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

/* SPDX-License-Identifier: GPL-2.0 */
/*
 * Block data types and constants.  Directly include this file only to
 * break include dependency loop.
 */
#ifndef __LINUX_BLK_TYPES_H
#define __LINUX_BLK_TYPES_H

#include <linux/types.h>
#include <linux/bvec.h>
#include <linux/ktime.h>

struct bio_set;
struct bio;
struct bio_integrity_payload;
struct page;
struct block_device;
struct io_context;
struct cgroup_subsys_state;
typedef void (bio_end_io_t) (struct bio *);

/*
 * Block error status values.  See block/blk-core:blk_errors for the details.
 * Alpha cannot write a byte atomically, so we need to use 32-bit value.
 */
#if defined(CONFIG_ALPHA) && !defined(__alpha_bwx__)
typedef u32 __bitwise blk_status_t;
#else
typedef u8 __bitwise blk_status_t;
#endif
#define	BLK_STS_OK 0
#define BLK_STS_NOTSUPP		((__force blk_status_t)1)
#define BLK_STS_TIMEOUT		((__force blk_status_t)2)
#define BLK_STS_NOSPC		((__force blk_status_t)3)
#define BLK_STS_TRANSPORT	((__force blk_status_t)4)
#define BLK_STS_TARGET		((__force blk_status_t)5)
#define BLK_STS_NEXUS		((__force blk_status_t)6)
#define BLK_STS_MEDIUM		((__force blk_status_t)7)
#define BLK_STS_PROTECTION	((__force blk_status_t)8)
#define BLK_STS_RESOURCE	((__force blk_status_t)9)
#define BLK_STS_IOERR		((__force blk_status_t)10)

/* hack for device mapper, don't use elsewhere: */
#define BLK_STS_DM_REQUEUE    ((__force blk_status_t)11)

#define BLK_STS_AGAIN		((__force blk_status_t)12)

/*
 * BLK_STS_DEV_RESOURCE is returned from the driver to the block layer if
 * device related resources are unavailable, but the driver can guarantee
 * that the queue will be rerun in the future once resources become
 * available again. This is typically the case for device specific
 * resources that are consumed for IO. If the driver fails allocating these
 * resources, we know that inflight (or pending) IO will free these
 * resource upon completion.
 *
 * This is different from BLK_STS_RESOURCE in that it explicitly references
 * a device specific resource. For resources of wider scope, allocation
 * failure can happen without having pending IO. This means that we can't
 * rely on request completions freeing these resources, as IO may not be in
 * flight. Examples of that are kernel memory allocations, DMA mappings, or
 * any other system wide resources.
 */
#define BLK_STS_DEV_RESOURCE	((__force blk_status_t)13)

/**
 * blk_path_error - returns true if error may be path related
 * @error: status the request was completed with
 *
 * Description:
 *     This classifies block error status into non-retryable errors and ones
 *     that may be successful if retried on a failover path.
 *
 * Return:
 *     %false - retrying failover path will not help
 *     %true  - may succeed if retried
 */
static inline bool blk_path_error(blk_status_t error)
{
	switch (error) {
	case BLK_STS_NOTSUPP:
	case BLK_STS_NOSPC:
	case BLK_STS_TARGET:
	case BLK_STS_NEXUS:
	case BLK_STS_MEDIUM:
	case BLK_STS_PROTECTION:
		return false;
	}

	/* Anything else could be a path failure, so should be retried */
	return true;
}

/*
 * From most significant bit:
 * 1 bit: reserved for other usage, see below
 * 12 bits: original size of bio
 * 51 bits: issue time of bio
 */
#define BIO_ISSUE_RES_BITS      1
#define BIO_ISSUE_SIZE_BITS     12
#define BIO_ISSUE_RES_SHIFT     (64 - BIO_ISSUE_RES_BITS)
#define BIO_ISSUE_SIZE_SHIFT    (BIO_ISSUE_RES_SHIFT - BIO_ISSUE_SIZE_BITS)
#define BIO_ISSUE_TIME_MASK     ((1ULL << BIO_ISSUE_SIZE_SHIFT) - 1)
#define BIO_ISSUE_SIZE_MASK     \
	(((1ULL << BIO_ISSUE_SIZE_BITS) - 1) << BIO_ISSUE_SIZE_SHIFT)
#define BIO_ISSUE_RES_MASK      (~((1ULL << BIO_ISSUE_RES_SHIFT) - 1))

/* Reserved bit for blk-throtl */
#define BIO_ISSUE_THROTL_SKIP_LATENCY (1ULL << 63)

struct bio_issue {
	u64 value;
};

static inline u64 __bio_issue_time(u64 time)
{
	return time & BIO_ISSUE_TIME_MASK;
}

static inline u64 bio_issue_time(struct bio_issue *issue)
{
	return __bio_issue_time(issue->value);
}

static inline sector_t bio_issue_size(struct bio_issue *issue)
{
	return ((issue->value & BIO_ISSUE_SIZE_MASK) >> BIO_ISSUE_SIZE_SHIFT);
}

static inline void bio_issue_init(struct bio_issue *issue,
				       sector_t size)
{
	size &= (1ULL << BIO_ISSUE_SIZE_BITS) - 1;
	issue->value = ((issue->value & BIO_ISSUE_RES_MASK) |
			(ktime_get_ns() & BIO_ISSUE_TIME_MASK) |
			((u64)size << BIO_ISSUE_SIZE_SHIFT));
}

/*
 * main unit of I/O for the block layer and lower layers (ie drivers and
 * stacking drivers)
 */
struct bio {
	struct bio		*bi_next;	/* request queue link */
	struct gendisk		*bi_disk;
	unsigned int		bi_opf;		/* bottom bits req flags,
						 * top bits REQ_OP. Use
						 * accessors.
						 */
	unsigned short		bi_flags;	/* status, etc and bvec pool number */
	unsigned short		bi_ioprio;
	unsigned short		bi_write_hint;
	blk_status_t		bi_status;
	u8			bi_partno;

	struct bvec_iter	bi_iter;

	atomic_t		__bi_remaining;
	bio_end_io_t		*bi_end_io;

	void			*bi_private;
#ifdef CONFIG_BLK_CGROUP
	/*
	 * Represents the association of the css and request_queue for the bio.
	 * If a bio goes direct to device, it will not have a blkg as it will
	 * not have a request_queue associated with it.  The reference is put
	 * on release of the bio.
	 */
	struct blkcg_gq		*bi_blkg;
	struct bio_issue	bi_issue;
#ifdef CONFIG_BLK_CGROUP_IOCOST
	u64			bi_iocost_cost;
#endif
#endif
	union {
#if defined(CONFIG_BLK_DEV_INTEGRITY)
		struct bio_integrity_payload *bi_integrity; /* data integrity */
#endif
	};

	unsigned short		bi_vcnt;	/* how many bio_vec's */

	/*
	 * Everything starting with bi_max_vecs will be preserved by bio_reset()
	 */

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

	atomic_t		__bi_cnt;	/* pin count */

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

	struct bio_set		*bi_pool;

	/*
	 * We can inline a number of vecs at the end of the bio, to avoid
	 * double allocations for a small number of bio_vecs. This member
	 * MUST obviously be kept at the very end of the bio.
	 */
	struct bio_vec		bi_inline_vecs[0];
};

#define BIO_RESET_BYTES		offsetof(struct bio, bi_max_vecs)

/*
 * bio flags
 */
enum {
	BIO_NO_PAGE_REF,	/* don't put release vec pages */
	BIO_CLONED,		/* doesn't own data */
	BIO_BOUNCED,		/* bio is a bounce bio */
	BIO_USER_MAPPED,	/* contains user pages */
	BIO_NULL_MAPPED,	/* contains invalid user pages */
	BIO_WORKINGSET,		/* contains userspace workingset pages */
	BIO_QUIET,		/* Make BIO Quiet */
	BIO_CHAIN,		/* chained bio, ->bi_remaining in effect */
	BIO_REFFED,		/* bio has elevated ->bi_cnt */
	BIO_THROTTLED,		/* This bio has already been subjected to
				 * throttling rules. Don't do it again. */
	BIO_TRACE_COMPLETION,	/* bio_endio() should trace the final completion
				 * of this bio. */
	BIO_QUEUE_ENTERED,	/* can use blk_queue_enter_live() */
	BIO_TRACKED,		/* set if bio goes through the rq_qos path */
	BIO_FLAG_LAST
};

/* See BVEC_POOL_OFFSET below before adding new flags */

/*
 * We support 6 different bvec pools, the last one is magic in that it
 * is backed by a mempool.
 */
#define BVEC_POOL_NR		6
#define BVEC_POOL_MAX		(BVEC_POOL_NR - 1)

/*
 * Top 3 bits of bio flags indicate the pool the bvecs came from.  We add
 * 1 to the actual index so that 0 indicates that there are no bvecs to be
 * freed.
 */
#define BVEC_POOL_BITS		(3)
#define BVEC_POOL_OFFSET	(16 - BVEC_POOL_BITS)
#define BVEC_POOL_IDX(bio)	((bio)->bi_flags >> BVEC_POOL_OFFSET)
#if (1<< BVEC_POOL_BITS) < (BVEC_POOL_NR+1)
# error "BVEC_POOL_BITS is too small"
#endif

/*
 * Flags starting here get preserved by bio_reset() - this includes
 * only BVEC_POOL_IDX()
 */
#define BIO_RESET_BITS	BVEC_POOL_OFFSET

typedef __u32 __bitwise blk_mq_req_flags_t;

/*
 * Operations and flags common to the bio and request structures.
 * We use 8 bits for encoding the operation, and the remaining 24 for flags.
 *
 * The least significant bit of the operation number indicates the data
 * transfer direction:
 *
 *   - if the least significant bit is set transfers are TO the device
 *   - if the least significant bit is not set transfers are FROM the device
 *
 * If a operation does not transfer data the least significant bit has no
 * meaning.
 */
#define REQ_OP_BITS	8
#define REQ_OP_MASK	((1 << REQ_OP_BITS) - 1)
#define REQ_FLAG_BITS	24

enum req_opf {
	/* read sectors from the device */
	REQ_OP_READ		= 0,
	/* write sectors to the device */
	REQ_OP_WRITE		= 1,
	/* flush the volatile write cache */
	REQ_OP_FLUSH		= 2,
	/* discard sectors */
	REQ_OP_DISCARD		= 3,
	/* securely erase sectors */
	REQ_OP_SECURE_ERASE	= 5,
	/* reset a zone write pointer */
	REQ_OP_ZONE_RESET	= 6,
	/* write the same sector many times */
	REQ_OP_WRITE_SAME	= 7,
	/* reset all the zone present on the device */
	REQ_OP_ZONE_RESET_ALL	= 8,
	/* write the zero filled sector many times */
	REQ_OP_WRITE_ZEROES	= 9,

	/* SCSI passthrough using struct scsi_request */
	REQ_OP_SCSI_IN		= 32,
	REQ_OP_SCSI_OUT		= 33,
	/* Driver private requests */
	REQ_OP_DRV_IN		= 34,
	REQ_OP_DRV_OUT		= 35,

	REQ_OP_LAST,
};

enum req_flag_bits {
	__REQ_FAILFAST_DEV =	/* no driver retries of device errors */
		REQ_OP_BITS,
	__REQ_FAILFAST_TRANSPORT, /* no driver retries of transport errors */
	__REQ_FAILFAST_DRIVER,	/* no driver retries of driver errors */
	__REQ_SYNC,		/* request is sync (sync write or read) */
	__REQ_META,		/* metadata io request */
	__REQ_PRIO,		/* boost priority in cfq */
	__REQ_NOMERGE,		/* don't touch this for merging */
	__REQ_IDLE,		/* anticipate more IO after this one */
	__REQ_INTEGRITY,	/* I/O includes block integrity payload */
	__REQ_FUA,		/* forced unit access */
	__REQ_PREFLUSH,		/* request for cache flush */
	__REQ_RAHEAD,		/* read ahead, can fail anytime */
	__REQ_BACKGROUND,	/* background IO */
	__REQ_NOWAIT,           /* Don't wait if request will block */
	__REQ_NOWAIT_INLINE,	/* Return would-block error inline */
	/*
	 * When a shared kthread needs to issue a bio for a cgroup, doing
	 * so synchronously can lead to priority inversions as the kthread
	 * can be trapped waiting for that cgroup.  CGROUP_PUNT flag makes
	 * submit_bio() punt the actual issuing to a dedicated per-blkcg
	 * work item to avoid such priority inversions.
	 */
	__REQ_CGROUP_PUNT,

	/* command specific flags for REQ_OP_WRITE_ZEROES: */
	__REQ_NOUNMAP,		/* do not free blocks when zeroing */

	__REQ_HIPRI,

	/* for driver use */
	__REQ_DRV,
	__REQ_SWAP,		/* swapping request. */
	__REQ_NR_BITS,		/* stops here */
};

#define REQ_FAILFAST_DEV	(1ULL << __REQ_FAILFAST_DEV)
#define REQ_FAILFAST_TRANSPORT	(1ULL << __REQ_FAILFAST_TRANSPORT)
#define REQ_FAILFAST_DRIVER	(1ULL << __REQ_FAILFAST_DRIVER)
#define REQ_SYNC		(1ULL << __REQ_SYNC)
#define REQ_META		(1ULL << __REQ_META)
#define REQ_PRIO		(1ULL << __REQ_PRIO)
#define REQ_NOMERGE		(1ULL << __REQ_NOMERGE)
#define REQ_IDLE		(1ULL << __REQ_IDLE)
#define REQ_INTEGRITY		(1ULL << __REQ_INTEGRITY)
#define REQ_FUA			(1ULL << __REQ_FUA)
#define REQ_PREFLUSH		(1ULL << __REQ_PREFLUSH)
#define REQ_RAHEAD		(1ULL << __REQ_RAHEAD)
#define REQ_BACKGROUND		(1ULL << __REQ_BACKGROUND)
#define REQ_NOWAIT		(1ULL << __REQ_NOWAIT)
#define REQ_NOWAIT_INLINE	(1ULL << __REQ_NOWAIT_INLINE)
#define REQ_CGROUP_PUNT		(1ULL << __REQ_CGROUP_PUNT)

#define REQ_NOUNMAP		(1ULL << __REQ_NOUNMAP)
#define REQ_HIPRI		(1ULL << __REQ_HIPRI)

#define REQ_DRV			(1ULL << __REQ_DRV)
#define REQ_SWAP		(1ULL << __REQ_SWAP)

#define REQ_FAILFAST_MASK \
	(REQ_FAILFAST_DEV | REQ_FAILFAST_TRANSPORT | REQ_FAILFAST_DRIVER)

#define REQ_NOMERGE_FLAGS \
	(REQ_NOMERGE | REQ_PREFLUSH | REQ_FUA)

enum stat_group {
	STAT_READ,
	STAT_WRITE,
	STAT_DISCARD,

	NR_STAT_GROUPS
};

#define bio_op(bio) \
	((bio)->bi_opf & REQ_OP_MASK)
#define req_op(req) \
	((req)->cmd_flags & REQ_OP_MASK)

/* obsolete, don't use in new code */
static inline void bio_set_op_attrs(struct bio *bio, unsigned op,
		unsigned op_flags)
{
	bio->bi_opf = op | op_flags;
}

static inline bool op_is_write(unsigned int op)
{
	return (op & 1);
}

/*
 * Check if the bio or request is one that needs special treatment in the
 * flush state machine.
 */
static inline bool op_is_flush(unsigned int op)
{
	return op & (REQ_FUA | REQ_PREFLUSH);
}

/*
 * Reads are always treated as synchronous, as are requests with the FUA or
 * PREFLUSH flag.  Other operations may be marked as synchronous using the
 * REQ_SYNC flag.
 */
static inline bool op_is_sync(unsigned int op)
{
	return (op & REQ_OP_MASK) == REQ_OP_READ ||
		(op & (REQ_SYNC | REQ_FUA | REQ_PREFLUSH));
}

static inline bool op_is_discard(unsigned int op)
{
	return (op & REQ_OP_MASK) == REQ_OP_DISCARD;
}

static inline int op_stat_group(unsigned int op)
{
	if (op_is_discard(op))
		return STAT_DISCARD;
	return op_is_write(op);
}

typedef unsigned int blk_qc_t;
#define BLK_QC_T_NONE		-1U
#define BLK_QC_T_EAGAIN		-2U
#define BLK_QC_T_SHIFT		16
#define BLK_QC_T_INTERNAL	(1U << 31)

static inline bool blk_qc_t_valid(blk_qc_t cookie)
{
	return cookie != BLK_QC_T_NONE && cookie != BLK_QC_T_EAGAIN;
}

static inline unsigned int blk_qc_t_to_queue_num(blk_qc_t cookie)
{
	return (cookie & ~BLK_QC_T_INTERNAL) >> BLK_QC_T_SHIFT;
}

static inline unsigned int blk_qc_t_to_tag(blk_qc_t cookie)
{
	return cookie & ((1u << BLK_QC_T_SHIFT) - 1);
}

static inline bool blk_qc_t_is_internal(blk_qc_t cookie)
{
	return (cookie & BLK_QC_T_INTERNAL) != 0;
}

struct blk_rq_stat {
	u64 mean;
	u64 min;
	u64 max;
	u32 nr_samples;
	u64 batch;
};

#endif /* __LINUX_BLK_TYPES_H */
Commit	Line	Data
b2441318	1	/* SPDX-License-Identifier: GPL-2.0 */
7cc01581 TH	2	/*
	3	* Block data types and constants. Directly include this file only to
	4	* break include dependency loop.
	5	*/
	6	#ifndef __LINUX_BLK_TYPES_H
	7	#define __LINUX_BLK_TYPES_H
	8
7cc01581	9	#include <linux/types.h>
0781e79e	10	#include <linux/bvec.h>
5238dcf4	11	#include <linux/ktime.h>
7cc01581 TH	12
	13	struct bio_set;
	14	struct bio;
	15	struct bio_integrity_payload;
	16	struct page;
	17	struct block_device;
852c788f TH	18	struct io_context;
852c788f TH	19	struct cgroup_subsys_state;
4246a0b6	20	typedef void (bio_end_io_t) (struct bio *);
7cc01581	21
2a842aca CH	22	/*
2a842aca CH	23	* Block error status values. See block/blk-core:blk_errors for the details.
6e2fb221	24	* Alpha cannot write a byte atomically, so we need to use 32-bit value.
2a842aca	25	*/
6e2fb221 MP	26	#if defined(CONFIG_ALPHA) && !defined(__alpha_bwx__)
	27	typedef u32 __bitwise blk_status_t;
	28	#else
2a842aca	29	typedef u8 __bitwise blk_status_t;
6e2fb221	30	#endif
2a842aca CH	31	#define BLK_STS_OK 0
	32	#define BLK_STS_NOTSUPP ((__force blk_status_t)1)
	33	#define BLK_STS_TIMEOUT ((__force blk_status_t)2)
	34	#define BLK_STS_NOSPC ((__force blk_status_t)3)
	35	#define BLK_STS_TRANSPORT ((__force blk_status_t)4)
	36	#define BLK_STS_TARGET ((__force blk_status_t)5)
	37	#define BLK_STS_NEXUS ((__force blk_status_t)6)
	38	#define BLK_STS_MEDIUM ((__force blk_status_t)7)
	39	#define BLK_STS_PROTECTION ((__force blk_status_t)8)
	40	#define BLK_STS_RESOURCE ((__force blk_status_t)9)
	41	#define BLK_STS_IOERR ((__force blk_status_t)10)
	42
4e4cbee9 CH	43	/* hack for device mapper, don't use elsewhere: */
	44	#define BLK_STS_DM_REQUEUE ((__force blk_status_t)11)
	45
03a07c92 GR	46	#define BLK_STS_AGAIN ((__force blk_status_t)12)
03a07c92 GR	47
86ff7c2a ML	48	/*
	49	* BLK_STS_DEV_RESOURCE is returned from the driver to the block layer if
	50	* device related resources are unavailable, but the driver can guarantee
	51	* that the queue will be rerun in the future once resources become
	52	* available again. This is typically the case for device specific
	53	* resources that are consumed for IO. If the driver fails allocating these
	54	* resources, we know that inflight (or pending) IO will free these
	55	* resource upon completion.
	56	*
	57	* This is different from BLK_STS_RESOURCE in that it explicitly references
	58	* a device specific resource. For resources of wider scope, allocation
	59	* failure can happen without having pending IO. This means that we can't
	60	* rely on request completions freeing these resources, as IO may not be in
	61	* flight. Examples of that are kernel memory allocations, DMA mappings, or
	62	* any other system wide resources.
	63	*/
	64	#define BLK_STS_DEV_RESOURCE ((__force blk_status_t)13)
	65
9111e568 KB	66	/**
	67	* blk_path_error - returns true if error may be path related
	68	* @error: status the request was completed with
	69	*
	70	* Description:
	71	* This classifies block error status into non-retryable errors and ones
	72	* that may be successful if retried on a failover path.
	73	*
	74	* Return:
	75	* %false - retrying failover path will not help
	76	* %true - may succeed if retried
	77	*/
	78	static inline bool blk_path_error(blk_status_t error)
	79	{
	80	switch (error) {
	81	case BLK_STS_NOTSUPP:
	82	case BLK_STS_NOSPC:
	83	case BLK_STS_TARGET:
	84	case BLK_STS_NEXUS:
	85	case BLK_STS_MEDIUM:
	86	case BLK_STS_PROTECTION:
	87	return false;
	88	}
	89
	90	/* Anything else could be a path failure, so should be retried */
	91	return true;
	92	}
	93
5238dcf4 OS	94	/*
	95	* From most significant bit:
	96	* 1 bit: reserved for other usage, see below
	97	* 12 bits: original size of bio
	98	* 51 bits: issue time of bio
	99	*/
	100	#define BIO_ISSUE_RES_BITS 1
	101	#define BIO_ISSUE_SIZE_BITS 12
	102	#define BIO_ISSUE_RES_SHIFT (64 - BIO_ISSUE_RES_BITS)
	103	#define BIO_ISSUE_SIZE_SHIFT (BIO_ISSUE_RES_SHIFT - BIO_ISSUE_SIZE_BITS)
	104	#define BIO_ISSUE_TIME_MASK ((1ULL << BIO_ISSUE_SIZE_SHIFT) - 1)
	105	#define BIO_ISSUE_SIZE_MASK \
	106	(((1ULL << BIO_ISSUE_SIZE_BITS) - 1) << BIO_ISSUE_SIZE_SHIFT)
	107	#define BIO_ISSUE_RES_MASK (~((1ULL << BIO_ISSUE_RES_SHIFT) - 1))
	108
	109	/* Reserved bit for blk-throtl */
	110	#define BIO_ISSUE_THROTL_SKIP_LATENCY (1ULL << 63)
	111
	112	struct bio_issue {
	113	u64 value;
	114	};
	115
	116	static inline u64 __bio_issue_time(u64 time)
	117	{
	118	return time & BIO_ISSUE_TIME_MASK;
	119	}
	120
	121	static inline u64 bio_issue_time(struct bio_issue *issue)
	122	{
	123	return __bio_issue_time(issue->value);
	124	}
	125
	126	static inline sector_t bio_issue_size(struct bio_issue *issue)
	127	{
	128	return ((issue->value & BIO_ISSUE_SIZE_MASK) >> BIO_ISSUE_SIZE_SHIFT);
	129	}
	130
	131	static inline void bio_issue_init(struct bio_issue *issue,
	132	sector_t size)
	133	{
	134	size &= (1ULL << BIO_ISSUE_SIZE_BITS) - 1;
	135	issue->value = ((issue->value & BIO_ISSUE_RES_MASK) \|
	136	(ktime_get_ns() & BIO_ISSUE_TIME_MASK) \|
	137	((u64)size << BIO_ISSUE_SIZE_SHIFT));
	138	}
	139
7cc01581 TH	140	/*
	141	* main unit of I/O for the block layer and lower layers (ie drivers and
	142	* stacking drivers)
	143	*/
	144	struct bio {
7cc01581	145	struct bio bi_next; / request queue link */
74d46992	146	struct gendisk *bi_disk;
1eff9d32 JA	147	unsigned int bi_opf; /* bottom bits req flags,
	148	* top bits REQ_OP. Use
	149	* accessors.
4e1b2d52	150	*/
dbde775c	151	unsigned short bi_flags; /* status, etc and bvec pool number */
43b62ce3	152	unsigned short bi_ioprio;
cb6934f8	153	unsigned short bi_write_hint;
111be883 SL	154	blk_status_t bi_status;
111be883 SL	155	u8 bi_partno;
7cc01581	156
111be883	157	struct bvec_iter bi_iter;
196d38bc	158
111be883	159	atomic_t __bi_remaining;
7cc01581 TH	160	bio_end_io_t *bi_end_io;
	161
	162	void *bi_private;
852c788f TH	163	#ifdef CONFIG_BLK_CGROUP
852c788f TH	164	/*
db6638d7 DZ	165	* Represents the association of the css and request_queue for the bio.
	166	* If a bio goes direct to device, it will not have a blkg as it will
	167	* not have a request_queue associated with it. The reference is put
	168	* on release of the bio.
852c788f	169	*/
08e18eab	170	struct blkcg_gq *bi_blkg;
5238dcf4	171	struct bio_issue bi_issue;
7caa4715 TH	172	#ifdef CONFIG_BLK_CGROUP_IOCOST
	173	u64 bi_iocost_cost;
	174	#endif
852c788f	175	#endif
180b2f95	176	union {
7cc01581	177	#if defined(CONFIG_BLK_DEV_INTEGRITY)
180b2f95	178	struct bio_integrity_payload bi_integrity; / data integrity */
7cc01581	179	#endif
180b2f95	180	};
7cc01581	181
4f024f37 KO	182	unsigned short bi_vcnt; /* how many bio_vec's */
4f024f37 KO	183
f44b48c7 KO	184	/*
	185	* Everything starting with bi_max_vecs will be preserved by bio_reset()
	186	*/
	187
4f024f37	188	unsigned short bi_max_vecs; /* max bvl_vecs we can hold */
f44b48c7	189
dac56212	190	atomic_t __bi_cnt; /* pin count */
f44b48c7 KO	191
	192	struct bio_vec bi_io_vec; / the actual vec list */
	193
395c72a7 KO	194	struct bio_set *bi_pool;
395c72a7 KO	195
7cc01581 TH	196	/*
	197	* We can inline a number of vecs at the end of the bio, to avoid
	198	* double allocations for a small number of bio_vecs. This member
	199	* MUST obviously be kept at the very end of the bio.
	200	*/
	201	struct bio_vec bi_inline_vecs[0];
	202	};
	203
f44b48c7 KO	204	#define BIO_RESET_BYTES offsetof(struct bio, bi_max_vecs)
f44b48c7 KO	205
7cc01581 TH	206	/*
	207	* bio flags
	208	*/
2b24e6f6 JT	209	enum {
2b24e6f6 JT	210	BIO_NO_PAGE_REF, /* don't put release vec pages */
2b24e6f6 JT	211	BIO_CLONED, /* doesn't own data */
	212	BIO_BOUNCED, /* bio is a bounce bio */
	213	BIO_USER_MAPPED, /* contains user pages */
	214	BIO_NULL_MAPPED, /* contains invalid user pages */
b8e24a93	215	BIO_WORKINGSET, /* contains userspace workingset pages */
2b24e6f6 JT	216	BIO_QUIET, /* Make BIO Quiet */
	217	BIO_CHAIN, /* chained bio, ->bi_remaining in effect */
	218	BIO_REFFED, /* bio has elevated ->bi_cnt */
	219	BIO_THROTTLED, /* This bio has already been subjected to
8d2bbd4c	220	* throttling rules. Don't do it again. */
2b24e6f6	221	BIO_TRACE_COMPLETION, /* bio_endio() should trace the final completion
fbbaf700	222	* of this bio. */
2b24e6f6 JT	223	BIO_QUEUE_ENTERED, /* can use blk_queue_enter_live() */
	224	BIO_TRACKED, /* set if bio goes through the rq_qos path */
	225	BIO_FLAG_LAST
	226	};
cd4a4ae4	227
dbde775c	228	/* See BVEC_POOL_OFFSET below before adding new flags */
f44b48c7	229
7cc01581	230	/*
ed996a52 CH	231	* We support 6 different bvec pools, the last one is magic in that it
ed996a52 CH	232	* is backed by a mempool.
7cc01581	233	*/
ed996a52 CH	234	#define BVEC_POOL_NR 6
	235	#define BVEC_POOL_MAX (BVEC_POOL_NR - 1)
	236
	237	/*
dbde775c	238	* Top 3 bits of bio flags indicate the pool the bvecs came from. We add
ed996a52 CH	239	* 1 to the actual index so that 0 indicates that there are no bvecs to be
	240	* freed.
	241	*/
dbde775c	242	#define BVEC_POOL_BITS (3)
c0acf12a	243	#define BVEC_POOL_OFFSET (16 - BVEC_POOL_BITS)
ed996a52	244	#define BVEC_POOL_IDX(bio) ((bio)->bi_flags >> BVEC_POOL_OFFSET)
dbde775c N	245	#if (1<< BVEC_POOL_BITS) < (BVEC_POOL_NR+1)
	246	# error "BVEC_POOL_BITS is too small"
	247	#endif
	248
	249	/*
	250	* Flags starting here get preserved by bio_reset() - this includes
	251	* only BVEC_POOL_IDX()
	252	*/
	253	#define BIO_RESET_BITS BVEC_POOL_OFFSET
7cc01581	254
9a95e4ef BVA	255	typedef __u32 __bitwise blk_mq_req_flags_t;
9a95e4ef BVA	256
7cc01581	257	/*
ef295ecf CH	258	* Operations and flags common to the bio and request structures.
ef295ecf CH	259	* We use 8 bits for encoding the operation, and the remaining 24 for flags.
87374179 CH	260	*
	261	* The least significant bit of the operation number indicates the data
	262	* transfer direction:
	263	*
	264	* - if the least significant bit is set transfers are TO the device
	265	* - if the least significant bit is not set transfers are FROM the device
	266	*
	267	* If a operation does not transfer data the least significant bit has no
	268	* meaning.
7cc01581	269	*/
ef295ecf CH	270	#define REQ_OP_BITS 8
	271	#define REQ_OP_MASK ((1 << REQ_OP_BITS) - 1)
	272	#define REQ_FLAG_BITS 24
	273
	274	enum req_opf {
87374179 CH	275	/* read sectors from the device */
	276	REQ_OP_READ = 0,
	277	/* write sectors to the device */
	278	REQ_OP_WRITE = 1,
	279	/* flush the volatile write cache */
	280	REQ_OP_FLUSH = 2,
	281	/* discard sectors */
	282	REQ_OP_DISCARD = 3,
87374179 CH	283	/* securely erase sectors */
87374179 CH	284	REQ_OP_SECURE_ERASE = 5,
8367de2c	285	/* reset a zone write pointer */
87374179 CH	286	REQ_OP_ZONE_RESET = 6,
	287	/* write the same sector many times */
	288	REQ_OP_WRITE_SAME = 7,
e84e8f06 CK	289	/* reset all the zone present on the device */
e84e8f06 CK	290	REQ_OP_ZONE_RESET_ALL = 8,
a6f0788e	291	/* write the zero filled sector many times */
1d62ac13	292	REQ_OP_WRITE_ZEROES = 9,
ef295ecf	293
aebf526b CH	294	/* SCSI passthrough using struct scsi_request */
	295	REQ_OP_SCSI_IN = 32,
	296	REQ_OP_SCSI_OUT = 33,
	297	/* Driver private requests */
	298	REQ_OP_DRV_IN = 34,
	299	REQ_OP_DRV_OUT = 35,
	300
ef295ecf CH	301	REQ_OP_LAST,
	302	};
	303
	304	enum req_flag_bits {
	305	__REQ_FAILFAST_DEV = /* no driver retries of device errors */
	306	REQ_OP_BITS,
7cc01581 TH	307	__REQ_FAILFAST_TRANSPORT, /* no driver retries of transport errors */
7cc01581 TH	308	__REQ_FAILFAST_DRIVER, /* no driver retries of driver errors */
7cc01581 TH	309	__REQ_SYNC, /* request is sync (sync write or read) */
7cc01581 TH	310	__REQ_META, /* metadata io request */
65299a3b	311	__REQ_PRIO, /* boost priority in cfq */
bd1c1c21	312	__REQ_NOMERGE, /* don't touch this for merging */
a2b80967	313	__REQ_IDLE, /* anticipate more IO after this one */
180b2f95	314	__REQ_INTEGRITY, /* I/O includes block integrity payload */
8e4bf844	315	__REQ_FUA, /* forced unit access */
28a8f0d3	316	__REQ_PREFLUSH, /* request for cache flush */
188bd2b1	317	__REQ_RAHEAD, /* read ahead, can fail anytime */
1d796d6a	318	__REQ_BACKGROUND, /* background IO */
8977f563	319	__REQ_NOWAIT, /* Don't wait if request will block */
893a1c97	320	__REQ_NOWAIT_INLINE, /* Return would-block error inline */
d3f77dfd TH	321	/*
	322	* When a shared kthread needs to issue a bio for a cgroup, doing
	323	* so synchronously can lead to priority inversions as the kthread
	324	* can be trapped waiting for that cgroup. CGROUP_PUNT flag makes
	325	* submit_bio() punt the actual issuing to a dedicated per-blkcg
	326	* work item to avoid such priority inversions.
	327	*/
	328	__REQ_CGROUP_PUNT,
d928be9f CH	329
	330	/* command specific flags for REQ_OP_WRITE_ZEROES: */
	331	__REQ_NOUNMAP, /* do not free blocks when zeroing */
	332
d1e36282 JA	333	__REQ_HIPRI,
d1e36282 JA	334
96222bcc CH	335	/* for driver use */
96222bcc CH	336	__REQ_DRV,
0d1e0c7c	337	__REQ_SWAP, /* swapping request. */
7cc01581 TH	338	__REQ_NR_BITS, /* stops here */
	339	};
	340
5953316d JA	341	#define REQ_FAILFAST_DEV (1ULL << __REQ_FAILFAST_DEV)
	342	#define REQ_FAILFAST_TRANSPORT (1ULL << __REQ_FAILFAST_TRANSPORT)
	343	#define REQ_FAILFAST_DRIVER (1ULL << __REQ_FAILFAST_DRIVER)
	344	#define REQ_SYNC (1ULL << __REQ_SYNC)
	345	#define REQ_META (1ULL << __REQ_META)
	346	#define REQ_PRIO (1ULL << __REQ_PRIO)
ef295ecf	347	#define REQ_NOMERGE (1ULL << __REQ_NOMERGE)
a2b80967	348	#define REQ_IDLE (1ULL << __REQ_IDLE)
180b2f95	349	#define REQ_INTEGRITY (1ULL << __REQ_INTEGRITY)
ef295ecf CH	350	#define REQ_FUA (1ULL << __REQ_FUA)
	351	#define REQ_PREFLUSH (1ULL << __REQ_PREFLUSH)
	352	#define REQ_RAHEAD (1ULL << __REQ_RAHEAD)
1d796d6a	353	#define REQ_BACKGROUND (1ULL << __REQ_BACKGROUND)
8977f563	354	#define REQ_NOWAIT (1ULL << __REQ_NOWAIT)
893a1c97	355	#define REQ_NOWAIT_INLINE (1ULL << __REQ_NOWAIT_INLINE)
d3f77dfd TH	356	#define REQ_CGROUP_PUNT (1ULL << __REQ_CGROUP_PUNT)
d3f77dfd TH	357
d928be9f	358	#define REQ_NOUNMAP (1ULL << __REQ_NOUNMAP)
d1e36282	359	#define REQ_HIPRI (1ULL << __REQ_HIPRI)
d928be9f	360
96222bcc	361	#define REQ_DRV (1ULL << __REQ_DRV)
0d1e0c7c	362	#define REQ_SWAP (1ULL << __REQ_SWAP)
d928be9f	363
7cc01581 TH	364	#define REQ_FAILFAST_MASK \
7cc01581 TH	365	(REQ_FAILFAST_DEV \| REQ_FAILFAST_TRANSPORT \| REQ_FAILFAST_DRIVER)
7cc01581	366
e2a60da7	367	#define REQ_NOMERGE_FLAGS \
e8064021	368	(REQ_NOMERGE \| REQ_PREFLUSH \| REQ_FUA)
e2a60da7	369
dbae2c55 MC	370	enum stat_group {
	371	STAT_READ,
	372	STAT_WRITE,
bdca3c87	373	STAT_DISCARD,
dbae2c55 MC	374
	375	NR_STAT_GROUPS
	376	};
	377
ef295ecf CH	378	#define bio_op(bio) \
	379	((bio)->bi_opf & REQ_OP_MASK)
	380	#define req_op(req) \
	381	((req)->cmd_flags & REQ_OP_MASK)
7cc01581	382
ef295ecf	383	/* obsolete, don't use in new code */
93c5bdf7 CH	384	static inline void bio_set_op_attrs(struct bio *bio, unsigned op,
	385	unsigned op_flags)
	386	{
	387	bio->bi_opf = op \| op_flags;
	388	}
c11f0c0b	389
87374179 CH	390	static inline bool op_is_write(unsigned int op)
	391	{
	392	return (op & 1);
	393	}
	394
f73f44eb CH	395	/*
	396	* Check if the bio or request is one that needs special treatment in the
	397	* flush state machine.
	398	*/
	399	static inline bool op_is_flush(unsigned int op)
	400	{
	401	return op & (REQ_FUA \| REQ_PREFLUSH);
	402	}
	403
b685d3d6 CH	404	/*
	405	* Reads are always treated as synchronous, as are requests with the FUA or
	406	* PREFLUSH flag. Other operations may be marked as synchronous using the
	407	* REQ_SYNC flag.
	408	*/
ef295ecf CH	409	static inline bool op_is_sync(unsigned int op)
ef295ecf CH	410	{
b685d3d6 CH	411	return (op & REQ_OP_MASK) == REQ_OP_READ \|\|
b685d3d6 CH	412	(op & (REQ_SYNC \| REQ_FUA \| REQ_PREFLUSH));
ef295ecf	413	}
c11f0c0b	414
bdca3c87 MC	415	static inline bool op_is_discard(unsigned int op)
	416	{
	417	return (op & REQ_OP_MASK) == REQ_OP_DISCARD;
	418	}
	419
ddcf35d3 MC	420	static inline int op_stat_group(unsigned int op)
ddcf35d3 MC	421	{
bdca3c87 MC	422	if (op_is_discard(op))
bdca3c87 MC	423	return STAT_DISCARD;
ddcf35d3 MC	424	return op_is_write(op);
	425	}
	426
dece1635	427	typedef unsigned int blk_qc_t;
fd2d3326	428	#define BLK_QC_T_NONE -1U
893a1c97	429	#define BLK_QC_T_EAGAIN -2U
fd2d3326 JA	430	#define BLK_QC_T_SHIFT 16
fd2d3326 JA	431	#define BLK_QC_T_INTERNAL (1U << 31)
dece1635 JA	432
	433	static inline bool blk_qc_t_valid(blk_qc_t cookie)
	434	{
893a1c97	435	return cookie != BLK_QC_T_NONE && cookie != BLK_QC_T_EAGAIN;
dece1635 JA	436	}
dece1635 JA	437
dece1635 JA	438	static inline unsigned int blk_qc_t_to_queue_num(blk_qc_t cookie)
dece1635 JA	439	{
fd2d3326	440	return (cookie & ~BLK_QC_T_INTERNAL) >> BLK_QC_T_SHIFT;
dece1635 JA	441	}
	442
	443	static inline unsigned int blk_qc_t_to_tag(blk_qc_t cookie)
	444	{
e3a7a3bf	445	return cookie & ((1u << BLK_QC_T_SHIFT) - 1);
dece1635 JA	446	}
dece1635 JA	447
fd2d3326 JA	448	static inline bool blk_qc_t_is_internal(blk_qc_t cookie)
	449	{
	450	return (cookie & BLK_QC_T_INTERNAL) != 0;
	451	}
	452
cf43e6be	453	struct blk_rq_stat {
eca8b53a	454	u64 mean;
cf43e6be JA	455	u64 min;
cf43e6be JA	456	u64 max;
eca8b53a	457	u32 nr_samples;
cf43e6be	458	u64 batch;
cf43e6be JA	459	};
cf43e6be JA	460
7cc01581	461	#endif /* __LINUX_BLK_TYPES_H */