[mirror_ubuntu-bionic-kernel.git] / drivers / md / raid1.h

/* SPDX-License-Identifier: GPL-2.0 */
#ifndef _RAID1_H
#define _RAID1_H

/*
 * each barrier unit size is 64MB fow now
 * note: it must be larger than RESYNC_DEPTH
 */
#define BARRIER_UNIT_SECTOR_BITS	17
#define BARRIER_UNIT_SECTOR_SIZE	(1<<17)
/*
 * In struct r1conf, the following members are related to I/O barrier
 * buckets,
 *	atomic_t	*nr_pending;
 *	atomic_t	*nr_waiting;
 *	atomic_t	*nr_queued;
 *	atomic_t	*barrier;
 * Each of them points to array of atomic_t variables, each array is
 * designed to have BARRIER_BUCKETS_NR elements and occupy a single
 * memory page. The data width of atomic_t variables is 4 bytes, equal
 * to 1<<(ilog2(sizeof(atomic_t))), BARRIER_BUCKETS_NR_BITS is defined
 * as (PAGE_SHIFT - ilog2(sizeof(int))) to make sure an array of
 * atomic_t variables with BARRIER_BUCKETS_NR elements just exactly
 * occupies a single memory page.
 */
#define BARRIER_BUCKETS_NR_BITS		(PAGE_SHIFT - ilog2(sizeof(atomic_t)))
#define BARRIER_BUCKETS_NR		(1<<BARRIER_BUCKETS_NR_BITS)

/* Note: raid1_info.rdev can be set to NULL asynchronously by raid1_remove_disk.
 * There are three safe ways to access raid1_info.rdev.
 * 1/ when holding mddev->reconfig_mutex
 * 2/ when resync/recovery is known to be happening - i.e. in code that is
 *    called as part of performing resync/recovery.
 * 3/ while holding rcu_read_lock(), use rcu_dereference to get the pointer
 *    and if it is non-NULL, increment rdev->nr_pending before dropping the
 *    RCU lock.
 * When .rdev is set to NULL, the nr_pending count checked again and if it has
 * been incremented, the pointer is put back in .rdev.
 */

struct raid1_info {
	struct md_rdev	*rdev;
	sector_t	head_position;

	/* When choose the best device for a read (read_balance())
	 * we try to keep sequential reads one the same device
	 */
	sector_t	next_seq_sect;
	sector_t	seq_start;
};

/*
 * 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 raid1_info	*mirrors;	/* twice 'raid_disks' to
						 * allow for replacements.
						 */
	int			raid_disks;

	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;
	/* A separate list of r1bio which just need raid_end_bio_io called.
	 * This mustn't happen for writes which had any errors if the superblock
	 * needs to be written.
	 */
	struct list_head	bio_end_io_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;
	atomic_t		nr_sync_pending;
	atomic_t		*nr_pending;
	atomic_t		*nr_waiting;
	atomic_t		*nr_queued;
	atomic_t		*barrier;
	int			array_frozen;

	/* 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;

	struct bio_set		*bio_split;

	/* 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;

	/* Keep track of cluster resync window to send to other
	 * nodes.
	 */
	sector_t		cluster_sync_low;
	sector_t		cluster_sync_high;

};

/*
 * 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;

	/*
	 * When R1BIO_BehindIO is set, we store pages for write behind
	 * in behind_master_bio.
	 */
	struct bio		*behind_master_bio;

	/*
	 * 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*/
};

/* bits for r1bio.state */
enum r1bio_state {
	R1BIO_Uptodate,
	R1BIO_IsSync,
	R1BIO_Degraded,
	R1BIO_BehindIO,
/* Set ReadError on bios that experience a readerror so that
 * raid1d knows what to do with them.
 */
	R1BIO_ReadError,
/* 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...
 */
	R1BIO_Returned,
/* If a write for this request means we can clear some
 * known-bad-block records, we set this flag
 */
	R1BIO_MadeGood,
	R1BIO_WriteError,
	R1BIO_FailFast,
};

static inline int sector_to_idx(sector_t sector)
{
	return hash_long(sector >> BARRIER_UNIT_SECTOR_BITS,
			 BARRIER_BUCKETS_NR_BITS);
}
#endif
Commit	Line	Data
b2441318	1	/* SPDX-License-Identifier: GPL-2.0 */
1da177e4 LT	2	#ifndef _RAID1_H
	3	#define _RAID1_H
	4
fd76863e	5	/*
	6	* each barrier unit size is 64MB fow now
	7	* note: it must be larger than RESYNC_DEPTH
	8	*/
	9	#define BARRIER_UNIT_SECTOR_BITS 17
	10	#define BARRIER_UNIT_SECTOR_SIZE (1<<17)
	11	/*
	12	* In struct r1conf, the following members are related to I/O barrier
	13	* buckets,
824e47da	14	* atomic_t *nr_pending;
	15	* atomic_t *nr_waiting;
	16	* atomic_t *nr_queued;
	17	* atomic_t *barrier;
	18	* Each of them points to array of atomic_t variables, each array is
	19	* designed to have BARRIER_BUCKETS_NR elements and occupy a single
	20	* memory page. The data width of atomic_t variables is 4 bytes, equal
	21	* to 1<<(ilog2(sizeof(atomic_t))), BARRIER_BUCKETS_NR_BITS is defined
	22	* as (PAGE_SHIFT - ilog2(sizeof(int))) to make sure an array of
	23	* atomic_t variables with BARRIER_BUCKETS_NR elements just exactly
	24	* occupies a single memory page.
fd76863e	25	*/
824e47da	26	#define BARRIER_BUCKETS_NR_BITS (PAGE_SHIFT - ilog2(sizeof(atomic_t)))
fd76863e	27	#define BARRIER_BUCKETS_NR (1<<BARRIER_BUCKETS_NR_BITS)
fd76863e	28
c19d783e N	29	/* Note: raid1_info.rdev can be set to NULL asynchronously by raid1_remove_disk.
	30	* There are three safe ways to access raid1_info.rdev.
	31	* 1/ when holding mddev->reconfig_mutex
	32	* 2/ when resync/recovery is known to be happening - i.e. in code that is
	33	* called as part of performing resync/recovery.
	34	* 3/ while holding rcu_read_lock(), use rcu_dereference to get the pointer
	35	* and if it is non-NULL, increment rdev->nr_pending before dropping the
	36	* RCU lock.
	37	* When .rdev is set to NULL, the nr_pending count checked again and if it has
	38	* been incremented, the pointer is put back in .rdev.
	39	*/
	40
0eaf822c	41	struct raid1_info {
3cb03002	42	struct md_rdev *rdev;
1da177e4	43	sector_t head_position;
be4d3280 SL	44
	45	/* When choose the best device for a read (read_balance())
	46	* we try to keep sequential reads one the same device
	47	*/
	48	sector_t next_seq_sect;
12cee5a8	49	sector_t seq_start;
1da177e4 LT	50	};
	51
	52	/*
	53	* memory pools need a pointer to the mddev, so they can force an unplug
	54	* when memory is tight, and a count of the number of drives that the
	55	* pool was allocated for, so they know how much to allocate and free.
	56	* mddev->raid_disks cannot be used, as it can change while a pool is active
	57	* These two datums are stored in a kmalloced struct.
8f19ccb2 N	58	* The 'raid_disks' here is twice the raid_disks in r1conf.
	59	* This allows space for each 'real' device can have a replacement in the
	60	* second half of the array.
1da177e4 LT	61	*/
	62
	63	struct pool_info {
fd01b88c	64	struct mddev *mddev;
1da177e4 LT	65	int raid_disks;
	66	};
	67
e8096360	68	struct r1conf {
fd01b88c	69	struct mddev *mddev;
0eaf822c	70	struct raid1_info mirrors; / twice 'raid_disks' to
8f19ccb2 N	71	* allow for replacements.
8f19ccb2 N	72	*/
1da177e4	73	int raid_disks;
ce550c20	74
1da177e4 LT	75	spinlock_t device_lock;
1da177e4 LT	76
9f2c9d12 N	77	/* list of 'struct r1bio' that need to be processed by raid1d,
	78	* whether to retry a read, writeout a resync or recovery
	79	* block, or anything else.
ce550c20	80	*/
1da177e4	81	struct list_head retry_list;
55ce74d4 N	82	/* A separate list of r1bio which just need raid_end_bio_io called.
	83	* This mustn't happen for writes which had any errors if the superblock
	84	* needs to be written.
	85	*/
	86	struct list_head bio_end_io_list;
191ea9b2	87
ce550c20 N	88	/* queue pending writes to be submitted on unplug */
ce550c20 N	89	struct bio_list pending_bio_list;
34db0cd6	90	int pending_count;
1da177e4	91
ce550c20 N	92	/* for use when syncing mirrors:
	93	* We don't allow both normal IO and resync/recovery IO at
	94	* the same time - resync/recovery can only happen when there
	95	* is no other IO. So when either is active, the other has to wait.
	96	* See more details description in raid1.c near raise_barrier().
	97	*/
	98	wait_queue_head_t wait_barrier;
1da177e4	99	spinlock_t resync_lock;
43ac9b84	100	atomic_t nr_sync_pending;
824e47da	101	atomic_t *nr_pending;
	102	atomic_t *nr_waiting;
	103	atomic_t *nr_queued;
	104	atomic_t *barrier;
b364e3d0	105	int array_frozen;
1da177e4	106
ce550c20 N	107	/* Set to 1 if a full sync is needed, (fresh device added).
	108	* Cleared when a sync completes.
	109	*/
	110	int fullsync;
1da177e4	111
ce550c20 N	112	/* When the same as mddev->recovery_disabled we don't allow
	113	* recovery to be attempted as we expect a read error.
	114	*/
	115	int recovery_disabled;
1da177e4	116
ce550c20 N	117	/* poolinfo contains information about the content of the
	118	* mempools - it changes when the array grows or shrinks
	119	*/
	120	struct pool_info *poolinfo;
9f2c9d12 N	121	mempool_t *r1bio_pool;
9f2c9d12 N	122	mempool_t *r1buf_pool;
709ae487	123
c230e7e5 N	124	struct bio_set *bio_split;
c230e7e5 N	125
ce550c20 N	126	/* temporary buffer to synchronous IO when attempting to repair
	127	* a read error.
	128	*/
	129	struct page *tmppage;
	130
709ae487 N	131	/* When taking over an array from a different personality, we store
	132	* the new thread here until we fully activate the array.
	133	*/
2b8bf345	134	struct md_thread *thread;
c40f341f GR	135
	136	/* Keep track of cluster resync window to send to other
	137	* nodes.
	138	*/
	139	sector_t cluster_sync_low;
	140	sector_t cluster_sync_high;
	141
1da177e4 LT	142	};
1da177e4 LT	143
1da177e4 LT	144	/*
	145	* this is our 'private' RAID1 bio.
	146	*
	147	* it contains information about what kind of IO operations were started
	148	* for this RAID1 operation, and about their status:
	149	*/
	150
9f2c9d12	151	struct r1bio {
1da177e4 LT	152	atomic_t remaining; /* 'have we finished' count,
	153	* used from IRQ handlers
	154	*/
4b6d287f N	155	atomic_t behind_remaining; /* number of write-behind ios remaining
	156	* in this BehindIO request
	157	*/
1da177e4 LT	158	sector_t sector;
	159	int sectors;
	160	unsigned long state;
fd01b88c	161	struct mddev *mddev;
1da177e4 LT	162	/*
	163	* original bio going to /dev/mdx
	164	*/
	165	struct bio *master_bio;
	166	/*
	167	* if the IO is in READ direction, then this is where we read
	168	*/
	169	int read_disk;
	170
	171	struct list_head retry_list;
841c1316 ML	172
	173	/*
	174	* When R1BIO_BehindIO is set, we store pages for write behind
	175	* in behind_master_bio.
	176	*/
	177	struct bio *behind_master_bio;
	178
1da177e4 LT	179	/*
	180	* if the IO is in WRITE direction, then multiple bios are used.
	181	* We choose the number when they are allocated.
	182	*/
	183	struct bio *bios[0];
191ea9b2	184	/* DO NOT PUT ANY NEW FIELDS HERE - bios array is contiguously alloced*/
1da177e4 LT	185	};
	186
	187	/* bits for r1bio.state */
be306c29 N	188	enum r1bio_state {
	189	R1BIO_Uptodate,
	190	R1BIO_IsSync,
	191	R1BIO_Degraded,
	192	R1BIO_BehindIO,
d2eb35ac N	193	/* Set ReadError on bios that experience a readerror so that
	194	* raid1d knows what to do with them.
	195	*/
be306c29	196	R1BIO_ReadError,
4b6d287f N	197	/* For write-behind requests, we call bi_end_io when
	198	* the last non-write-behind device completes, providing
	199	* any write was successful. Otherwise we call when
	200	* any write-behind write succeeds, otherwise we call
	201	* with failure when last write completes (and all failed).
	202	* Record that bi_end_io was called with this flag...
	203	*/
be306c29	204	R1BIO_Returned,
4367af55 N	205	/* If a write for this request means we can clear some
	206	* known-bad-block records, we set this flag
	207	*/
be306c29 N	208	R1BIO_MadeGood,
be306c29 N	209	R1BIO_WriteError,
2e52d449	210	R1BIO_FailFast,
be306c29	211	};
fd76863e	212
	213	static inline int sector_to_idx(sector_t sector)
	214	{
	215	return hash_long(sector >> BARRIER_UNIT_SECTOR_BITS,
	216	BARRIER_BUCKETS_NR_BITS);
	217	}
1da177e4	218	#endif