[deliverable/linux.git] / fs / xfs / linux-2.6 / xfs_buf.h

/*
 * Copyright (c) 2000-2005 Silicon Graphics, Inc.
 * 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 as
 * published by the Free Software Foundation.
 *
 * This program is distributed in the hope that it would 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 the Free Software Foundation,
 * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
 */
#ifndef __XFS_BUF_H__
#define __XFS_BUF_H__

#include <linux/list.h>
#include <linux/types.h>
#include <linux/spinlock.h>
#include <asm/system.h>
#include <linux/mm.h>
#include <linux/fs.h>
#include <linux/buffer_head.h>
#include <linux/uio.h>

/*
 *	Base types
 */

#define XFS_BUF_DADDR_NULL	((xfs_daddr_t) (-1LL))

#define xfs_buf_ctob(pp)	((pp) * PAGE_CACHE_SIZE)
#define xfs_buf_btoc(dd)	(((dd) + PAGE_CACHE_SIZE-1) >> PAGE_CACHE_SHIFT)
#define xfs_buf_btoct(dd)	((dd) >> PAGE_CACHE_SHIFT)
#define xfs_buf_poff(aa)	((aa) & ~PAGE_CACHE_MASK)

typedef enum {
	XBRW_READ = 1,			/* transfer into target memory */
	XBRW_WRITE = 2,			/* transfer from target memory */
	XBRW_ZERO = 3,			/* Zero target memory */
} xfs_buf_rw_t;

#define XBF_READ	(1 << 0) /* buffer intended for reading from device */
#define XBF_WRITE	(1 << 1) /* buffer intended for writing to device */
#define XBF_MAPPED	(1 << 2) /* buffer mapped (b_addr valid) */
#define XBF_ASYNC	(1 << 4) /* initiator will not wait for completion */
#define XBF_DONE	(1 << 5) /* all pages in the buffer uptodate */
#define XBF_DELWRI	(1 << 6) /* buffer has dirty pages */
#define XBF_STALE	(1 << 7) /* buffer has been staled, do not find it */
#define XBF_ORDERED	(1 << 11)/* use ordered writes */
#define XBF_READ_AHEAD	(1 << 12)/* asynchronous read-ahead */
#define XBF_LOG_BUFFER	(1 << 13)/* this is a buffer used for the log */

/* flags used only as arguments to access routines */
#define XBF_LOCK	(1 << 14)/* lock requested */
#define XBF_TRYLOCK	(1 << 15)/* lock requested, but do not wait */
#define XBF_DONT_BLOCK	(1 << 16)/* do not block in current thread */

/* flags used only internally */
#define _XBF_PAGE_CACHE	(1 << 17)/* backed by pagecache */
#define _XBF_PAGES	(1 << 18)/* backed by refcounted pages */
#define	_XBF_RUN_QUEUES	(1 << 19)/* run block device task queue	*/
#define _XBF_DELWRI_Q	(1 << 21)/* buffer on delwri queue */

/*
 * Special flag for supporting metadata blocks smaller than a FSB.
 *
 * In this case we can have multiple xfs_buf_t on a single page and
 * need to lock out concurrent xfs_buf_t readers as they only
 * serialise access to the buffer.
 *
 * If the FSB size >= PAGE_CACHE_SIZE case, we have no serialisation
 * between reads of the page. Hence we can have one thread read the
 * page and modify it, but then race with another thread that thinks
 * the page is not up-to-date and hence reads it again.
 *
 * The result is that the first modifcation to the page is lost.
 * This sort of AGF/AGI reading race can happen when unlinking inodes
 * that require truncation and results in the AGI unlinked list
 * modifications being lost.
 */
#define _XBF_PAGE_LOCKED	(1 << 22)

typedef unsigned int xfs_buf_flags_t;

#define XFS_BUF_FLAGS \
	{ XBF_READ,		"READ" }, \
	{ XBF_WRITE,		"WRITE" }, \
	{ XBF_MAPPED,		"MAPPED" }, \
	{ XBF_ASYNC,		"ASYNC" }, \
	{ XBF_DONE,		"DONE" }, \
	{ XBF_DELWRI,		"DELWRI" }, \
	{ XBF_STALE,		"STALE" }, \
	{ XBF_ORDERED,		"ORDERED" }, \
	{ XBF_READ_AHEAD,	"READ_AHEAD" }, \
	{ XBF_LOCK,		"LOCK" },  	/* should never be set */\
	{ XBF_TRYLOCK,		"TRYLOCK" }, 	/* ditto */\
	{ XBF_DONT_BLOCK,	"DONT_BLOCK" },	/* ditto */\
	{ _XBF_PAGE_CACHE,	"PAGE_CACHE" }, \
	{ _XBF_PAGES,		"PAGES" }, \
	{ _XBF_RUN_QUEUES,	"RUN_QUEUES" }, \
	{ _XBF_DELWRI_Q,	"DELWRI_Q" }, \
	{ _XBF_PAGE_LOCKED,	"PAGE_LOCKED" }


typedef enum {
	XBT_FORCE_SLEEP = 0,
	XBT_FORCE_FLUSH = 1,
} xfs_buftarg_flags_t;

typedef struct xfs_bufhash {
	struct list_head	bh_list;
	spinlock_t		bh_lock;
} xfs_bufhash_t;

typedef struct xfs_buftarg {
	dev_t			bt_dev;
	struct block_device	*bt_bdev;
	struct address_space	*bt_mapping;
	struct xfs_mount	*bt_mount;
	unsigned int		bt_bsize;
	unsigned int		bt_sshift;
	size_t			bt_smask;

	/* per device delwri queue */
	struct task_struct	*bt_task;
	struct list_head	bt_delwrite_queue;
	spinlock_t		bt_delwrite_lock;
	unsigned long		bt_flags;

	/* LRU control structures */
	struct shrinker		bt_shrinker;
	struct list_head	bt_lru;
	spinlock_t		bt_lru_lock;
	unsigned int		bt_lru_nr;
} xfs_buftarg_t;

/*
 *	xfs_buf_t:  Buffer structure for pagecache-based buffers
 *
 * This buffer structure is used by the pagecache buffer management routines
 * to refer to an assembly of pages forming a logical buffer.
 *
 * The buffer structure is used on a temporary basis only, and discarded when
 * released.  The real data storage is recorded in the pagecache. Buffers are
 * hashed to the block device on which the file system resides.
 */

struct xfs_buf;
typedef void (*xfs_buf_iodone_t)(struct xfs_buf *);

#define XB_PAGES	2

typedef struct xfs_buf {
	/*
	 * first cacheline holds all the fields needed for an uncontended cache
	 * hit to be fully processed. The semaphore straddles the cacheline
	 * boundary, but the counter and lock sits on the first cacheline,
	 * which is the only bit that is touched if we hit the semaphore
	 * fast-path on locking.
	 */
	struct rb_node		b_rbnode;	/* rbtree node */
	xfs_off_t		b_file_offset;	/* offset in file */
	size_t			b_buffer_length;/* size of buffer in bytes */
	atomic_t		b_hold;		/* reference count */
	atomic_t		b_lru_ref;	/* lru reclaim ref count */
	xfs_buf_flags_t		b_flags;	/* status flags */
	struct semaphore	b_sema;		/* semaphore for lockables */

	struct list_head	b_lru;		/* lru list */
	wait_queue_head_t	b_waiters;	/* unpin waiters */
	struct list_head	b_list;
	struct xfs_perag	*b_pag;		/* contains rbtree root */
	xfs_buftarg_t		*b_target;	/* buffer target (device) */
	xfs_daddr_t		b_bn;		/* block number for I/O */
	size_t			b_count_desired;/* desired transfer size */
	void			*b_addr;	/* virtual address of buffer */
	struct work_struct	b_iodone_work;
	xfs_buf_iodone_t	b_iodone;	/* I/O completion function */
	struct completion	b_iowait;	/* queue for I/O waiters */
	void			*b_fspriv;
	void			*b_fspriv2;
	struct page		**b_pages;	/* array of page pointers */
	struct page		*b_page_array[XB_PAGES]; /* inline pages */
	unsigned long		b_queuetime;	/* time buffer was queued */
	atomic_t		b_pin_count;	/* pin count */
	atomic_t		b_io_remaining;	/* #outstanding I/O requests */
	unsigned int		b_page_count;	/* size of page array */
	unsigned int		b_offset;	/* page offset in first page */
	unsigned short		b_error;	/* error code on I/O */
#ifdef XFS_BUF_LOCK_TRACKING
	int			b_last_holder;
#endif
} xfs_buf_t;


/* Finding and Reading Buffers */
extern xfs_buf_t *_xfs_buf_find(xfs_buftarg_t *, xfs_off_t, size_t,
				xfs_buf_flags_t, xfs_buf_t *);
#define xfs_incore(buftarg,blkno,len,lockit) \
	_xfs_buf_find(buftarg, blkno ,len, lockit, NULL)

extern xfs_buf_t *xfs_buf_get(xfs_buftarg_t *, xfs_off_t, size_t,
				xfs_buf_flags_t);
extern xfs_buf_t *xfs_buf_read(xfs_buftarg_t *, xfs_off_t, size_t,
				xfs_buf_flags_t);

extern xfs_buf_t *xfs_buf_get_empty(size_t, xfs_buftarg_t *);
extern xfs_buf_t *xfs_buf_get_uncached(struct xfs_buftarg *, size_t, int);
extern int xfs_buf_associate_memory(xfs_buf_t *, void *, size_t);
extern void xfs_buf_hold(xfs_buf_t *);
extern void xfs_buf_readahead(xfs_buftarg_t *, xfs_off_t, size_t);
struct xfs_buf *xfs_buf_read_uncached(struct xfs_mount *mp,
				struct xfs_buftarg *target,
				xfs_daddr_t daddr, size_t length, int flags);

/* Releasing Buffers */
extern void xfs_buf_free(xfs_buf_t *);
extern void xfs_buf_rele(xfs_buf_t *);

/* Locking and Unlocking Buffers */
extern int xfs_buf_cond_lock(xfs_buf_t *);
extern int xfs_buf_lock_value(xfs_buf_t *);
extern void xfs_buf_lock(xfs_buf_t *);
extern void xfs_buf_unlock(xfs_buf_t *);

/* Buffer Read and Write Routines */
extern int xfs_bwrite(struct xfs_mount *mp, struct xfs_buf *bp);
extern void xfs_bdwrite(void *mp, xfs_buf_t *bp);

extern void xfsbdstrat(struct xfs_mount *, struct xfs_buf *);
extern int xfs_bdstrat_cb(struct xfs_buf *);

extern void xfs_buf_ioend(xfs_buf_t *,	int);
extern void xfs_buf_ioerror(xfs_buf_t *, int);
extern int xfs_buf_iorequest(xfs_buf_t *);
extern int xfs_buf_iowait(xfs_buf_t *);
extern void xfs_buf_iomove(xfs_buf_t *, size_t, size_t, void *,
				xfs_buf_rw_t);
#define xfs_buf_zero(bp, off, len) \
	    xfs_buf_iomove((bp), (off), (len), NULL, XBRW_ZERO)

static inline int xfs_buf_geterror(xfs_buf_t *bp)
{
	return bp ? bp->b_error : ENOMEM;
}

/* Buffer Utility Routines */
extern xfs_caddr_t xfs_buf_offset(xfs_buf_t *, size_t);

/* Delayed Write Buffer Routines */
extern void xfs_buf_delwri_dequeue(xfs_buf_t *);
extern void xfs_buf_delwri_promote(xfs_buf_t *);

/* Buffer Daemon Setup Routines */
extern int xfs_buf_init(void);
extern void xfs_buf_terminate(void);

#define xfs_buf_target_name(target)	\
	({ char __b[BDEVNAME_SIZE]; bdevname((target)->bt_bdev, __b); __b; })


#define XFS_BUF_BFLAGS(bp)	((bp)->b_flags)
#define XFS_BUF_ZEROFLAGS(bp)	((bp)->b_flags &= \
		~(XBF_READ|XBF_WRITE|XBF_ASYNC|XBF_DELWRI|XBF_ORDERED))

void xfs_buf_stale(struct xfs_buf *bp);
#define XFS_BUF_STALE(bp)	xfs_buf_stale(bp);
#define XFS_BUF_UNSTALE(bp)	((bp)->b_flags &= ~XBF_STALE)
#define XFS_BUF_ISSTALE(bp)	((bp)->b_flags & XBF_STALE)
#define XFS_BUF_SUPER_STALE(bp)	do {				\
					XFS_BUF_STALE(bp);	\
					xfs_buf_delwri_dequeue(bp);	\
					XFS_BUF_DONE(bp);	\
				} while (0)

#define XFS_BUF_DELAYWRITE(bp)		((bp)->b_flags |= XBF_DELWRI)
#define XFS_BUF_UNDELAYWRITE(bp)	xfs_buf_delwri_dequeue(bp)
#define XFS_BUF_ISDELAYWRITE(bp)	((bp)->b_flags & XBF_DELWRI)

#define XFS_BUF_ERROR(bp,no)	xfs_buf_ioerror(bp,no)
#define XFS_BUF_GETERROR(bp)	xfs_buf_geterror(bp)
#define XFS_BUF_ISERROR(bp)	(xfs_buf_geterror(bp) ? 1 : 0)

#define XFS_BUF_DONE(bp)	((bp)->b_flags |= XBF_DONE)
#define XFS_BUF_UNDONE(bp)	((bp)->b_flags &= ~XBF_DONE)
#define XFS_BUF_ISDONE(bp)	((bp)->b_flags & XBF_DONE)

#define XFS_BUF_BUSY(bp)	do { } while (0)
#define XFS_BUF_UNBUSY(bp)	do { } while (0)
#define XFS_BUF_ISBUSY(bp)	(1)

#define XFS_BUF_ASYNC(bp)	((bp)->b_flags |= XBF_ASYNC)
#define XFS_BUF_UNASYNC(bp)	((bp)->b_flags &= ~XBF_ASYNC)
#define XFS_BUF_ISASYNC(bp)	((bp)->b_flags & XBF_ASYNC)

#define XFS_BUF_ORDERED(bp)	((bp)->b_flags |= XBF_ORDERED)
#define XFS_BUF_UNORDERED(bp)	((bp)->b_flags &= ~XBF_ORDERED)
#define XFS_BUF_ISORDERED(bp)	((bp)->b_flags & XBF_ORDERED)

#define XFS_BUF_HOLD(bp)	xfs_buf_hold(bp)
#define XFS_BUF_READ(bp)	((bp)->b_flags |= XBF_READ)
#define XFS_BUF_UNREAD(bp)	((bp)->b_flags &= ~XBF_READ)
#define XFS_BUF_ISREAD(bp)	((bp)->b_flags & XBF_READ)

#define XFS_BUF_WRITE(bp)	((bp)->b_flags |= XBF_WRITE)
#define XFS_BUF_UNWRITE(bp)	((bp)->b_flags &= ~XBF_WRITE)
#define XFS_BUF_ISWRITE(bp)	((bp)->b_flags & XBF_WRITE)

#define XFS_BUF_IODONE_FUNC(bp)			((bp)->b_iodone)
#define XFS_BUF_SET_IODONE_FUNC(bp, func)	((bp)->b_iodone = (func))
#define XFS_BUF_CLR_IODONE_FUNC(bp)		((bp)->b_iodone = NULL)

#define XFS_BUF_FSPRIVATE(bp, type)		((type)(bp)->b_fspriv)
#define XFS_BUF_SET_FSPRIVATE(bp, val)		((bp)->b_fspriv = (void*)(val))
#define XFS_BUF_FSPRIVATE2(bp, type)		((type)(bp)->b_fspriv2)
#define XFS_BUF_SET_FSPRIVATE2(bp, val)		((bp)->b_fspriv2 = (void*)(val))
#define XFS_BUF_SET_START(bp)			do { } while (0)

#define XFS_BUF_PTR(bp)			(xfs_caddr_t)((bp)->b_addr)
#define XFS_BUF_SET_PTR(bp, val, cnt)	xfs_buf_associate_memory(bp, val, cnt)
#define XFS_BUF_ADDR(bp)		((bp)->b_bn)
#define XFS_BUF_SET_ADDR(bp, bno)	((bp)->b_bn = (xfs_daddr_t)(bno))
#define XFS_BUF_OFFSET(bp)		((bp)->b_file_offset)
#define XFS_BUF_SET_OFFSET(bp, off)	((bp)->b_file_offset = (off))
#define XFS_BUF_COUNT(bp)		((bp)->b_count_desired)
#define XFS_BUF_SET_COUNT(bp, cnt)	((bp)->b_count_desired = (cnt))
#define XFS_BUF_SIZE(bp)		((bp)->b_buffer_length)
#define XFS_BUF_SET_SIZE(bp, cnt)	((bp)->b_buffer_length = (cnt))

static inline void
xfs_buf_set_ref(
	struct xfs_buf	*bp,
	int		lru_ref)
{
	atomic_set(&bp->b_lru_ref, lru_ref);
}
#define XFS_BUF_SET_VTYPE_REF(bp, type, ref)	xfs_buf_set_ref(bp, ref)
#define XFS_BUF_SET_VTYPE(bp, type)		do { } while (0)

#define XFS_BUF_ISPINNED(bp)	atomic_read(&((bp)->b_pin_count))

#define XFS_BUF_VALUSEMA(bp)	xfs_buf_lock_value(bp)
#define XFS_BUF_CPSEMA(bp)	(xfs_buf_cond_lock(bp) == 0)
#define XFS_BUF_VSEMA(bp)	xfs_buf_unlock(bp)
#define XFS_BUF_PSEMA(bp,x)	xfs_buf_lock(bp)
#define XFS_BUF_FINISH_IOWAIT(bp)	complete(&bp->b_iowait);

#define XFS_BUF_SET_TARGET(bp, target)	((bp)->b_target = (target))
#define XFS_BUF_TARGET(bp)		((bp)->b_target)
#define XFS_BUFTARG_NAME(target)	xfs_buf_target_name(target)

static inline void xfs_buf_relse(xfs_buf_t *bp)
{
	xfs_buf_unlock(bp);
	xfs_buf_rele(bp);
}

/*
 *	Handling of buftargs.
 */
extern xfs_buftarg_t *xfs_alloc_buftarg(struct xfs_mount *,
			struct block_device *, int, const char *);
extern void xfs_free_buftarg(struct xfs_mount *, struct xfs_buftarg *);
extern void xfs_wait_buftarg(xfs_buftarg_t *);
extern int xfs_setsize_buftarg(xfs_buftarg_t *, unsigned int, unsigned int);
extern int xfs_flush_buftarg(xfs_buftarg_t *, int);

#ifdef CONFIG_KDB_MODULES
extern struct list_head *xfs_get_buftarg_list(void);
#endif

#define xfs_getsize_buftarg(buftarg)	block_size((buftarg)->bt_bdev)
#define xfs_readonly_buftarg(buftarg)	bdev_read_only((buftarg)->bt_bdev)

#define xfs_binval(buftarg)		xfs_flush_buftarg(buftarg, 1)
#define XFS_bflush(buftarg)		xfs_flush_buftarg(buftarg, 1)

#endif	/* __XFS_BUF_H__ */
Commit	Line	Data
1da177e4	1	/*
7b718769 NS	2	* Copyright (c) 2000-2005 Silicon Graphics, Inc.
7b718769 NS	3	* All Rights Reserved.
1da177e4	4	*
7b718769 NS	5	* This program is free software; you can redistribute it and/or
7b718769 NS	6	* modify it under the terms of the GNU General Public License as
1da177e4 LT	7	* published by the Free Software Foundation.
1da177e4 LT	8	*
7b718769 NS	9	* This program is distributed in the hope that it would be useful,
	10	* but WITHOUT ANY WARRANTY; without even the implied warranty of
	11	* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
	12	* GNU General Public License for more details.
1da177e4	13	*
7b718769 NS	14	* You should have received a copy of the GNU General Public License
	15	* along with this program; if not, write the Free Software Foundation,
	16	* Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
1da177e4	17	*/
1da177e4 LT	18	#ifndef __XFS_BUF_H__
	19	#define __XFS_BUF_H__
	20
1da177e4 LT	21	#include <linux/list.h>
	22	#include <linux/types.h>
	23	#include <linux/spinlock.h>
	24	#include <asm/system.h>
	25	#include <linux/mm.h>
	26	#include <linux/fs.h>
	27	#include <linux/buffer_head.h>
	28	#include <linux/uio.h>
	29
	30	/*
	31	* Base types
	32	*/
	33
ce8e922c NS	34	#define XFS_BUF_DADDR_NULL ((xfs_daddr_t) (-1LL))
	35
	36	#define xfs_buf_ctob(pp) ((pp) * PAGE_CACHE_SIZE)
	37	#define xfs_buf_btoc(dd) (((dd) + PAGE_CACHE_SIZE-1) >> PAGE_CACHE_SHIFT)
	38	#define xfs_buf_btoct(dd) ((dd) >> PAGE_CACHE_SHIFT)
	39	#define xfs_buf_poff(aa) ((aa) & ~PAGE_CACHE_MASK)
	40
	41	typedef enum {
	42	XBRW_READ = 1, /* transfer into target memory */
	43	XBRW_WRITE = 2, /* transfer from target memory */
	44	XBRW_ZERO = 3, /* Zero target memory */
	45	} xfs_buf_rw_t;
	46
807cbbdb CH	47	#define XBF_READ (1 << 0) /* buffer intended for reading from device */
	48	#define XBF_WRITE (1 << 1) /* buffer intended for writing to device */
	49	#define XBF_MAPPED (1 << 2) /* buffer mapped (b_addr valid) */
	50	#define XBF_ASYNC (1 << 4) /* initiator will not wait for completion */
	51	#define XBF_DONE (1 << 5) /* all pages in the buffer uptodate */
	52	#define XBF_DELWRI (1 << 6) /* buffer has dirty pages */
	53	#define XBF_STALE (1 << 7) /* buffer has been staled, do not find it */
807cbbdb CH	54	#define XBF_ORDERED (1 << 11)/* use ordered writes */
	55	#define XBF_READ_AHEAD (1 << 12)/* asynchronous read-ahead */
	56	#define XBF_LOG_BUFFER (1 << 13)/* this is a buffer used for the log */
1da177e4	57
807cbbdb CH	58	/* flags used only as arguments to access routines */
	59	#define XBF_LOCK (1 << 14)/* lock requested */
	60	#define XBF_TRYLOCK (1 << 15)/* lock requested, but do not wait */
	61	#define XBF_DONT_BLOCK (1 << 16)/* do not block in current thread */
1da177e4	62
807cbbdb CH	63	/* flags used only internally */
	64	#define _XBF_PAGE_CACHE (1 << 17)/* backed by pagecache */
	65	#define _XBF_PAGES (1 << 18)/* backed by refcounted pages */
	66	#define _XBF_RUN_QUEUES (1 << 19)/* run block device task queue */
	67	#define _XBF_DELWRI_Q (1 << 21)/* buffer on delwri queue */
6ab455ee	68
807cbbdb CH	69	/*
	70	* Special flag for supporting metadata blocks smaller than a FSB.
	71	*
	72	* In this case we can have multiple xfs_buf_t on a single page and
	73	* need to lock out concurrent xfs_buf_t readers as they only
	74	* serialise access to the buffer.
	75	*
	76	* If the FSB size >= PAGE_CACHE_SIZE case, we have no serialisation
	77	* between reads of the page. Hence we can have one thread read the
	78	* page and modify it, but then race with another thread that thinks
	79	* the page is not up-to-date and hence reads it again.
	80	*
	81	* The result is that the first modifcation to the page is lost.
	82	* This sort of AGF/AGI reading race can happen when unlinking inodes
	83	* that require truncation and results in the AGI unlinked list
	84	* modifications being lost.
	85	*/
	86	#define _XBF_PAGE_LOCKED (1 << 22)
	87
807cbbdb	88	typedef unsigned int xfs_buf_flags_t;
1da177e4	89
0b1b213f CH	90	#define XFS_BUF_FLAGS \
	91	{ XBF_READ, "READ" }, \
	92	{ XBF_WRITE, "WRITE" }, \
	93	{ XBF_MAPPED, "MAPPED" }, \
	94	{ XBF_ASYNC, "ASYNC" }, \
	95	{ XBF_DONE, "DONE" }, \
	96	{ XBF_DELWRI, "DELWRI" }, \
	97	{ XBF_STALE, "STALE" }, \
0b1b213f CH	98	{ XBF_ORDERED, "ORDERED" }, \
	99	{ XBF_READ_AHEAD, "READ_AHEAD" }, \
	100	{ XBF_LOCK, "LOCK" }, /* should never be set */\
	101	{ XBF_TRYLOCK, "TRYLOCK" }, /* ditto */\
	102	{ XBF_DONT_BLOCK, "DONT_BLOCK" }, /* ditto */\
	103	{ _XBF_PAGE_CACHE, "PAGE_CACHE" }, \
	104	{ _XBF_PAGES, "PAGES" }, \
	105	{ _XBF_RUN_QUEUES, "RUN_QUEUES" }, \
	106	{ _XBF_DELWRI_Q, "DELWRI_Q" }, \
80f6c29d	107	{ _XBF_PAGE_LOCKED, "PAGE_LOCKED" }
0b1b213f CH	108
0b1b213f CH	109
ce8e922c	110	typedef enum {
5e6a07df DC	111	XBT_FORCE_SLEEP = 0,
5e6a07df DC	112	XBT_FORCE_FLUSH = 1,
ce8e922c	113	} xfs_buftarg_flags_t;
1da177e4 LT	114
	115	typedef struct xfs_bufhash {
	116	struct list_head bh_list;
	117	spinlock_t bh_lock;
	118	} xfs_bufhash_t;
	119
	120	typedef struct xfs_buftarg {
ce8e922c NS	121	dev_t bt_dev;
	122	struct block_device *bt_bdev;
	123	struct address_space *bt_mapping;
ebad861b	124	struct xfs_mount *bt_mount;
ce8e922c NS	125	unsigned int bt_bsize;
	126	unsigned int bt_sshift;
	127	size_t bt_smask;
	128
a6867a68 DC	129	/* per device delwri queue */
a6867a68 DC	130	struct task_struct *bt_task;
a6867a68 DC	131	struct list_head bt_delwrite_queue;
a6867a68 DC	132	spinlock_t bt_delwrite_lock;
ce8e922c	133	unsigned long bt_flags;
ff57ab21 DC	134
	135	/* LRU control structures */
	136	struct shrinker bt_shrinker;
430cbeb8 DC	137	struct list_head bt_lru;
	138	spinlock_t bt_lru_lock;
	139	unsigned int bt_lru_nr;
1da177e4 LT	140	} xfs_buftarg_t;
	141
	142	/*
ce8e922c	143	* xfs_buf_t: Buffer structure for pagecache-based buffers
1da177e4	144	*
ce8e922c NS	145	* This buffer structure is used by the pagecache buffer management routines
	146	* to refer to an assembly of pages forming a logical buffer.
	147	*
	148	* The buffer structure is used on a temporary basis only, and discarded when
	149	* released. The real data storage is recorded in the pagecache. Buffers are
1da177e4 LT	150	* hashed to the block device on which the file system resides.
	151	*/
	152
	153	struct xfs_buf;
ce8e922c	154	typedef void (xfs_buf_iodone_t)(struct xfs_buf );
1da177e4	155
ce8e922c	156	#define XB_PAGES 2
1da177e4 LT	157
1da177e4 LT	158	typedef struct xfs_buf {
50f59e8e DC	159	/*
	160	* first cacheline holds all the fields needed for an uncontended cache
	161	* hit to be fully processed. The semaphore straddles the cacheline
	162	* boundary, but the counter and lock sits on the first cacheline,
	163	* which is the only bit that is touched if we hit the semaphore
	164	* fast-path on locking.
	165	*/
	166	struct rb_node b_rbnode; /* rbtree node */
	167	xfs_off_t b_file_offset; /* offset in file */
	168	size_t b_buffer_length;/* size of buffer in bytes */
	169	atomic_t b_hold; /* reference count */
430cbeb8	170	atomic_t b_lru_ref; /* lru reclaim ref count */
50f59e8e	171	xfs_buf_flags_t b_flags; /* status flags */
ce8e922c	172	struct semaphore b_sema; /* semaphore for lockables */
50f59e8e	173
430cbeb8	174	struct list_head b_lru; /* lru list */
ce8e922c NS	175	wait_queue_head_t b_waiters; /* unpin waiters */
ce8e922c NS	176	struct list_head b_list;
74f75a0c	177	struct xfs_perag b_pag; / contains rbtree root */
ce8e922c	178	xfs_buftarg_t b_target; / buffer target (device) */
ce8e922c	179	xfs_daddr_t b_bn; /* block number for I/O */
ce8e922c NS	180	size_t b_count_desired;/* desired transfer size */
	181	void b_addr; / virtual address of buffer */
	182	struct work_struct b_iodone_work;
ce8e922c	183	xfs_buf_iodone_t b_iodone; /* I/O completion function */
b4dd330b	184	struct completion b_iowait; /* queue for I/O waiters */
ce8e922c NS	185	void *b_fspriv;
ce8e922c NS	186	void *b_fspriv2;
ce8e922c NS	187	struct page *b_pages; / array of page pointers */
ce8e922c NS	188	struct page b_page_array[XB_PAGES]; / inline pages */
50f59e8e DC	189	unsigned long b_queuetime; /* time buffer was queued */
	190	atomic_t b_pin_count; /* pin count */
	191	atomic_t b_io_remaining; /* #outstanding I/O requests */
	192	unsigned int b_page_count; /* size of page array */
	193	unsigned int b_offset; /* page offset in first page */
	194	unsigned short b_error; /* error code on I/O */
ce8e922c NS	195	#ifdef XFS_BUF_LOCK_TRACKING
ce8e922c NS	196	int b_last_holder;
1da177e4 LT	197	#endif
	198	} xfs_buf_t;
	199
	200
	201	/* Finding and Reading Buffers */
ce8e922c NS	202	extern xfs_buf_t _xfs_buf_find(xfs_buftarg_t , xfs_off_t, size_t,
ce8e922c NS	203	xfs_buf_flags_t, xfs_buf_t *);
1da177e4	204	#define xfs_incore(buftarg,blkno,len,lockit) \
ce8e922c	205	_xfs_buf_find(buftarg, blkno ,len, lockit, NULL)
1da177e4	206
6ad112bf	207	extern xfs_buf_t xfs_buf_get(xfs_buftarg_t , xfs_off_t, size_t,
ce8e922c	208	xfs_buf_flags_t);
6ad112bf	209	extern xfs_buf_t xfs_buf_read(xfs_buftarg_t , xfs_off_t, size_t,
ce8e922c	210	xfs_buf_flags_t);
1da177e4	211
ce8e922c	212	extern xfs_buf_t xfs_buf_get_empty(size_t, xfs_buftarg_t );
686865f7	213	extern xfs_buf_t xfs_buf_get_uncached(struct xfs_buftarg , size_t, int);
ce8e922c NS	214	extern int xfs_buf_associate_memory(xfs_buf_t , void , size_t);
ce8e922c NS	215	extern void xfs_buf_hold(xfs_buf_t *);
1a1a3e97	216	extern void xfs_buf_readahead(xfs_buftarg_t *, xfs_off_t, size_t);
5adc94c2 DC	217	struct xfs_buf xfs_buf_read_uncached(struct xfs_mount mp,
	218	struct xfs_buftarg *target,
	219	xfs_daddr_t daddr, size_t length, int flags);
1da177e4 LT	220
1da177e4 LT	221	/* Releasing Buffers */
ce8e922c NS	222	extern void xfs_buf_free(xfs_buf_t *);
ce8e922c NS	223	extern void xfs_buf_rele(xfs_buf_t *);
1da177e4 LT	224
1da177e4 LT	225	/* Locking and Unlocking Buffers */
ce8e922c NS	226	extern int xfs_buf_cond_lock(xfs_buf_t *);
	227	extern int xfs_buf_lock_value(xfs_buf_t *);
	228	extern void xfs_buf_lock(xfs_buf_t *);
	229	extern void xfs_buf_unlock(xfs_buf_t *);
1da177e4 LT	230
1da177e4 LT	231	/* Buffer Read and Write Routines */
64e0bc7d	232	extern int xfs_bwrite(struct xfs_mount mp, struct xfs_buf bp);
5d765b97	233	extern void xfs_bdwrite(void mp, xfs_buf_t bp);
4e23471a CH	234
	235	extern void xfsbdstrat(struct xfs_mount , struct xfs_buf );
	236	extern int xfs_bdstrat_cb(struct xfs_buf *);
	237
ce8e922c NS	238	extern void xfs_buf_ioend(xfs_buf_t *, int);
ce8e922c NS	239	extern void xfs_buf_ioerror(xfs_buf_t *, int);
ce8e922c NS	240	extern int xfs_buf_iorequest(xfs_buf_t *);
ce8e922c NS	241	extern int xfs_buf_iowait(xfs_buf_t *);
b9c48649	242	extern void xfs_buf_iomove(xfs_buf_t , size_t, size_t, void ,
ce8e922c	243	xfs_buf_rw_t);
1a1a3e97 CH	244	#define xfs_buf_zero(bp, off, len) \
1a1a3e97 CH	245	xfs_buf_iomove((bp), (off), (len), NULL, XBRW_ZERO)
ce8e922c	246
ce8e922c	247	static inline int xfs_buf_geterror(xfs_buf_t *bp)
1da177e4	248	{
ce8e922c	249	return bp ? bp->b_error : ENOMEM;
1da177e4 LT	250	}
	251
	252	/* Buffer Utility Routines */
ce8e922c	253	extern xfs_caddr_t xfs_buf_offset(xfs_buf_t *, size_t);
1da177e4	254
1da177e4	255	/* Delayed Write Buffer Routines */
ce8e922c	256	extern void xfs_buf_delwri_dequeue(xfs_buf_t *);
d808f617	257	extern void xfs_buf_delwri_promote(xfs_buf_t *);
1da177e4 LT	258
1da177e4 LT	259	/* Buffer Daemon Setup Routines */
ce8e922c NS	260	extern int xfs_buf_init(void);
ce8e922c NS	261	extern void xfs_buf_terminate(void);
1da177e4	262
ce8e922c NS	263	#define xfs_buf_target_name(target) \
ce8e922c NS	264	({ char __b[BDEVNAME_SIZE]; bdevname((target)->bt_bdev, __b); __b; })
1da177e4 LT	265
1da177e4 LT	266
ce8e922c	267	#define XFS_BUF_BFLAGS(bp) ((bp)->b_flags)
f5faad79 NS	268	#define XFS_BUF_ZEROFLAGS(bp) ((bp)->b_flags &= \
f5faad79 NS	269	~(XBF_READ\|XBF_WRITE\|XBF_ASYNC\|XBF_DELWRI\|XBF_ORDERED))
ce8e922c	270
430cbeb8 DC	271	void xfs_buf_stale(struct xfs_buf *bp);
430cbeb8 DC	272	#define XFS_BUF_STALE(bp) xfs_buf_stale(bp);
0cadda1c CH	273	#define XFS_BUF_UNSTALE(bp) ((bp)->b_flags &= ~XBF_STALE)
0cadda1c CH	274	#define XFS_BUF_ISSTALE(bp) ((bp)->b_flags & XBF_STALE)
ce8e922c NS	275	#define XFS_BUF_SUPER_STALE(bp) do { \
	276	XFS_BUF_STALE(bp); \
	277	xfs_buf_delwri_dequeue(bp); \
	278	XFS_BUF_DONE(bp); \
	279	} while (0)
1da177e4	280
ce8e922c NS	281	#define XFS_BUF_DELAYWRITE(bp) ((bp)->b_flags \|= XBF_DELWRI)
	282	#define XFS_BUF_UNDELAYWRITE(bp) xfs_buf_delwri_dequeue(bp)
	283	#define XFS_BUF_ISDELAYWRITE(bp) ((bp)->b_flags & XBF_DELWRI)
	284
	285	#define XFS_BUF_ERROR(bp,no) xfs_buf_ioerror(bp,no)
	286	#define XFS_BUF_GETERROR(bp) xfs_buf_geterror(bp)
	287	#define XFS_BUF_ISERROR(bp) (xfs_buf_geterror(bp) ? 1 : 0)
	288
	289	#define XFS_BUF_DONE(bp) ((bp)->b_flags \|= XBF_DONE)
	290	#define XFS_BUF_UNDONE(bp) ((bp)->b_flags &= ~XBF_DONE)
	291	#define XFS_BUF_ISDONE(bp) ((bp)->b_flags & XBF_DONE)
	292
	293	#define XFS_BUF_BUSY(bp) do { } while (0)
	294	#define XFS_BUF_UNBUSY(bp) do { } while (0)
	295	#define XFS_BUF_ISBUSY(bp) (1)
	296
	297	#define XFS_BUF_ASYNC(bp) ((bp)->b_flags \|= XBF_ASYNC)
	298	#define XFS_BUF_UNASYNC(bp) ((bp)->b_flags &= ~XBF_ASYNC)
	299	#define XFS_BUF_ISASYNC(bp) ((bp)->b_flags & XBF_ASYNC)
	300
	301	#define XFS_BUF_ORDERED(bp) ((bp)->b_flags \|= XBF_ORDERED)
	302	#define XFS_BUF_UNORDERED(bp) ((bp)->b_flags &= ~XBF_ORDERED)
	303	#define XFS_BUF_ISORDERED(bp) ((bp)->b_flags & XBF_ORDERED)
ce8e922c NS	304
	305	#define XFS_BUF_HOLD(bp) xfs_buf_hold(bp)
	306	#define XFS_BUF_READ(bp) ((bp)->b_flags \|= XBF_READ)
	307	#define XFS_BUF_UNREAD(bp) ((bp)->b_flags &= ~XBF_READ)
	308	#define XFS_BUF_ISREAD(bp) ((bp)->b_flags & XBF_READ)
	309
	310	#define XFS_BUF_WRITE(bp) ((bp)->b_flags \|= XBF_WRITE)
	311	#define XFS_BUF_UNWRITE(bp) ((bp)->b_flags &= ~XBF_WRITE)
	312	#define XFS_BUF_ISWRITE(bp) ((bp)->b_flags & XBF_WRITE)
	313
ce8e922c NS	314	#define XFS_BUF_IODONE_FUNC(bp) ((bp)->b_iodone)
	315	#define XFS_BUF_SET_IODONE_FUNC(bp, func) ((bp)->b_iodone = (func))
	316	#define XFS_BUF_CLR_IODONE_FUNC(bp) ((bp)->b_iodone = NULL)
ce8e922c NS	317
	318	#define XFS_BUF_FSPRIVATE(bp, type) ((type)(bp)->b_fspriv)
	319	#define XFS_BUF_SET_FSPRIVATE(bp, val) ((bp)->b_fspriv = (void*)(val))
	320	#define XFS_BUF_FSPRIVATE2(bp, type) ((type)(bp)->b_fspriv2)
	321	#define XFS_BUF_SET_FSPRIVATE2(bp, val) ((bp)->b_fspriv2 = (void*)(val))
ce8e922c	322	#define XFS_BUF_SET_START(bp) do { } while (0)
ce8e922c NS	323
	324	#define XFS_BUF_PTR(bp) (xfs_caddr_t)((bp)->b_addr)
	325	#define XFS_BUF_SET_PTR(bp, val, cnt) xfs_buf_associate_memory(bp, val, cnt)
	326	#define XFS_BUF_ADDR(bp) ((bp)->b_bn)
	327	#define XFS_BUF_SET_ADDR(bp, bno) ((bp)->b_bn = (xfs_daddr_t)(bno))
	328	#define XFS_BUF_OFFSET(bp) ((bp)->b_file_offset)
	329	#define XFS_BUF_SET_OFFSET(bp, off) ((bp)->b_file_offset = (off))
	330	#define XFS_BUF_COUNT(bp) ((bp)->b_count_desired)
	331	#define XFS_BUF_SET_COUNT(bp, cnt) ((bp)->b_count_desired = (cnt))
	332	#define XFS_BUF_SIZE(bp) ((bp)->b_buffer_length)
	333	#define XFS_BUF_SET_SIZE(bp, cnt) ((bp)->b_buffer_length = (cnt))
	334
821eb21d DC	335	static inline void
	336	xfs_buf_set_ref(
	337	struct xfs_buf *bp,
	338	int lru_ref)
	339	{
	340	atomic_set(&bp->b_lru_ref, lru_ref);
	341	}
	342	#define XFS_BUF_SET_VTYPE_REF(bp, type, ref) xfs_buf_set_ref(bp, ref)
ce8e922c	343	#define XFS_BUF_SET_VTYPE(bp, type) do { } while (0)
ce8e922c	344
4d16e924	345	#define XFS_BUF_ISPINNED(bp) atomic_read(&((bp)->b_pin_count))
ce8e922c NS	346
	347	#define XFS_BUF_VALUSEMA(bp) xfs_buf_lock_value(bp)
	348	#define XFS_BUF_CPSEMA(bp) (xfs_buf_cond_lock(bp) == 0)
	349	#define XFS_BUF_VSEMA(bp) xfs_buf_unlock(bp)
	350	#define XFS_BUF_PSEMA(bp,x) xfs_buf_lock(bp)
b4dd330b	351	#define XFS_BUF_FINISH_IOWAIT(bp) complete(&bp->b_iowait);
ce8e922c NS	352
	353	#define XFS_BUF_SET_TARGET(bp, target) ((bp)->b_target = (target))
	354	#define XFS_BUF_TARGET(bp) ((bp)->b_target)
	355	#define XFS_BUFTARG_NAME(target) xfs_buf_target_name(target)
	356
ce8e922c	357	static inline void xfs_buf_relse(xfs_buf_t *bp)
1da177e4	358	{
bfc60177	359	xfs_buf_unlock(bp);
ce8e922c	360	xfs_buf_rele(bp);
1da177e4 LT	361	}
1da177e4 LT	362
1da177e4 LT	363	/*
	364	* Handling of buftargs.
	365	*/
ebad861b DC	366	extern xfs_buftarg_t xfs_alloc_buftarg(struct xfs_mount ,
ebad861b DC	367	struct block_device , int, const char );
c141b292	368	extern void xfs_free_buftarg(struct xfs_mount , struct xfs_buftarg );
1da177e4 LT	369	extern void xfs_wait_buftarg(xfs_buftarg_t *);
1da177e4 LT	370	extern int xfs_setsize_buftarg(xfs_buftarg_t *, unsigned int, unsigned int);
1da177e4	371	extern int xfs_flush_buftarg(xfs_buftarg_t *, int);
d808f617	372
e6a0e9cd TS	373	#ifdef CONFIG_KDB_MODULES
	374	extern struct list_head *xfs_get_buftarg_list(void);
	375	#endif
1da177e4	376
ce8e922c NS	377	#define xfs_getsize_buftarg(buftarg) block_size((buftarg)->bt_bdev)
	378	#define xfs_readonly_buftarg(buftarg) bdev_read_only((buftarg)->bt_bdev)
	379
	380	#define xfs_binval(buftarg) xfs_flush_buftarg(buftarg, 1)
	381	#define XFS_bflush(buftarg) xfs_flush_buftarg(buftarg, 1)
1da177e4 LT	382
1da177e4 LT	383	#endif /* __XFS_BUF_H__ */