO. So when either is active, the other has to wait.
	60	* See more details description in raid1.c near raise_barrier().
	61	*/
	62	wait_queue_head_t wait_barrier;
1da177e4	63	spinlock_t resync_lock;
191ea9b2	64	int nr_pending;
17999be4	65	int nr_waiting;
ddaf22ab	66	int nr_queued;
191ea9b2	67	int barrier;
1da177e4	68
ce550c20 N	69	/* Set to 1 if a full sync is needed, (fresh device added).
	70	* Cleared when a sync completes.
	71	*/
	72	int fullsync;
1da177e4	73
ce550c20 N	74	/* When the same as mddev->recovery_disabled we don't allow
	75	* recovery to be attempted as we expect a read error.
	76	*/
	77	int recovery_disabled;
1da177e4	78
ddaf22ab	79
ce550c20 N	80	/* poolinfo contains information about the content of the
	81	* mempools - it changes when the array grows or shrinks
	82	*/
	83	struct pool_info *poolinfo;
9f2c9d12 N	84	mempool_t *r1bio_pool;
9f2c9d12 N	85	mempool_t *r1buf_pool;
709ae487	86
ce550c20 N	87	/* temporary buffer to synchronous IO when attempting to repair
	88	* a read error.
	89	*/
	90	struct page *tmppage;
	91
	92
709ae487 N	93	/* When taking over an array from a different personality, we store
	94	* the new thread here until we fully activate the array.
	95	*/
2b8bf345	96	struct md_thread *thread;
1da177e4 LT	97	};
1da177e4 LT	98
1da177e4 LT	99	/*
	100	* this is our 'private' RAID1 bio.
	101	*
	102	* it contains information about what kind of IO operations were started
	103	* for this RAID1 operation, and about their status:
	104	*/
	105
9f2c9d12	106	struct r1bio {
1da177e4 LT	107	atomic_t remaining; /* 'have we finished' count,
	108	* used from IRQ handlers
	109	*/
4b6d287f N	110	atomic_t behind_remaining; /* number of write-behind ios remaining
	111	* in this BehindIO request
	112	*/
1da177e4 LT	113	sector_t sector;
	114	int sectors;
	115	unsigned long state;
fd01b88c	116	struct mddev *mddev;
1da177e4 LT	117	/*
	118	* original bio going to /dev/mdx
	119	*/
	120	struct bio *master_bio;
	121	/*
	122	* if the IO is in READ direction, then this is where we read
	123	*/
	124	int read_disk;
	125
	126	struct list_head retry_list;
af6d7b76	127	/* Next two are only valid when R1BIO_BehindIO is set */
2ca68f5e	128	struct bio_vec *behind_bvecs;
af6d7b76	129	int behind_page_count;
1da177e4 LT	130	/*
	131	* if the IO is in WRITE direction, then multiple bios are used.
	132	* We choose the number when they are allocated.
	133	*/
	134	struct bio *bios[0];
191ea9b2	135	/* DO NOT PUT ANY NEW FIELDS HERE - bios array is contiguously alloced*/
1da177e4 LT	136	};
1da177e4 LT	137
cf30a473 N	138	/* when we get a read error on a read-only array, we redirect to another
	139	* device without failing the first device, or trying to over-write to
	140	* correct the read error. To keep track of bad blocks on a per-bio
	141	* level, we store IO_BLOCKED in the appropriate 'bios' pointer
	142	*/
4367af55 N	143	#define IO_BLOCKED ((struct bio *)1)
	144	/* When we successfully write to a known bad-block, we need to remove the
	145	* bad-block marking which must be done from process context. So we record
	146	* the success by setting bios[n] to IO_MADE_GOOD
	147	*/
	148	#define IO_MADE_GOOD ((struct bio *)2)
	149
	150	#define BIO_SPECIAL(bio) ((unsigned long)bio <= 2)
cf30a473	151
1da177e4 LT	152	/* bits for r1bio.state */
	153	#define R1BIO_Uptodate 0
	154	#define R1BIO_IsSync 1
191ea9b2	155	#define R1BIO_Degraded 2
a9701a30	156	#define R1BIO_BehindIO 3
d2eb35ac N	157	/* Set ReadError on bios that experience a readerror so that
	158	* raid1d knows what to do with them.
	159	*/
	160	#define R1BIO_ReadError 4
4b6d287f N	161	/* For write-behind requests, we call bi_end_io when
	162	* the last non-write-behind device completes, providing
	163	* any write was successful. Otherwise we call when
	164	* any write-behind write succeeds, otherwise we call
	165	* with failure when last write completes (and all failed).
	166	* Record that bi_end_io was called with this flag...
	167	*/
9e71f9c8	168	#define R1BIO_Returned 6
4367af55 N	169	/* If a write for this request means we can clear some
	170	* known-bad-block records, we set this flag
	171	*/
cd5ff9a1 N	172	#define R1BIO_MadeGood 7
cd5ff9a1 N	173	#define R1BIO_WriteError 8
4b6d287f	174
fd01b88c	175	extern int md_raid1_congested(struct mddev *mddev, int bits);
1ed7242e	176
1da177e4	177	#endif