[mirror_ubuntu-zesty-kernel.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 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 */
	int 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


extern int register_md_personality(struct mdk_personality *p);
extern int unregister_md_personality(struct mdk_personality *p);
extern mdk_thread_t * md_register_thread(void (*run) (mddev_t *mddev),
				mddev_t *mddev, const char *name);
extern void md_unregister_thread(mdk_thread_t *thread);
extern void md_wakeup_thread(mdk_thread_t *thread);
extern void md_check_recovery(mddev_t *mddev);
extern void md_write_start(mddev_t *mddev, struct bio *bi);
extern void md_write_end(mddev_t *mddev);
extern void md_done_sync(mddev_t *mddev, int blocks, int ok);
extern void md_error(mddev_t *mddev, mdk_rdev_t *rdev);

extern void md_super_write(mddev_t *mddev, mdk_rdev_t *rdev,
			   sector_t sector, int size, struct page *page);
extern void md_super_wait(mddev_t *mddev);
extern int sync_page_io(struct block_device *bdev, sector_t sector, int size,
			struct page *page, int rw);
extern void md_do_sync(mddev_t *mddev);
extern void md_new_event(mddev_t *mddev);
extern int md_allow_write(mddev_t *mddev);
extern void md_wait_for_blocked_rdev(mdk_rdev_t *rdev, mddev_t *mddev);
Commit	Line	Data
1da177e4 LT	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
06d91a5f N	18	/* and dm-bio-list.h is not under include/linux because.... ??? */
	19	#include "../../../drivers/md/dm-bio-list.h"
	20
9361401e DH	21	#ifdef CONFIG_BLOCK
9361401e DH	22
1da177e4	23	#define MaxSector (~(sector_t)0)
1da177e4	24
1da177e4 LT	25	typedef struct mddev_s mddev_t;
	26	typedef struct mdk_rdev_s mdk_rdev_t;
	27
1da177e4 LT	28	/*
	29	* options passed in raidrun:
	30	*/
	31
e8703fe1	32	/* Currently this must fit in an 'int' */
8932c2e0	33	#define MAX_CHUNK_SIZE (1<<30)
1da177e4	34
1da177e4 LT	35	/*
	36	* MD's 'extended' device
	37	*/
	38	struct mdk_rdev_s
	39	{
	40	struct list_head same_set; /* RAID devices within the same set */
	41
	42	sector_t size; /* Device size (in blocks) */
	43	mddev_t mddev; / RAID array if running */
eea1bf38	44	int last_events; /* IO event timestamp */
1da177e4 LT	45
	46	struct block_device bdev; / block device handle */
	47
	48	struct page *sb_page;
	49	int sb_loaded;
42543769	50	__u64 sb_events;
1da177e4	51	sector_t data_offset; /* start of data in array */
0f420358	52	sector_t sb_start; /* offset of the super block (in 512byte sectors) */
0002b271	53	int sb_size; /* bytes in the superblock */
1da177e4 LT	54	int preferred_minor; /* autorun support */
1da177e4 LT	55
86e6ffdd N	56	struct kobject kobj;
86e6ffdd N	57
1da177e4 LT	58	/* A device can be in one of three states based on two flags:
	59	* Not working: faulty==1 in_sync==0
	60	* Fully working: faulty==0 in_sync==1
	61	* Working, but not
	62	* in sync with array
	63	* faulty==0 in_sync==0
	64	*
	65	* It can never have faulty==1, in_sync==1
	66	* This reduces the burden of testing multiple flags in many cases
	67	*/
1da177e4	68
b2d444d7 N	69	unsigned long flags;
	70	#define Faulty 1 /* device is known to have a fault */
	71	#define In_sync 2 /* device is in_sync with rest of array */
8ddf9efe	72	#define WriteMostly 4 /* Avoid reading if at all possible */
a9701a30	73	#define BarriersNotsupp 5 /* BIO_RW_BARRIER is not supported */
c5d79adb N	74	#define AllReserved 6 /* If whole device is reserved for
c5d79adb N	75	* one array */
d0fae18f	76	#define AutoDetected 7 /* added by auto-detect */
6bfe0b49 DW	77	#define Blocked 8 /* An error occured on an externally
	78	* managed array, don't allow writes
	79	* until it is cleared */
52664732 NB	80	#define StateChanged 9 /* Faulty or Blocked has changed during
	81	* interrupt, so it needs to be
	82	* notified by the thread */
6bfe0b49	83	wait_queue_head_t blocked_wait;
8ddf9efe	84
1da177e4 LT	85	int desc_nr; /* descriptor index in the superblock */
1da177e4 LT	86	int raid_disk; /* role of device in array */
41158c7e N	87	int saved_raid_disk; /* role that device used to have in the
	88	* array and could again if we did a partial
	89	* resync from the bitmap
	90	*/
5fd6c1dc N	91	sector_t recovery_offset;/* If this device has been partially
	92	* recovered, this is where we were
	93	* up to.
	94	*/
1da177e4 LT	95
	96	atomic_t nr_pending; /* number of pending requests.
	97	* only maintained for arrays that
	98	* support hot removal
	99	*/
ba22dcbf N	100	atomic_t read_errors; /* number of consecutive read errors that
	101	* we have tried to ignore.
	102	*/
4dbcdc75 N	103	atomic_t corrected_errors; /* number of corrected read errors,
	104	* for reporting to userspace and storing
	105	* in superblock.
	106	*/
5792a285	107	struct work_struct del_work; /* used for delayed sysfs removal */
3c0ee63a N	108
	109	struct sysfs_dirent sysfs_state; / handle for 'state'
	110	* sysfs entry */
1da177e4 LT	111	};
1da177e4 LT	112
1da177e4 LT	113	struct mddev_s
	114	{
	115	void *private;
2604b703	116	struct mdk_personality *pers;
1da177e4 LT	117	dev_t unit;
	118	int md_minor;
	119	struct list_head disks;
850b2b42 N	120	unsigned long flags;
	121	#define MD_CHANGE_DEVS 0 /* Some device status has changed */
	122	#define MD_CHANGE_CLEAN 1 /* transition to or from 'clean' */
	123	#define MD_CHANGE_PENDING 2 /* superblock update in progress */
	124
1da177e4 LT	125	int ro;
	126
	127	struct gendisk *gendisk;
	128
eae1701f	129	struct kobject kobj;
d3374825 N	130	int hold_active;
d3374825 N	131	#define UNTIL_IOCTL 1
efeb53c0	132	#define UNTIL_STOP 2
eae1701f	133
1da177e4 LT	134	/* Superblock information */
	135	int major_version,
	136	minor_version,
	137	patch_version;
	138	int persistent;
e691063a N	139	int external; /* metadata is
	140	* managed externally */
	141	char metadata_type[17]; /* externally set*/
1da177e4 LT	142	int chunk_size;
	143	time_t ctime, utime;
	144	int level, layout;
d9d166c2	145	char clevel[16];
1da177e4 LT	146	int raid_disks;
	147	int max_disks;
	148	sector_t size; /* used size of component devices */
f233ea5c	149	sector_t array_sectors; /* exported array size */
1da177e4 LT	150	__u64 events;
	151
	152	char uuid[16];
	153
f6705578 N	154	/* If the array is being reshaped, we need to record the
	155	* new shape and an indication of where we are up to.
	156	* This is written to the superblock.
	157	* If reshape_position is MaxSector, then no reshape is happening (yet).
	158	*/
	159	sector_t reshape_position;
	160	int delta_disks, new_level, new_layout, new_chunk;
	161
1da177e4 LT	162	struct mdk_thread_s thread; / management thread */
1da177e4 LT	163	struct mdk_thread_s sync_thread; / doing resync or reconstruct */
ff4e8d9a	164	sector_t curr_resync; /* last block scheduled */
1da177e4 LT	165	unsigned long resync_mark; /* a recent timestamp */
1da177e4 LT	166	sector_t resync_mark_cnt;/* blocks written at resync_mark */
ff4e8d9a	167	sector_t curr_mark_cnt; /* blocks scheduled now */
1da177e4 LT	168
1da177e4 LT	169	sector_t resync_max_sectors; /* may be set by personality */
9d88883e N	170
	171	sector_t resync_mismatches; /* count of sectors where
	172	* parity/replica mismatch found
	173	*/
e464eafd N	174
	175	/* allow user-space to request suspension of IO to regions of the array */
	176	sector_t suspend_lo;
	177	sector_t suspend_hi;
88202a0c N	178	/* if zero, use the system-wide default */
	179	int sync_speed_min;
	180	int sync_speed_max;
	181
90b08710 BS	182	/* resync even though the same disks are shared among md-devices */
	183	int parallel_resync;
	184
6ff8d8ec	185	int ok_start_degraded;
1da177e4 LT	186	/* recovery/resync flags
	187	* NEEDED: we might need to start a resync/recover
	188	* RUNNING: a thread is running, or about to be started
	189	* SYNC: actually doing a resync, not a recovery
72a23c21	190	* RECOVER: doing recovery, or need to try it.
dfc70645	191	* INTR: resync needs to be aborted for some reason
1da177e4	192	* DONE: thread is done and is waiting to be reaped
24dd469d N	193	* REQUEST: user-space has requested a sync (used with SYNC)
24dd469d N	194	* CHECK: user-space request for for check-only, no repair
ccfcc3c1 N	195	* RESHAPE: A reshape is happening
	196	*
	197	* If neither SYNC or RESHAPE are set, then it is a recovery.
1da177e4 LT	198	*/
	199	#define MD_RECOVERY_RUNNING 0
	200	#define MD_RECOVERY_SYNC 1
72a23c21	201	#define MD_RECOVERY_RECOVER 2
1da177e4 LT	202	#define MD_RECOVERY_INTR 3
	203	#define MD_RECOVERY_DONE 4
	204	#define MD_RECOVERY_NEEDED 5
24dd469d N	205	#define MD_RECOVERY_REQUESTED 6
24dd469d N	206	#define MD_RECOVERY_CHECK 7
ccfcc3c1	207	#define MD_RECOVERY_RESHAPE 8
5fd6c1dc N	208	#define MD_RECOVERY_FROZEN 9
5fd6c1dc N	209
1da177e4	210	unsigned long recovery;
4044ba58 N	211	int recovery_disabled; /* if we detect that recovery
	212	* will always fail, set this
	213	* so we don't loop trying */
1da177e4 LT	214
1da177e4 LT	215	int in_sync; /* know to not need resync */
df5b89b3	216	struct mutex reconfig_mutex;
f2ea68cf N	217	atomic_t active; /* general refcount */
f2ea68cf N	218	atomic_t openers; /* number of active opens */
1da177e4	219
44ce6294	220	int changed; /* true if we might need to reread partition info */
1da177e4 LT	221	int degraded; /* whether md should consider
	222	* adding a spare
	223	*/
a9701a30 N	224	int barriers_work; /* initialised to true, cleared as soon
	225	* as a barrier request to slave
	226	* fails. Only supported
	227	*/
	228	struct bio biolist; / bios that need to be retried
	229	* because BIO_RW_BARRIER is not supported
	230	*/
1da177e4 LT	231
	232	atomic_t recovery_active; /* blocks scheduled, but not written */
	233	wait_queue_head_t recovery_wait;
	234	sector_t recovery_cp;
5e96ee65 NB	235	sector_t resync_min; /* user requested sync
5e96ee65 NB	236	* starts here */
c6207277 N	237	sector_t resync_max; /* resync should pause
c6207277 N	238	* when it gets here */
06d91a5f	239
b62b7590 N	240	struct sysfs_dirent sysfs_state; / handle for 'array_state'
	241	* file in sysfs.
	242	*/
0c3573f1	243	struct sysfs_dirent sysfs_action; / handle for 'sync_action' */
b62b7590	244
d3374825 N	245	struct work_struct del_work; /* used for delayed sysfs removal */
d3374825 N	246
06d91a5f	247	spinlock_t write_lock;
3d310eb7	248	wait_queue_head_t sb_wait; /* for waiting on superblock updates */
7bfa19f2	249	atomic_t pending_writes; /* number of active superblock writes */
06d91a5f	250
1da177e4 LT	251	unsigned int safemode; /* if set, update "clean" superblock
	252	* when no writes pending.
	253	*/
	254	unsigned int safemode_delay;
	255	struct timer_list safemode_timer;
	256	atomic_t writes_pending;
165125e1	257	struct request_queue queue; / for plugging ... */
1da177e4	258
4b6d287f N	259	atomic_t write_behind; /* outstanding async IO */
	260	unsigned int max_write_behind; /* 0 = sync */
	261
32a7627c N	262	struct bitmap bitmap; / the bitmap for the device */
32a7627c N	263	struct file bitmap_file; / the bitmap file */
a654b9d8 N	264	long bitmap_offset; /* offset from superblock of
	265	* start of bitmap. May be
	266	* negative, but not '0'
	267	*/
36fa3063 N	268	long default_bitmap_offset; /* this is the offset to use when
	269	* hot-adding a bitmap. It should
	270	* eventually be settable by sysfs.
	271	*/
32a7627c	272
1da177e4 LT	273	struct list_head all_mddevs;
	274	};
	275
	276
	277	static inline void rdev_dec_pending(mdk_rdev_t rdev, mddev_t mddev)
	278	{
b2d444d7	279	int faulty = test_bit(Faulty, &rdev->flags);
1da177e4 LT	280	if (atomic_dec_and_test(&rdev->nr_pending) && faulty)
	281	set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
	282	}
	283
	284	static inline void md_sync_acct(struct block_device *bdev, unsigned long nr_sectors)
	285	{
	286	atomic_add(nr_sectors, &bdev->bd_contains->bd_disk->sync_io);
	287	}
	288
2604b703	289	struct mdk_personality
1da177e4 LT	290	{
1da177e4 LT	291	char *name;
2604b703 N	292	int level;
2604b703 N	293	struct list_head list;
1da177e4	294	struct module *owner;
165125e1	295	int (make_request)(struct request_queue q, struct bio *bio);
1da177e4 LT	296	int (run)(mddev_t mddev);
	297	int (stop)(mddev_t mddev);
	298	void (status)(struct seq_file seq, mddev_t *mddev);
	299	/* error_handler must set ->faulty and clear ->in_sync
	300	* if appropriate, and should abort recovery if needed
	301	*/
	302	void (error_handler)(mddev_t mddev, mdk_rdev_t *rdev);
	303	int (hot_add_disk) (mddev_t mddev, mdk_rdev_t *rdev);
	304	int (hot_remove_disk) (mddev_t mddev, int number);
	305	int (spare_active) (mddev_t mddev);
57afd89f	306	sector_t (sync_request)(mddev_t mddev, sector_t sector_nr, int *skipped, int go_faster);
1da177e4	307	int (resize) (mddev_t mddev, sector_t sectors);
63c70c4f N	308	int (check_reshape) (mddev_t mddev);
63c70c4f N	309	int (start_reshape) (mddev_t mddev);
1da177e4	310	int (reconfig) (mddev_t mddev, int layout, int chunk_size);
36fa3063 N	311	/* quiesce moves between quiescence states
	312	* 0 - fully active
	313	* 1 - no new requests allowed
	314	* others - reserved
	315	*/
	316	void (quiesce) (mddev_t mddev, int state);
1da177e4 LT	317	};
	318
	319
007583c9 N	320	struct md_sysfs_entry {
	321	struct attribute attr;
	322	ssize_t (show)(mddev_t , char *);
	323	ssize_t (store)(mddev_t , const char *, size_t);
	324	};
	325
	326
1da177e4 LT	327	static inline char * mdname (mddev_t * mddev)
	328	{
	329	return mddev->gendisk ? mddev->gendisk->disk_name : "mdX";
	330	}
	331
1da177e4 LT	332	/*
	333	* iterates through some rdev ringlist. It's safe to remove the
	334	* current 'rdev'. Dont touch 'tmp' though.
	335	*/
159ec1fc CR	336	#define rdev_for_each_list(rdev, tmp, head) \
	337	list_for_each_entry_safe(rdev, tmp, head, same_set)
	338
1da177e4 LT	339	/*
	340	* iterates through the 'same array disks' ringlist
	341	*/
d089c6af	342	#define rdev_for_each(rdev, tmp, mddev) \
159ec1fc	343	list_for_each_entry_safe(rdev, tmp, &((mddev)->disks), same_set)
1da177e4	344
4b80991c N	345	#define rdev_for_each_rcu(rdev, mddev) \
	346	list_for_each_entry_rcu(rdev, &((mddev)->disks), same_set)
	347
1da177e4 LT	348	typedef struct mdk_thread_s {
	349	void (run) (mddev_t mddev);
	350	mddev_t *mddev;
	351	wait_queue_head_t wqueue;
	352	unsigned long flags;
1da177e4	353	struct task_struct *tsk;
32a7627c	354	unsigned long timeout;
1da177e4 LT	355	} mdk_thread_t;
	356
	357	#define THREAD_WAKEUP 0
	358
	359	#define __wait_event_lock_irq(wq, condition, lock, cmd) \
	360	do { \
	361	wait_queue_t __wait; \
	362	init_waitqueue_entry(&__wait, current); \
	363	\
	364	add_wait_queue(&wq, &__wait); \
	365	for (;;) { \
	366	set_current_state(TASK_UNINTERRUPTIBLE); \
	367	if (condition) \
	368	break; \
	369	spin_unlock_irq(&lock); \
	370	cmd; \
	371	schedule(); \
	372	spin_lock_irq(&lock); \
	373	} \
	374	current->state = TASK_RUNNING; \
	375	remove_wait_queue(&wq, &__wait); \
	376	} while (0)
	377
	378	#define wait_event_lock_irq(wq, condition, lock, cmd) \
	379	do { \
	380	if (condition) \
	381	break; \
	382	__wait_event_lock_irq(wq, condition, lock, cmd); \
	383	} while (0)
	384
1345b1d8 N	385	static inline void safe_put_page(struct page *p)
	386	{
	387	if (p) put_page(p);
	388	}
	389
9361401e	390	#endif /* CONFIG_BLOCK */
1da177e4 LT	391	#endif
1da177e4 LT	392
92022950 N	393
	394	extern int register_md_personality(struct mdk_personality *p);
	395	extern int unregister_md_personality(struct mdk_personality *p);
	396	extern mdk_thread_t * md_register_thread(void (run) (mddev_t mddev),
	397	mddev_t mddev, const char name);
	398	extern void md_unregister_thread(mdk_thread_t *thread);
	399	extern void md_wakeup_thread(mdk_thread_t *thread);
	400	extern void md_check_recovery(mddev_t *mddev);
	401	extern void md_write_start(mddev_t mddev, struct bio bi);
	402	extern void md_write_end(mddev_t *mddev);
	403	extern void md_done_sync(mddev_t *mddev, int blocks, int ok);
	404	extern void md_error(mddev_t mddev, mdk_rdev_t rdev);
	405
	406	extern void md_super_write(mddev_t mddev, mdk_rdev_t rdev,
	407	sector_t sector, int size, struct page *page);
	408	extern void md_super_wait(mddev_t *mddev);
	409	extern int sync_page_io(struct block_device *bdev, sector_t sector, int size,
	410	struct page *page, int rw);
	411	extern void md_do_sync(mddev_t *mddev);
	412	extern void md_new_event(mddev_t *mddev);
	413	extern int md_allow_write(mddev_t *mddev);
	414	extern void md_wait_for_blocked_rdev(mdk_rdev_t rdev, mddev_t mddev);