[linux-block.git] / include / linux / bio.h

/*
 * 2.5 block I/O model
 *
 * Copyright (C) 2001 Jens Axboe <axboe@suse.de>
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License version 2 as
 * published by the Free Software Foundation.
 *
 * This program is distributed in the hope that it will 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 Licens
 * along with this program; if not, write to the Free Software
 * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-
 */
#ifndef __LINUX_BIO_H
#define __LINUX_BIO_H

#include <linux/highmem.h>
#include <linux/mempool.h>
#include <linux/ioprio.h>

#ifdef CONFIG_BLOCK

/* Platforms may set this to teach the BIO layer about IOMMU hardware. */
#include <asm/io.h>

#if defined(BIO_VMERGE_MAX_SIZE) && defined(BIO_VMERGE_BOUNDARY)
#define BIOVEC_VIRT_START_SIZE(x) (bvec_to_phys(x) & (BIO_VMERGE_BOUNDARY - 1))
#define BIOVEC_VIRT_OVERSIZE(x)	((x) > BIO_VMERGE_MAX_SIZE)
#else
#define BIOVEC_VIRT_START_SIZE(x)	0
#define BIOVEC_VIRT_OVERSIZE(x)		0
#endif

#ifndef BIO_VMERGE_BOUNDARY
#define BIO_VMERGE_BOUNDARY	0
#endif

#define BIO_DEBUG

#ifdef BIO_DEBUG
#define BIO_BUG_ON	BUG_ON
#else
#define BIO_BUG_ON
#endif

#define BIO_MAX_PAGES		256
#define BIO_MAX_SIZE		(BIO_MAX_PAGES << PAGE_CACHE_SHIFT)
#define BIO_MAX_SECTORS		(BIO_MAX_SIZE >> 9)

/*
 * was unsigned short, but we might as well be ready for > 64kB I/O pages
 */
struct bio_vec {
	struct page	*bv_page;
	unsigned int	bv_len;
	unsigned int	bv_offset;
};

struct bio_set;
struct bio;
typedef void (bio_end_io_t) (struct bio *, int);
typedef void (bio_destructor_t) (struct bio *);

/*
 * main unit of I/O for the block layer and lower layers (ie drivers and
 * stacking drivers)
 */
struct bio {
	sector_t		bi_sector;	/* device address in 512 byte
						   sectors */
	struct bio		*bi_next;	/* request queue link */
	struct block_device	*bi_bdev;
	unsigned long		bi_flags;	/* status, command, etc */
	unsigned long		bi_rw;		/* bottom bits READ/WRITE,
						 * top bits priority
						 */

	unsigned short		bi_vcnt;	/* how many bio_vec's */
	unsigned short		bi_idx;		/* current index into bvl_vec */

	/* Number of segments in this BIO after
	 * physical address coalescing is performed.
	 */
	unsigned short		bi_phys_segments;

	/* Number of segments after physical and DMA remapping
	 * hardware coalescing is performed.
	 */
	unsigned short		bi_hw_segments;

	unsigned int		bi_size;	/* residual I/O count */

	/*
	 * To keep track of the max hw size, we account for the
	 * sizes of the first and last virtually mergeable segments
	 * in this bio
	 */
	unsigned int		bi_hw_front_size;
	unsigned int		bi_hw_back_size;

	unsigned int		bi_max_vecs;	/* max bvl_vecs we can hold */

	struct bio_vec		*bi_io_vec;	/* the actual vec list */

	bio_end_io_t		*bi_end_io;
	atomic_t		bi_cnt;		/* pin count */

	void			*bi_private;

	bio_destructor_t	*bi_destructor;	/* destructor */
};

/*
 * bio flags
 */
#define BIO_UPTODATE	0	/* ok after I/O completion */
#define BIO_RW_BLOCK	1	/* RW_AHEAD set, and read/write would block */
#define BIO_EOF		2	/* out-out-bounds error */
#define BIO_SEG_VALID	3	/* nr_hw_seg valid */
#define BIO_CLONED	4	/* doesn't own data */
#define BIO_BOUNCED	5	/* bio is a bounce bio */
#define BIO_USER_MAPPED 6	/* contains user pages */
#define BIO_EOPNOTSUPP	7	/* not supported */
#define bio_flagged(bio, flag)	((bio)->bi_flags & (1 << (flag)))

/*
 * top 4 bits of bio flags indicate the pool this bio came from
 */
#define BIO_POOL_BITS		(4)
#define BIO_POOL_OFFSET		(BITS_PER_LONG - BIO_POOL_BITS)
#define BIO_POOL_MASK		(1UL << BIO_POOL_OFFSET)
#define BIO_POOL_IDX(bio)	((bio)->bi_flags >> BIO_POOL_OFFSET)	

/*
 * bio bi_rw flags
 *
 * bit 0 -- read (not set) or write (set)
 * bit 1 -- rw-ahead when set
 * bit 2 -- barrier
 * bit 3 -- fail fast, don't want low level driver retries
 * bit 4 -- synchronous I/O hint: the block layer will unplug immediately
 */
#define BIO_RW		0
#define BIO_RW_AHEAD	1
#define BIO_RW_BARRIER	2
#define BIO_RW_FAILFAST	3
#define BIO_RW_SYNC	4
#define BIO_RW_META	5

/*
 * upper 16 bits of bi_rw define the io priority of this bio
 */
#define BIO_PRIO_SHIFT	(8 * sizeof(unsigned long) - IOPRIO_BITS)
#define bio_prio(bio)	((bio)->bi_rw >> BIO_PRIO_SHIFT)
#define bio_prio_valid(bio)	ioprio_valid(bio_prio(bio))

#define bio_set_prio(bio, prio)		do {			\
	WARN_ON(prio >= (1 << IOPRIO_BITS));			\
	(bio)->bi_rw &= ((1UL << BIO_PRIO_SHIFT) - 1);		\
	(bio)->bi_rw |= ((unsigned long) (prio) << BIO_PRIO_SHIFT);	\
} while (0)

/*
 * various member access, note that bio_data should of course not be used
 * on highmem page vectors
 */
#define bio_iovec_idx(bio, idx)	(&((bio)->bi_io_vec[(idx)]))
#define bio_iovec(bio)		bio_iovec_idx((bio), (bio)->bi_idx)
#define bio_page(bio)		bio_iovec((bio))->bv_page
#define bio_offset(bio)		bio_iovec((bio))->bv_offset
#define bio_segments(bio)	((bio)->bi_vcnt - (bio)->bi_idx)
#define bio_sectors(bio)	((bio)->bi_size >> 9)
#define bio_barrier(bio)	((bio)->bi_rw & (1 << BIO_RW_BARRIER))
#define bio_sync(bio)		((bio)->bi_rw & (1 << BIO_RW_SYNC))
#define bio_failfast(bio)	((bio)->bi_rw & (1 << BIO_RW_FAILFAST))
#define bio_rw_ahead(bio)	((bio)->bi_rw & (1 << BIO_RW_AHEAD))
#define bio_rw_meta(bio)	((bio)->bi_rw & (1 << BIO_RW_META))
#define bio_empty_barrier(bio)	(bio_barrier(bio) && !(bio)->bi_size)

static inline unsigned int bio_cur_sectors(struct bio *bio)
{
	if (bio->bi_vcnt)
		return bio_iovec(bio)->bv_len >> 9;

	return 0;
}

static inline void *bio_data(struct bio *bio)
{
	if (bio->bi_vcnt)
		return page_address(bio_page(bio)) + bio_offset(bio);

	return NULL;
}

/*
 * will die
 */
#define bio_to_phys(bio)	(page_to_phys(bio_page((bio))) + (unsigned long) bio_offset((bio)))
#define bvec_to_phys(bv)	(page_to_phys((bv)->bv_page) + (unsigned long) (bv)->bv_offset)

/*
 * queues that have highmem support enabled may still need to revert to
 * PIO transfers occasionally and thus map high pages temporarily. For
 * permanent PIO fall back, user is probably better off disabling highmem
 * I/O completely on that queue (see ide-dma for example)
 */
#define __bio_kmap_atomic(bio, idx, kmtype)				\
	(kmap_atomic(bio_iovec_idx((bio), (idx))->bv_page, kmtype) +	\
		bio_iovec_idx((bio), (idx))->bv_offset)

#define __bio_kunmap_atomic(addr, kmtype) kunmap_atomic(addr, kmtype)

/*
 * merge helpers etc
 */

#define __BVEC_END(bio)		bio_iovec_idx((bio), (bio)->bi_vcnt - 1)
#define __BVEC_START(bio)	bio_iovec_idx((bio), (bio)->bi_idx)

/*
 * allow arch override, for eg virtualized architectures (put in asm/io.h)
 */
#ifndef BIOVEC_PHYS_MERGEABLE
#define BIOVEC_PHYS_MERGEABLE(vec1, vec2)	\
	((bvec_to_phys((vec1)) + (vec1)->bv_len) == bvec_to_phys((vec2)))
#endif

#define BIOVEC_VIRT_MERGEABLE(vec1, vec2)	\
	((((bvec_to_phys((vec1)) + (vec1)->bv_len) | bvec_to_phys((vec2))) & (BIO_VMERGE_BOUNDARY - 1)) == 0)
#define __BIO_SEG_BOUNDARY(addr1, addr2, mask) \
	(((addr1) | (mask)) == (((addr2) - 1) | (mask)))
#define BIOVEC_SEG_BOUNDARY(q, b1, b2) \
	__BIO_SEG_BOUNDARY(bvec_to_phys((b1)), bvec_to_phys((b2)) + (b2)->bv_len, (q)->seg_boundary_mask)
#define BIO_SEG_BOUNDARY(q, b1, b2) \
	BIOVEC_SEG_BOUNDARY((q), __BVEC_END((b1)), __BVEC_START((b2)))

#define bio_io_error(bio) bio_endio((bio), -EIO)

/*
 * drivers should not use the __ version unless they _really_ want to
 * run through the entire bio and not just pending pieces
 */
#define __bio_for_each_segment(bvl, bio, i, start_idx)			\
	for (bvl = bio_iovec_idx((bio), (start_idx)), i = (start_idx);	\
	     i < (bio)->bi_vcnt;					\
	     bvl++, i++)

#define bio_for_each_segment(bvl, bio, i)				\
	__bio_for_each_segment(bvl, bio, i, (bio)->bi_idx)

/*
 * get a reference to a bio, so it won't disappear. the intended use is
 * something like:
 *
 * bio_get(bio);
 * submit_bio(rw, bio);
 * if (bio->bi_flags ...)
 *	do_something
 * bio_put(bio);
 *
 * without the bio_get(), it could potentially complete I/O before submit_bio
 * returns. and then bio would be freed memory when if (bio->bi_flags ...)
 * runs
 */
#define bio_get(bio)	atomic_inc(&(bio)->bi_cnt)


/*
 * A bio_pair is used when we need to split a bio.
 * This can only happen for a bio that refers to just one
 * page of data, and in the unusual situation when the
 * page crosses a chunk/device boundary
 *
 * The address of the master bio is stored in bio1.bi_private
 * The address of the pool the pair was allocated from is stored
 *   in bio2.bi_private
 */
struct bio_pair {
	struct bio	bio1, bio2;
	struct bio_vec	bv1, bv2;
	atomic_t	cnt;
	int		error;
};
extern struct bio_pair *bio_split(struct bio *bi, mempool_t *pool,
				  int first_sectors);
extern mempool_t *bio_split_pool;
extern void bio_pair_release(struct bio_pair *dbio);

extern struct bio_set *bioset_create(int, int);
extern void bioset_free(struct bio_set *);

extern struct bio *bio_alloc(gfp_t, int);
extern struct bio *bio_alloc_bioset(gfp_t, int, struct bio_set *);
extern void bio_put(struct bio *);
extern void bio_free(struct bio *, struct bio_set *);

extern void bio_endio(struct bio *, int);
struct request_queue;
extern int bio_phys_segments(struct request_queue *, struct bio *);
extern int bio_hw_segments(struct request_queue *, struct bio *);

extern void __bio_clone(struct bio *, struct bio *);
extern struct bio *bio_clone(struct bio *, gfp_t);

extern void bio_init(struct bio *);

extern int bio_add_page(struct bio *, struct page *, unsigned int,unsigned int);
extern int bio_add_pc_page(struct request_queue *, struct bio *, struct page *,
			   unsigned int, unsigned int);
extern int bio_get_nr_vecs(struct block_device *);
extern struct bio *bio_map_user(struct request_queue *, struct block_device *,
				unsigned long, unsigned int, int);
struct sg_iovec;
extern struct bio *bio_map_user_iov(struct request_queue *,
				    struct block_device *,
				    struct sg_iovec *, int, int);
extern void bio_unmap_user(struct bio *);
extern struct bio *bio_map_kern(struct request_queue *, void *, unsigned int,
				gfp_t);
extern struct bio *bio_copy_kern(struct request_queue *, void *, unsigned int,
				 gfp_t, int);
extern void bio_set_pages_dirty(struct bio *bio);
extern void bio_check_pages_dirty(struct bio *bio);
extern struct bio *bio_copy_user(struct request_queue *, unsigned long, unsigned int, int);
extern struct bio *bio_copy_user_iov(struct request_queue *, struct sg_iovec *,
				     int, int);
extern int bio_uncopy_user(struct bio *);
void zero_fill_bio(struct bio *bio);
extern struct bio_vec *bvec_alloc_bs(gfp_t, int, unsigned long *, struct bio_set *);

/*
 * bio_set is used to allow other portions of the IO system to
 * allocate their own private memory pools for bio and iovec structures.
 * These memory pools in turn all allocate from the bio_slab
 * and the bvec_slabs[].
 */
#define BIO_POOL_SIZE 2
#define BIOVEC_NR_POOLS 6

struct bio_set {
	mempool_t *bio_pool;
	mempool_t *bvec_pools[BIOVEC_NR_POOLS];
};

struct biovec_slab {
	int nr_vecs;
	char *name;
	struct kmem_cache *slab;
};

extern struct bio_set *fs_bio_set;

/*
 * a small number of entries is fine, not going to be performance critical.
 * basically we just need to survive
 */
#define BIO_SPLIT_ENTRIES 2

#ifdef CONFIG_HIGHMEM
/*
 * remember to add offset! and never ever reenable interrupts between a
 * bvec_kmap_irq and bvec_kunmap_irq!!
 *
 * This function MUST be inlined - it plays with the CPU interrupt flags.
 */
static inline char *bvec_kmap_irq(struct bio_vec *bvec, unsigned long *flags)
{
	unsigned long addr;

	/*
	 * might not be a highmem page, but the preempt/irq count
	 * balancing is a lot nicer this way
	 */
	local_irq_save(*flags);
	addr = (unsigned long) kmap_atomic(bvec->bv_page, KM_BIO_SRC_IRQ);

	BUG_ON(addr & ~PAGE_MASK);

	return (char *) addr + bvec->bv_offset;
}

static inline void bvec_kunmap_irq(char *buffer, unsigned long *flags)
{
	unsigned long ptr = (unsigned long) buffer & PAGE_MASK;

	kunmap_atomic((void *) ptr, KM_BIO_SRC_IRQ);
	local_irq_restore(*flags);
}

#else
#define bvec_kmap_irq(bvec, flags)	(page_address((bvec)->bv_page) + (bvec)->bv_offset)
#define bvec_kunmap_irq(buf, flags)	do { *(flags) = 0; } while (0)
#endif

static inline char *__bio_kmap_irq(struct bio *bio, unsigned short idx,
				   unsigned long *flags)
{
	return bvec_kmap_irq(bio_iovec_idx(bio, idx), flags);
}
#define __bio_kunmap_irq(buf, flags)	bvec_kunmap_irq(buf, flags)

#define bio_kmap_irq(bio, flags) \
	__bio_kmap_irq((bio), (bio)->bi_idx, (flags))
#define bio_kunmap_irq(buf,flags)	__bio_kunmap_irq(buf, flags)

#endif /* CONFIG_BLOCK */
#endif /* __LINUX_BIO_H */
Commit	Line	Data
1da177e4 LT	1	/*
	2	* 2.5 block I/O model
	3	*
	4	* Copyright (C) 2001 Jens Axboe <axboe@suse.de>
	5	*
	6	* This program is free software; you can redistribute it and/or modify
	7	* it under the terms of the GNU General Public License version 2 as
	8	* published by the Free Software Foundation.
	9	*
	10	* This program is distributed in the hope that it will be useful,
	11	* but WITHOUT ANY WARRANTY; without even the implied warranty of
	12
	13	* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
	14	* GNU General Public License for more details.
	15	*
	16	* You should have received a copy of the GNU General Public Licens
	17	* along with this program; if not, write to the Free Software
	18	* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-
	19	*/
	20	#ifndef __LINUX_BIO_H
	21	#define __LINUX_BIO_H
	22
	23	#include <linux/highmem.h>
	24	#include <linux/mempool.h>
22e2c507	25	#include <linux/ioprio.h>
1da177e4	26
02a5e0ac DH	27	#ifdef CONFIG_BLOCK
02a5e0ac DH	28
1da177e4 LT	29	/* Platforms may set this to teach the BIO layer about IOMMU hardware. */
	30	#include <asm/io.h>
	31
	32	#if defined(BIO_VMERGE_MAX_SIZE) && defined(BIO_VMERGE_BOUNDARY)
	33	#define BIOVEC_VIRT_START_SIZE(x) (bvec_to_phys(x) & (BIO_VMERGE_BOUNDARY - 1))
	34	#define BIOVEC_VIRT_OVERSIZE(x) ((x) > BIO_VMERGE_MAX_SIZE)
	35	#else
	36	#define BIOVEC_VIRT_START_SIZE(x) 0
	37	#define BIOVEC_VIRT_OVERSIZE(x) 0
	38	#endif
	39
	40	#ifndef BIO_VMERGE_BOUNDARY
	41	#define BIO_VMERGE_BOUNDARY 0
	42	#endif
	43
	44	#define BIO_DEBUG
	45
	46	#ifdef BIO_DEBUG
	47	#define BIO_BUG_ON BUG_ON
	48	#else
	49	#define BIO_BUG_ON
	50	#endif
	51
d84a8477	52	#define BIO_MAX_PAGES 256
1da177e4 LT	53	#define BIO_MAX_SIZE (BIO_MAX_PAGES << PAGE_CACHE_SHIFT)
	54	#define BIO_MAX_SECTORS (BIO_MAX_SIZE >> 9)
	55
	56	/*
	57	* was unsigned short, but we might as well be ready for > 64kB I/O pages
	58	*/
	59	struct bio_vec {
	60	struct page *bv_page;
	61	unsigned int bv_len;
	62	unsigned int bv_offset;
	63	};
	64
	65	struct bio_set;
	66	struct bio;
6712ecf8	67	typedef void (bio_end_io_t) (struct bio *, int);
1da177e4 LT	68	typedef void (bio_destructor_t) (struct bio *);
	69
	70	/*
	71	* main unit of I/O for the block layer and lower layers (ie drivers and
	72	* stacking drivers)
	73	*/
	74	struct bio {
2c2345c2 RG	75	sector_t bi_sector; /* device address in 512 byte
2c2345c2 RG	76	sectors */
1da177e4 LT	77	struct bio bi_next; / request queue link */
	78	struct block_device *bi_bdev;
	79	unsigned long bi_flags; /* status, command, etc */
	80	unsigned long bi_rw; /* bottom bits READ/WRITE,
	81	* top bits priority
	82	*/
	83
	84	unsigned short bi_vcnt; /* how many bio_vec's */
	85	unsigned short bi_idx; /* current index into bvl_vec */
	86
	87	/* Number of segments in this BIO after
	88	* physical address coalescing is performed.
	89	*/
	90	unsigned short bi_phys_segments;
	91
	92	/* Number of segments after physical and DMA remapping
	93	* hardware coalescing is performed.
	94	*/
	95	unsigned short bi_hw_segments;
	96
	97	unsigned int bi_size; /* residual I/O count */
	98
	99	/*
	100	* To keep track of the max hw size, we account for the
	101	* sizes of the first and last virtually mergeable segments
	102	* in this bio
	103	*/
	104	unsigned int bi_hw_front_size;
	105	unsigned int bi_hw_back_size;
	106
	107	unsigned int bi_max_vecs; /* max bvl_vecs we can hold */
	108
	109	struct bio_vec bi_io_vec; / the actual vec list */
	110
	111	bio_end_io_t *bi_end_io;
	112	atomic_t bi_cnt; /* pin count */
	113
	114	void *bi_private;
	115
	116	bio_destructor_t bi_destructor; / destructor */
1da177e4 LT	117	};
	118
	119	/*
	120	* bio flags
	121	*/
	122	#define BIO_UPTODATE 0 /* ok after I/O completion */
	123	#define BIO_RW_BLOCK 1 /* RW_AHEAD set, and read/write would block */
	124	#define BIO_EOF 2 /* out-out-bounds error */
	125	#define BIO_SEG_VALID 3 /* nr_hw_seg valid */
	126	#define BIO_CLONED 4 /* doesn't own data */
	127	#define BIO_BOUNCED 5 /* bio is a bounce bio */
	128	#define BIO_USER_MAPPED 6 /* contains user pages */
	129	#define BIO_EOPNOTSUPP 7 /* not supported */
	130	#define bio_flagged(bio, flag) ((bio)->bi_flags & (1 << (flag)))
	131
	132	/*
	133	* top 4 bits of bio flags indicate the pool this bio came from
	134	*/
	135	#define BIO_POOL_BITS (4)
	136	#define BIO_POOL_OFFSET (BITS_PER_LONG - BIO_POOL_BITS)
	137	#define BIO_POOL_MASK (1UL << BIO_POOL_OFFSET)
	138	#define BIO_POOL_IDX(bio) ((bio)->bi_flags >> BIO_POOL_OFFSET)
	139
	140	/*
	141	* bio bi_rw flags
	142	*
	143	* bit 0 -- read (not set) or write (set)
	144	* bit 1 -- rw-ahead when set
	145	* bit 2 -- barrier
	146	* bit 3 -- fail fast, don't want low level driver retries
	147	* bit 4 -- synchronous I/O hint: the block layer will unplug immediately
	148	*/
	149	#define BIO_RW 0
	150	#define BIO_RW_AHEAD 1
	151	#define BIO_RW_BARRIER 2
	152	#define BIO_RW_FAILFAST 3
	153	#define BIO_RW_SYNC 4
5404bc7a	154	#define BIO_RW_META 5
1da177e4	155
22e2c507 JA	156	/*
	157	* upper 16 bits of bi_rw define the io priority of this bio
	158	*/
	159	#define BIO_PRIO_SHIFT (8 * sizeof(unsigned long) - IOPRIO_BITS)
	160	#define bio_prio(bio) ((bio)->bi_rw >> BIO_PRIO_SHIFT)
	161	#define bio_prio_valid(bio) ioprio_valid(bio_prio(bio))
	162
	163	#define bio_set_prio(bio, prio) do { \
	164	WARN_ON(prio >= (1 << IOPRIO_BITS)); \
	165	(bio)->bi_rw &= ((1UL << BIO_PRIO_SHIFT) - 1); \
	166	(bio)->bi_rw \|= ((unsigned long) (prio) << BIO_PRIO_SHIFT); \
	167	} while (0)
	168
1da177e4 LT	169	/*
	170	* various member access, note that bio_data should of course not be used
	171	* on highmem page vectors
	172	*/
	173	#define bio_iovec_idx(bio, idx) (&((bio)->bi_io_vec[(idx)]))
	174	#define bio_iovec(bio) bio_iovec_idx((bio), (bio)->bi_idx)
	175	#define bio_page(bio) bio_iovec((bio))->bv_page
	176	#define bio_offset(bio) bio_iovec((bio))->bv_offset
	177	#define bio_segments(bio) ((bio)->bi_vcnt - (bio)->bi_idx)
	178	#define bio_sectors(bio) ((bio)->bi_size >> 9)
1da177e4 LT	179	#define bio_barrier(bio) ((bio)->bi_rw & (1 << BIO_RW_BARRIER))
	180	#define bio_sync(bio) ((bio)->bi_rw & (1 << BIO_RW_SYNC))
	181	#define bio_failfast(bio) ((bio)->bi_rw & (1 << BIO_RW_FAILFAST))
	182	#define bio_rw_ahead(bio) ((bio)->bi_rw & (1 << BIO_RW_AHEAD))
5404bc7a	183	#define bio_rw_meta(bio) ((bio)->bi_rw & (1 << BIO_RW_META))
bf2de6f5 JA	184	#define bio_empty_barrier(bio) (bio_barrier(bio) && !(bio)->bi_size)
	185
	186	static inline unsigned int bio_cur_sectors(struct bio *bio)
	187	{
	188	if (bio->bi_vcnt)
	189	return bio_iovec(bio)->bv_len >> 9;
	190
	191	return 0;
	192	}
	193
	194	static inline void bio_data(struct bio bio)
	195	{
	196	if (bio->bi_vcnt)
	197	return page_address(bio_page(bio)) + bio_offset(bio);
	198
	199	return NULL;
	200	}
1da177e4 LT	201
	202	/*
	203	* will die
	204	*/
	205	#define bio_to_phys(bio) (page_to_phys(bio_page((bio))) + (unsigned long) bio_offset((bio)))
	206	#define bvec_to_phys(bv) (page_to_phys((bv)->bv_page) + (unsigned long) (bv)->bv_offset)
	207
	208	/*
	209	* queues that have highmem support enabled may still need to revert to
	210	* PIO transfers occasionally and thus map high pages temporarily. For
	211	* permanent PIO fall back, user is probably better off disabling highmem
	212	* I/O completely on that queue (see ide-dma for example)
	213	*/
	214	#define __bio_kmap_atomic(bio, idx, kmtype) \
	215	(kmap_atomic(bio_iovec_idx((bio), (idx))->bv_page, kmtype) + \
	216	bio_iovec_idx((bio), (idx))->bv_offset)
	217
	218	#define __bio_kunmap_atomic(addr, kmtype) kunmap_atomic(addr, kmtype)
	219
	220	/*
	221	* merge helpers etc
	222	*/
	223
	224	#define __BVEC_END(bio) bio_iovec_idx((bio), (bio)->bi_vcnt - 1)
	225	#define __BVEC_START(bio) bio_iovec_idx((bio), (bio)->bi_idx)
	226
	227	/*
	228	* allow arch override, for eg virtualized architectures (put in asm/io.h)
	229	*/
	230	#ifndef BIOVEC_PHYS_MERGEABLE
	231	#define BIOVEC_PHYS_MERGEABLE(vec1, vec2) \
	232	((bvec_to_phys((vec1)) + (vec1)->bv_len) == bvec_to_phys((vec2)))
	233	#endif
	234
	235	#define BIOVEC_VIRT_MERGEABLE(vec1, vec2) \
	236	((((bvec_to_phys((vec1)) + (vec1)->bv_len) \| bvec_to_phys((vec2))) & (BIO_VMERGE_BOUNDARY - 1)) == 0)
	237	#define __BIO_SEG_BOUNDARY(addr1, addr2, mask) \
	238	(((addr1) \| (mask)) == (((addr2) - 1) \| (mask)))
	239	#define BIOVEC_SEG_BOUNDARY(q, b1, b2) \
	240	__BIO_SEG_BOUNDARY(bvec_to_phys((b1)), bvec_to_phys((b2)) + (b2)->bv_len, (q)->seg_boundary_mask)
	241	#define BIO_SEG_BOUNDARY(q, b1, b2) \
	242	BIOVEC_SEG_BOUNDARY((q), __BVEC_END((b1)), __BVEC_START((b2)))
	243
6712ecf8	244	#define bio_io_error(bio) bio_endio((bio), -EIO)
1da177e4 LT	245
	246	/*
	247	* drivers should not use the __ version unless they _really_ want to
	248	* run through the entire bio and not just pending pieces
	249	*/
	250	#define __bio_for_each_segment(bvl, bio, i, start_idx) \
	251	for (bvl = bio_iovec_idx((bio), (start_idx)), i = (start_idx); \
	252	i < (bio)->bi_vcnt; \
	253	bvl++, i++)
	254
	255	#define bio_for_each_segment(bvl, bio, i) \
	256	__bio_for_each_segment(bvl, bio, i, (bio)->bi_idx)
	257
	258	/*
	259	* get a reference to a bio, so it won't disappear. the intended use is
	260	* something like:
	261	*
	262	* bio_get(bio);
	263	* submit_bio(rw, bio);
	264	* if (bio->bi_flags ...)
	265	* do_something
	266	* bio_put(bio);
	267	*
	268	* without the bio_get(), it could potentially complete I/O before submit_bio
	269	* returns. and then bio would be freed memory when if (bio->bi_flags ...)
	270	* runs
	271	*/
	272	#define bio_get(bio) atomic_inc(&(bio)->bi_cnt)
	273
	274
	275	/*
	276	* A bio_pair is used when we need to split a bio.
	277	* This can only happen for a bio that refers to just one
	278	* page of data, and in the unusual situation when the
	279	* page crosses a chunk/device boundary
	280	*
	281	* The address of the master bio is stored in bio1.bi_private
	282	* The address of the pool the pair was allocated from is stored
	283	* in bio2.bi_private
	284	*/
	285	struct bio_pair {
	286	struct bio bio1, bio2;
	287	struct bio_vec bv1, bv2;
	288	atomic_t cnt;
	289	int error;
	290	};
	291	extern struct bio_pair bio_split(struct bio bi, mempool_t *pool,
	292	int first_sectors);
	293	extern mempool_t *bio_split_pool;
	294	extern void bio_pair_release(struct bio_pair *dbio);
	295
5972511b	296	extern struct bio_set *bioset_create(int, int);
1da177e4 LT	297	extern void bioset_free(struct bio_set *);
1da177e4 LT	298
dd0fc66f AV	299	extern struct bio *bio_alloc(gfp_t, int);
dd0fc66f AV	300	extern struct bio bio_alloc_bioset(gfp_t, int, struct bio_set );
1da177e4	301	extern void bio_put(struct bio *);
3676347a	302	extern void bio_free(struct bio , struct bio_set );
1da177e4	303
6712ecf8	304	extern void bio_endio(struct bio *, int);
1da177e4 LT	305	struct request_queue;
	306	extern int bio_phys_segments(struct request_queue , struct bio );
	307	extern int bio_hw_segments(struct request_queue , struct bio );
	308
	309	extern void __bio_clone(struct bio , struct bio );
dd0fc66f	310	extern struct bio bio_clone(struct bio , gfp_t);
1da177e4 LT	311
	312	extern void bio_init(struct bio *);
	313
	314	extern int bio_add_page(struct bio , struct page , unsigned int,unsigned int);
6e68af66 MC	315	extern int bio_add_pc_page(struct request_queue , struct bio , struct page *,
6e68af66 MC	316	unsigned int, unsigned int);
1da177e4 LT	317	extern int bio_get_nr_vecs(struct block_device *);
	318	extern struct bio bio_map_user(struct request_queue , struct block_device *,
	319	unsigned long, unsigned int, int);
f1970baf JB	320	struct sg_iovec;
	321	extern struct bio bio_map_user_iov(struct request_queue ,
	322	struct block_device *,
	323	struct sg_iovec *, int, int);
1da177e4	324	extern void bio_unmap_user(struct bio *);
df46b9a4	325	extern struct bio bio_map_kern(struct request_queue , void *, unsigned int,
27496a8c	326	gfp_t);
68154e90 FT	327	extern struct bio bio_copy_kern(struct request_queue , void *, unsigned int,
68154e90 FT	328	gfp_t, int);
1da177e4 LT	329	extern void bio_set_pages_dirty(struct bio *bio);
	330	extern void bio_check_pages_dirty(struct bio *bio);
	331	extern struct bio bio_copy_user(struct request_queue , unsigned long, unsigned int, int);
c5dec1c3 FT	332	extern struct bio bio_copy_user_iov(struct request_queue , struct sg_iovec *,
c5dec1c3 FT	333	int, int);
1da177e4 LT	334	extern int bio_uncopy_user(struct bio *);
1da177e4 LT	335	void zero_fill_bio(struct bio *bio);
51d654e1 MP	336	extern struct bio_vec bvec_alloc_bs(gfp_t, int, unsigned long , struct bio_set *);
	337
	338	/*
	339	* bio_set is used to allow other portions of the IO system to
	340	* allocate their own private memory pools for bio and iovec structures.
	341	* These memory pools in turn all allocate from the bio_slab
	342	* and the bvec_slabs[].
	343	*/
	344	#define BIO_POOL_SIZE 2
	345	#define BIOVEC_NR_POOLS 6
	346
	347	struct bio_set {
	348	mempool_t *bio_pool;
	349	mempool_t *bvec_pools[BIOVEC_NR_POOLS];
	350	};
	351
	352	struct biovec_slab {
	353	int nr_vecs;
	354	char *name;
	355	struct kmem_cache *slab;
	356	};
	357
	358	extern struct bio_set *fs_bio_set;
	359
	360	/*
	361	* a small number of entries is fine, not going to be performance critical.
	362	* basically we just need to survive
	363	*/
	364	#define BIO_SPLIT_ENTRIES 2
1da177e4 LT	365
	366	#ifdef CONFIG_HIGHMEM
	367	/*
	368	* remember to add offset! and never ever reenable interrupts between a
	369	* bvec_kmap_irq and bvec_kunmap_irq!!
	370	*
	371	* This function MUST be inlined - it plays with the CPU interrupt flags.
1da177e4	372	*/
c2d08dad	373	static inline char bvec_kmap_irq(struct bio_vec bvec, unsigned long *flags)
1da177e4 LT	374	{
	375	unsigned long addr;
	376
	377	/*
	378	* might not be a highmem page, but the preempt/irq count
	379	* balancing is a lot nicer this way
	380	*/
	381	local_irq_save(*flags);
	382	addr = (unsigned long) kmap_atomic(bvec->bv_page, KM_BIO_SRC_IRQ);
	383
	384	BUG_ON(addr & ~PAGE_MASK);
	385
	386	return (char *) addr + bvec->bv_offset;
	387	}
	388
c2d08dad	389	static inline void bvec_kunmap_irq(char buffer, unsigned long flags)
1da177e4 LT	390	{
	391	unsigned long ptr = (unsigned long) buffer & PAGE_MASK;
	392
	393	kunmap_atomic((void *) ptr, KM_BIO_SRC_IRQ);
	394	local_irq_restore(*flags);
	395	}
	396
	397	#else
	398	#define bvec_kmap_irq(bvec, flags) (page_address((bvec)->bv_page) + (bvec)->bv_offset)
	399	#define bvec_kunmap_irq(buf, flags) do { *(flags) = 0; } while (0)
	400	#endif
	401
c2d08dad	402	static inline char __bio_kmap_irq(struct bio bio, unsigned short idx,
1da177e4 LT	403	unsigned long *flags)
	404	{
	405	return bvec_kmap_irq(bio_iovec_idx(bio, idx), flags);
	406	}
	407	#define __bio_kunmap_irq(buf, flags) bvec_kunmap_irq(buf, flags)
	408
	409	#define bio_kmap_irq(bio, flags) \
	410	__bio_kmap_irq((bio), (bio)->bi_idx, (flags))
	411	#define bio_kunmap_irq(buf,flags) __bio_kunmap_irq(buf, flags)
	412
02a5e0ac	413	#endif /* CONFIG_BLOCK */
1da177e4	414	#endif /* __LINUX_BIO_H */