[linux-block.git] / include / linux / raid / md_k.h

/*
   md_k.h : kernel internal structure of the Linux MD driver
          Copyright (C) 1996-98 Ingo Molnar, Gadi Oxman
	  
   This program is free software; you can redistribute it and/or modify
   it under the terms of the GNU General Public License as published by
   the Free Software Foundation; either version 2, or (at your option)
   any later version.
   
   You should have received a copy of the GNU General Public License
   (for example /usr/src/linux/COPYING); if not, write to the Free
   Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.  
*/

#ifndef _MD_K_H
#define _MD_K_H

/* and dm-bio-list.h is not under include/linux because.... ??? */
#include "../../../drivers/md/dm-bio-list.h"

#ifdef CONFIG_BLOCK

#define	LEVEL_MULTIPATH		(-4)
#define	LEVEL_LINEAR		(-1)
#define	LEVEL_FAULTY		(-5)

/* we need a value for 'no level specified' and 0
 * means 'raid0', so we need something else.  This is
 * for internal use only
 */
#define	LEVEL_NONE		(-1000000)

#define MaxSector (~(sector_t)0)

typedef struct mddev_s mddev_t;
typedef struct mdk_rdev_s mdk_rdev_t;

/*
 * options passed in raidrun:
 */

/* Currently this must fit in an 'int' */
#define MAX_CHUNK_SIZE (1<<30)

/*
 * MD's 'extended' device
 */
struct mdk_rdev_s
{
	struct list_head same_set;	/* RAID devices within the same set */

	sector_t size;			/* Device size (in blocks) */
	mddev_t *mddev;			/* RAID array if running */
	long last_events;		/* IO event timestamp */

	struct block_device *bdev;	/* block device handle */

	struct page	*sb_page;
	int		sb_loaded;
	__u64		sb_events;
	sector_t	data_offset;	/* start of data in array */
	sector_t 	sb_start;	/* offset of the super block (in 512byte sectors) */
	int		sb_size;	/* bytes in the superblock */
	int		preferred_minor;	/* autorun support */

	struct kobject	kobj;

	/* A device can be in one of three states based on two flags:
	 * Not working:   faulty==1 in_sync==0
	 * Fully working: faulty==0 in_sync==1
	 * Working, but not
	 * in sync with array
	 *                faulty==0 in_sync==0
	 *
	 * It can never have faulty==1, in_sync==1
	 * This reduces the burden of testing multiple flags in many cases
	 */

	unsigned long	flags;
#define	Faulty		1		/* device is known to have a fault */
#define	In_sync		2		/* device is in_sync with rest of array */
#define	WriteMostly	4		/* Avoid reading if at all possible */
#define	BarriersNotsupp	5		/* BIO_RW_BARRIER is not supported */
#define	AllReserved	6		/* If whole device is reserved for
					 * one array */
#define	AutoDetected	7		/* added by auto-detect */
#define Blocked		8		/* An error occured on an externally
					 * managed array, don't allow writes
					 * until it is cleared */
#define StateChanged	9		/* Faulty or Blocked has changed during
					 * interrupt, so it needs to be
					 * notified by the thread */
	wait_queue_head_t blocked_wait;

	int desc_nr;			/* descriptor index in the superblock */
	int raid_disk;			/* role of device in array */
	int saved_raid_disk;		/* role that device used to have in the
					 * array and could again if we did a partial
					 * resync from the bitmap
					 */
	sector_t	recovery_offset;/* If this device has been partially
					 * recovered, this is where we were
					 * up to.
					 */

	atomic_t	nr_pending;	/* number of pending requests.
					 * only maintained for arrays that
					 * support hot removal
					 */
	atomic_t	read_errors;	/* number of consecutive read errors that
					 * we have tried to ignore.
					 */
	atomic_t	corrected_errors; /* number of corrected read errors,
					   * for reporting to userspace and storing
					   * in superblock.
					   */
	struct work_struct del_work;	/* used for delayed sysfs removal */

	struct sysfs_dirent *sysfs_state; /* handle for 'state'
					   * sysfs entry */
};

struct mddev_s
{
	void				*private;
	struct mdk_personality		*pers;
	dev_t				unit;
	int				md_minor;
	struct list_head 		disks;
	unsigned long			flags;
#define MD_CHANGE_DEVS	0	/* Some device status has changed */
#define MD_CHANGE_CLEAN 1	/* transition to or from 'clean' */
#define MD_CHANGE_PENDING 2	/* superblock update in progress */

	int				ro;

	struct gendisk			*gendisk;

	struct kobject			kobj;
	int				hold_active;
#define	UNTIL_IOCTL	1
#define	UNTIL_STOP	2

	/* Superblock information */
	int				major_version,
					minor_version,
					patch_version;
	int				persistent;
	int 				external;	/* metadata is
							 * managed externally */
	char				metadata_type[17]; /* externally set*/
	int				chunk_size;
	time_t				ctime, utime;
	int				level, layout;
	char				clevel[16];
	int				raid_disks;
	int				max_disks;
	sector_t			size; /* used size of component devices */
	sector_t			array_sectors; /* exported array size */
	__u64				events;

	char				uuid[16];

	/* If the array is being reshaped, we need to record the
	 * new shape and an indication of where we are up to.
	 * This is written to the superblock.
	 * If reshape_position is MaxSector, then no reshape is happening (yet).
	 */
	sector_t			reshape_position;
	int				delta_disks, new_level, new_layout, new_chunk;

	struct mdk_thread_s		*thread;	/* management thread */
	struct mdk_thread_s		*sync_thread;	/* doing resync or reconstruct */
	sector_t			curr_resync;	/* last block scheduled */
	unsigned long			resync_mark;	/* a recent timestamp */
	sector_t			resync_mark_cnt;/* blocks written at resync_mark */
	sector_t			curr_mark_cnt; /* blocks scheduled now */

	sector_t			resync_max_sectors; /* may be set by personality */

	sector_t			resync_mismatches; /* count of sectors where
							    * parity/replica mismatch found
							    */

	/* allow user-space to request suspension of IO to regions of the array */
	sector_t			suspend_lo;
	sector_t			suspend_hi;
	/* if zero, use the system-wide default */
	int				sync_speed_min;
	int				sync_speed_max;

	/* resync even though the same disks are shared among md-devices */
	int				parallel_resync;

	int				ok_start_degraded;
	/* recovery/resync flags 
	 * NEEDED:   we might need to start a resync/recover
	 * RUNNING:  a thread is running, or about to be started
	 * SYNC:     actually doing a resync, not a recovery
	 * RECOVER:  doing recovery, or need to try it.
	 * INTR:     resync needs to be aborted for some reason
	 * DONE:     thread is done and is waiting to be reaped
	 * REQUEST:  user-space has requested a sync (used with SYNC)
	 * CHECK:    user-space request for for check-only, no repair
	 * RESHAPE:  A reshape is happening
	 *
	 * If neither SYNC or RESHAPE are set, then it is a recovery.
	 */
#define	MD_RECOVERY_RUNNING	0
#define	MD_RECOVERY_SYNC	1
#define	MD_RECOVERY_RECOVER	2
#define	MD_RECOVERY_INTR	3
#define	MD_RECOVERY_DONE	4
#define	MD_RECOVERY_NEEDED	5
#define	MD_RECOVERY_REQUESTED	6
#define	MD_RECOVERY_CHECK	7
#define MD_RECOVERY_RESHAPE	8
#define	MD_RECOVERY_FROZEN	9

	unsigned long			recovery;
	int				recovery_disabled; /* if we detect that recovery
							    * will always fail, set this
							    * so we don't loop trying */

	int				in_sync;	/* know to not need resync */
	struct mutex			reconfig_mutex;
	atomic_t			active;		/* general refcount */
	atomic_t			openers;	/* number of active opens */

	int				changed;	/* true if we might need to reread partition info */
	int				degraded;	/* whether md should consider
							 * adding a spare
							 */
	int				barriers_work;	/* initialised to true, cleared as soon
							 * as a barrier request to slave
							 * fails.  Only supported
							 */
	struct bio			*biolist; 	/* bios that need to be retried
							 * because BIO_RW_BARRIER is not supported
							 */

	atomic_t			recovery_active; /* blocks scheduled, but not written */
	wait_queue_head_t		recovery_wait;
	sector_t			recovery_cp;
	sector_t			resync_min;	/* user requested sync
							 * starts here */
	sector_t			resync_max;	/* resync should pause
							 * when it gets here */

	struct sysfs_dirent		*sysfs_state;	/* handle for 'array_state'
							 * file in sysfs.
							 */
	struct sysfs_dirent		*sysfs_action;  /* handle for 'sync_action' */

	struct work_struct del_work;	/* used for delayed sysfs removal */

	spinlock_t			write_lock;
	wait_queue_head_t		sb_wait;	/* for waiting on superblock updates */
	atomic_t			pending_writes;	/* number of active superblock writes */

	unsigned int			safemode;	/* if set, update "clean" superblock
							 * when no writes pending.
							 */ 
	unsigned int			safemode_delay;
	struct timer_list		safemode_timer;
	atomic_t			writes_pending; 
	struct request_queue		*queue;	/* for plugging ... */

	atomic_t                        write_behind; /* outstanding async IO */
	unsigned int                    max_write_behind; /* 0 = sync */

	struct bitmap                   *bitmap; /* the bitmap for the device */
	struct file			*bitmap_file; /* the bitmap file */
	long				bitmap_offset; /* offset from superblock of
							* start of bitmap. May be
							* negative, but not '0'
							*/
	long				default_bitmap_offset; /* this is the offset to use when
								* hot-adding a bitmap.  It should
								* eventually be settable by sysfs.
								*/

	struct list_head		all_mddevs;
};


static inline void rdev_dec_pending(mdk_rdev_t *rdev, mddev_t *mddev)
{
	int faulty = test_bit(Faulty, &rdev->flags);
	if (atomic_dec_and_test(&rdev->nr_pending) && faulty)
		set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
}

static inline void md_sync_acct(struct block_device *bdev, unsigned long nr_sectors)
{
        atomic_add(nr_sectors, &bdev->bd_contains->bd_disk->sync_io);
}

struct mdk_personality
{
	char *name;
	int level;
	struct list_head list;
	struct module *owner;
	int (*make_request)(struct request_queue *q, struct bio *bio);
	int (*run)(mddev_t *mddev);
	int (*stop)(mddev_t *mddev);
	void (*status)(struct seq_file *seq, mddev_t *mddev);
	/* error_handler must set ->faulty and clear ->in_sync
	 * if appropriate, and should abort recovery if needed 
	 */
	void (*error_handler)(mddev_t *mddev, mdk_rdev_t *rdev);
	int (*hot_add_disk) (mddev_t *mddev, mdk_rdev_t *rdev);
	int (*hot_remove_disk) (mddev_t *mddev, int number);
	int (*spare_active) (mddev_t *mddev);
	sector_t (*sync_request)(mddev_t *mddev, sector_t sector_nr, int *skipped, int go_faster);
	int (*resize) (mddev_t *mddev, sector_t sectors);
	int (*check_reshape) (mddev_t *mddev);
	int (*start_reshape) (mddev_t *mddev);
	int (*reconfig) (mddev_t *mddev, int layout, int chunk_size);
	/* quiesce moves between quiescence states
	 * 0 - fully active
	 * 1 - no new requests allowed
	 * others - reserved
	 */
	void (*quiesce) (mddev_t *mddev, int state);
};


struct md_sysfs_entry {
	struct attribute attr;
	ssize_t (*show)(mddev_t *, char *);
	ssize_t (*store)(mddev_t *, const char *, size_t);
};


static inline char * mdname (mddev_t * mddev)
{
	return mddev->gendisk ? mddev->gendisk->disk_name : "mdX";
}

/*
 * iterates through some rdev ringlist. It's safe to remove the
 * current 'rdev'. Dont touch 'tmp' though.
 */
#define rdev_for_each_list(rdev, tmp, head)				\
	list_for_each_entry_safe(rdev, tmp, head, same_set)

/*
 * iterates through the 'same array disks' ringlist
 */
#define rdev_for_each(rdev, tmp, mddev)				\
	list_for_each_entry_safe(rdev, tmp, &((mddev)->disks), same_set)

#define rdev_for_each_rcu(rdev, mddev)				\
	list_for_each_entry_rcu(rdev, &((mddev)->disks), same_set)

typedef struct mdk_thread_s {
	void			(*run) (mddev_t *mddev);
	mddev_t			*mddev;
	wait_queue_head_t	wqueue;
	unsigned long           flags;
	struct task_struct	*tsk;
	unsigned long		timeout;
} mdk_thread_t;

#define THREAD_WAKEUP  0

#define __wait_event_lock_irq(wq, condition, lock, cmd) 		\
do {									\
	wait_queue_t __wait;						\
	init_waitqueue_entry(&__wait, current);				\
									\
	add_wait_queue(&wq, &__wait);					\
	for (;;) {							\
		set_current_state(TASK_UNINTERRUPTIBLE);		\
		if (condition)						\
			break;						\
		spin_unlock_irq(&lock);					\
		cmd;							\
		schedule();						\
		spin_lock_irq(&lock);					\
	}								\
	current->state = TASK_RUNNING;					\
	remove_wait_queue(&wq, &__wait);				\
} while (0)

#define wait_event_lock_irq(wq, condition, lock, cmd) 			\
do {									\
	if (condition)	 						\
		break;							\
	__wait_event_lock_irq(wq, condition, lock, cmd);		\
} while (0)

static inline void safe_put_page(struct page *p)
{
	if (p) put_page(p);
}

#endif /* CONFIG_BLOCK */
#endif
Commit	Line	Data
	1	/*
	2	md_k.h : kernel internal structure of the Linux MD driver
	3	Copyright (C) 1996-98 Ingo Molnar, Gadi Oxman
	4
	5	This program is free software; you can redistribute it and/or modify
	6	it under the terms of the GNU General Public License as published by
	7	the Free Software Foundation; either version 2, or (at your option)
	8	any later version.
	9
	10	You should have received a copy of the GNU General Public License
	11	(for example /usr/src/linux/COPYING); if not, write to the Free
	12	Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
	13	*/
	14
	15	#ifndef _MD_K_H
	16	#define _MD_K_H
	17
	18	/* and dm-bio-list.h is not under include/linux because.... ??? */
	19	#include "../../../drivers/md/dm-bio-list.h"
	20
	21	#ifdef CONFIG_BLOCK
	22
	23	#define LEVEL_MULTIPATH (-4)
	24	#define LEVEL_LINEAR (-1)
	25	#define LEVEL_FAULTY (-5)
	26
	27	/* we need a value for 'no level specified' and 0
	28	* means 'raid0', so we need something else. This is
	29	* for internal use only
	30	*/
	31	#define LEVEL_NONE (-1000000)
	32
	33	#define MaxSector (~(sector_t)0)
	34
	35	typedef struct mddev_s mddev_t;
	36	typedef struct mdk_rdev_s mdk_rdev_t;
	37
	38	/*
	39	* options passed in raidrun:
	40	*/
	41
	42	/* Currently this must fit in an 'int' */
	43	#define MAX_CHUNK_SIZE (1<<30)
	44
	45	/*
	46	* MD's 'extended' device
	47	*/
	48	struct mdk_rdev_s
	49	{
	50	struct list_head same_set; /* RAID devices within the same set */
	51
	52	sector_t size; /* Device size (in blocks) */
	53	mddev_t mddev; / RAID array if running */
	54	long last_events; /* IO event timestamp */
	55
	56	struct block_device bdev; / block device handle */
	57
	58	struct page *sb_page;
	59	int sb_loaded;
	60	__u64 sb_events;
	61	sector_t data_offset; /* start of data in array */
	62	sector_t sb_start; /* offset of the super block (in 512byte sectors) */
	63	int sb_size; /* bytes in the superblock */
	64	int preferred_minor; /* autorun support */
	65
	66	struct kobject kobj;
	67
	68	/* A device can be in one of three states based on two flags:
	69	* Not working: faulty==1 in_sync==0
	70	* Fully working: faulty==0 in_sync==1
	71	* Working, but not
	72	* in sync with array
	73	* faulty==0 in_sync==0
	74	*
	75	* It can never have faulty==1, in_sync==1
	76	* This reduces the burden of testing multiple flags in many cases
	77	*/
	78
	79	unsigned long flags;
	80	#define Faulty 1 /* device is known to have a fault */
	81	#define In_sync 2 /* device is in_sync with rest of array */
	82	#define WriteMostly 4 /* Avoid reading if at all possible */
	83	#define BarriersNotsupp 5 /* BIO_RW_BARRIER is not supported */
	84	#define AllReserved 6 /* If whole device is reserved for
	85	* one array */
	86	#define AutoDetected 7 /* added by auto-detect */
	87	#define Blocked 8 /* An error occured on an externally
	88	* managed array, don't allow writes
	89	* until it is cleared */
	90	#define StateChanged 9 /* Faulty or Blocked has changed during
	91	* interrupt, so it needs to be
	92	* notified by the thread */
	93	wait_queue_head_t blocked_wait;
	94
	95	int desc_nr; /* descriptor index in the superblock */
	96	int raid_disk; /* role of device in array */
	97	int saved_raid_disk; /* role that device used to have in the
	98	* array and could again if we did a partial
	99	* resync from the bitmap
	100	*/
	101	sector_t recovery_offset;/* If this device has been partially
	102	* recovered, this is where we were
	103	* up to.
	104	*/
	105
	106	atomic_t nr_pending; /* number of pending requests.
	107	* only maintained for arrays that
	108	* support hot removal
	109	*/
	110	atomic_t read_errors; /* number of consecutive read errors that
	111	* we have tried to ignore.
	112	*/
	113	atomic_t corrected_errors; /* number of corrected read errors,
	114	* for reporting to userspace and storing
	115	* in superblock.
	116	*/
	117	struct work_struct del_work; /* used for delayed sysfs removal */
	118
	119	struct sysfs_dirent sysfs_state; / handle for 'state'
	120	* sysfs entry */
	121	};
	122
	123	struct mddev_s
	124	{
	125	void *private;
	126	struct mdk_personality *pers;
	127	dev_t unit;
	128	int md_minor;
	129	struct list_head disks;
	130	unsigned long flags;
	131	#define MD_CHANGE_DEVS 0 /* Some device status has changed */
	132	#define MD_CHANGE_CLEAN 1 /* transition to or from 'clean' */
	133	#define MD_CHANGE_PENDING 2 /* superblock update in progress */
	134
	135	int ro;
	136
	137	struct gendisk *gendisk;
	138
	139	struct kobject kobj;
	140	int hold_active;
	141	#define UNTIL_IOCTL 1
	142	#define UNTIL_STOP 2
	143
	144	/* Superblock information */
	145	int major_version,
	146	minor_version,
	147	patch_version;
	148	int persistent;
	149	int external; /* metadata is
	150	* managed externally */
	151	char metadata_type[17]; /* externally set*/
	152	int chunk_size;
	153	time_t ctime, utime;
	154	int level, layout;
	155	char clevel[16];
	156	int raid_disks;
	157	int max_disks;
	158	sector_t size; /* used size of component devices */
	159	sector_t array_sectors; /* exported array size */
	160	__u64 events;
	161
	162	char uuid[16];
	163
	164	/* If the array is being reshaped, we need to record the
	165	* new shape and an indication of where we are up to.
	166	* This is written to the superblock.
	167	* If reshape_position is MaxSector, then no reshape is happening (yet).
	168	*/
	169	sector_t reshape_position;
	170	int delta_disks, new_level, new_layout, new_chunk;
	171
	172	struct mdk_thread_s thread; / management thread */
	173	struct mdk_thread_s sync_thread; / doing resync or reconstruct */
	174	sector_t curr_resync; /* last block scheduled */
	175	unsigned long resync_mark; /* a recent timestamp */
	176	sector_t resync_mark_cnt;/* blocks written at resync_mark */
	177	sector_t curr_mark_cnt; /* blocks scheduled now */
	178
	179	sector_t resync_max_sectors; /* may be set by personality */
	180
	181	sector_t resync_mismatches; /* count of sectors where
	182	* parity/replica mismatch found
	183	*/
	184
	185	/* allow user-space to request suspension of IO to regions of the array */
	186	sector_t suspend_lo;
	187	sector_t suspend_hi;
	188	/* if zero, use the system-wide default */
	189	int sync_speed_min;
	190	int sync_speed_max;
	191
	192	/* resync even though the same disks are shared among md-devices */
	193	int parallel_resync;
	194
	195	int ok_start_degraded;
	196	/* recovery/resync flags
	197	* NEEDED: we might need to start a resync/recover
	198	* RUNNING: a thread is running, or about to be started
	199	* SYNC: actually doing a resync, not a recovery
	200	* RECOVER: doing recovery, or need to try it.
	201	* INTR: resync needs to be aborted for some reason
	202	* DONE: thread is done and is waiting to be reaped
	203	* REQUEST: user-space has requested a sync (used with SYNC)
	204	* CHECK: user-space request for for check-only, no repair
	205	* RESHAPE: A reshape is happening
	206	*
	207	* If neither SYNC or RESHAPE are set, then it is a recovery.
	208	*/
	209	#define MD_RECOVERY_RUNNING 0
	210	#define MD_RECOVERY_SYNC 1
	211	#define MD_RECOVERY_RECOVER 2
	212	#define MD_RECOVERY_INTR 3
	213	#define MD_RECOVERY_DONE 4
	214	#define MD_RECOVERY_NEEDED 5
	215	#define MD_RECOVERY_REQUESTED 6
	216	#define MD_RECOVERY_CHECK 7
	217	#define MD_RECOVERY_RESHAPE 8
	218	#define MD_RECOVERY_FROZEN 9
	219
	220	unsigned long recovery;
	221	int recovery_disabled; /* if we detect that recovery
	222	* will always fail, set this
	223	* so we don't loop trying */
	224
	225	int in_sync; /* know to not need resync */
	226	struct mutex reconfig_mutex;
	227	atomic_t active; /* general refcount */
	228	atomic_t openers; /* number of active opens */
	229
	230	int changed; /* true if we might need to reread partition info */
	231	int degraded; /* whether md should consider
	232	* adding a spare
	233	*/
	234	int barriers_work; /* initialised to true, cleared as soon
	235	* as a barrier request to slave
	236	* fails. Only supported
	237	*/
	238	struct bio biolist; / bios that need to be retried
	239	* because BIO_RW_BARRIER is not supported
	240	*/
	241
	242	atomic_t recovery_active; /* blocks scheduled, but not written */
	243	wait_queue_head_t recovery_wait;
	244	sector_t recovery_cp;
	245	sector_t resync_min; /* user requested sync
	246	* starts here */
	247	sector_t resync_max; /* resync should pause
	248	* when it gets here */
	249
	250	struct sysfs_dirent sysfs_state; / handle for 'array_state'
	251	* file in sysfs.
	252	*/
	253	struct sysfs_dirent sysfs_action; / handle for 'sync_action' */
	254
	255	struct work_struct del_work; /* used for delayed sysfs removal */
	256
	257	spinlock_t write_lock;
	258	wait_queue_head_t sb_wait; /* for waiting on superblock updates */
	259	atomic_t pending_writes; /* number of active superblock writes */
	260
	261	unsigned int safemode; /* if set, update "clean" superblock
	262	* when no writes pending.
	263	*/
	264	unsigned int safemode_delay;
	265	struct timer_list safemode_timer;
	266	atomic_t writes_pending;
	267	struct request_queue queue; / for plugging ... */
	268
	269	atomic_t write_behind; /* outstanding async IO */
	270	unsigned int max_write_behind; /* 0 = sync */
	271
	272	struct bitmap bitmap; / the bitmap for the device */
	273	struct file bitmap_file; / the bitmap file */
	274	long bitmap_offset; /* offset from superblock of
	275	* start of bitmap. May be
	276	* negative, but not '0'
	277	*/
	278	long default_bitmap_offset; /* this is the offset to use when
	279	* hot-adding a bitmap. It should
	280	* eventually be settable by sysfs.
	281	*/
	282
	283	struct list_head all_mddevs;
	284	};
	285
	286
	287	static inline void rdev_dec_pending(mdk_rdev_t rdev, mddev_t mddev)
	288	{
	289	int faulty = test_bit(Faulty, &rdev->flags);
	290	if (atomic_dec_and_test(&rdev->nr_pending) && faulty)
	291	set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
	292	}
	293
	294	static inline void md_sync_acct(struct block_device *bdev, unsigned long nr_sectors)
	295	{
	296	atomic_add(nr_sectors, &bdev->bd_contains->bd_disk->sync_io);
	297	}
	298
	299	struct mdk_personality
	300	{
	301	char *name;
	302	int level;
	303	struct list_head list;
	304	struct module *owner;
	305	int (make_request)(struct request_queue q, struct bio *bio);
	306	int (run)(mddev_t mddev);
	307	int (stop)(mddev_t mddev);
	308	void (status)(struct seq_file seq, mddev_t *mddev);
	309	/* error_handler must set ->faulty and clear ->in_sync
	310	* if appropriate, and should abort recovery if needed
	311	*/
	312	void (error_handler)(mddev_t mddev, mdk_rdev_t *rdev);
	313	int (hot_add_disk) (mddev_t mddev, mdk_rdev_t *rdev);
	314	int (hot_remove_disk) (mddev_t mddev, int number);
	315	int (spare_active) (mddev_t mddev);
	316	sector_t (sync_request)(mddev_t mddev, sector_t sector_nr, int *skipped, int go_faster);
	317	int (resize) (mddev_t mddev, sector_t sectors);
	318	int (check_reshape) (mddev_t mddev);
	319	int (start_reshape) (mddev_t mddev);
	320	int (reconfig) (mddev_t mddev, int layout, int chunk_size);
	321	/* quiesce moves between quiescence states
	322	* 0 - fully active
	323	* 1 - no new requests allowed
	324	* others - reserved
	325	*/
	326	void (quiesce) (mddev_t mddev, int state);
	327	};
	328
	329
	330	struct md_sysfs_entry {
	331	struct attribute attr;
	332	ssize_t (show)(mddev_t , char *);
	333	ssize_t (store)(mddev_t , const char *, size_t);
	334	};
	335
	336
	337	static inline char * mdname (mddev_t * mddev)
	338	{
	339	return mddev->gendisk ? mddev->gendisk->disk_name : "mdX";
	340	}
	341
	342	/*
	343	* iterates through some rdev ringlist. It's safe to remove the
	344	* current 'rdev'. Dont touch 'tmp' though.
	345	*/
	346	#define rdev_for_each_list(rdev, tmp, head) \
	347	list_for_each_entry_safe(rdev, tmp, head, same_set)
	348
	349	/*
	350	* iterates through the 'same array disks' ringlist
	351	*/
	352	#define rdev_for_each(rdev, tmp, mddev) \
	353	list_for_each_entry_safe(rdev, tmp, &((mddev)->disks), same_set)
	354
	355	#define rdev_for_each_rcu(rdev, mddev) \
	356	list_for_each_entry_rcu(rdev, &((mddev)->disks), same_set)
	357
	358	typedef struct mdk_thread_s {
	359	void (run) (mddev_t mddev);
	360	mddev_t *mddev;
	361	wait_queue_head_t wqueue;
	362	unsigned long flags;
	363	struct task_struct *tsk;
	364	unsigned long timeout;
	365	} mdk_thread_t;
	366
	367	#define THREAD_WAKEUP 0
	368
	369	#define __wait_event_lock_irq(wq, condition, lock, cmd) \
	370	do { \
	371	wait_queue_t __wait; \
	372	init_waitqueue_entry(&__wait, current); \
	373	\
	374	add_wait_queue(&wq, &__wait); \
	375	for (;;) { \
	376	set_current_state(TASK_UNINTERRUPTIBLE); \
	377	if (condition) \
	378	break; \
	379	spin_unlock_irq(&lock); \
	380	cmd; \
	381	schedule(); \
	382	spin_lock_irq(&lock); \
	383	} \
	384	current->state = TASK_RUNNING; \
	385	remove_wait_queue(&wq, &__wait); \
	386	} while (0)
	387
	388	#define wait_event_lock_irq(wq, condition, lock, cmd) \
	389	do { \
	390	if (condition) \
	391	break; \
	392	__wait_event_lock_irq(wq, condition, lock, cmd); \
	393	} while (0)
	394
	395	static inline void safe_put_page(struct page *p)
	396	{
	397	if (p) put_page(p);
	398	}
	399
	400	#endif /* CONFIG_BLOCK */
	401	#endif
	402