[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 io_context;
struct cgroup_subsys_state;
typedef void (bio_end_io_t) (struct bio *);
struct bio_crypt_ctx;

struct block_device {
	dev_t			bd_dev;
	int			bd_openers;
	struct inode *		bd_inode;	/* will die */
	struct super_block *	bd_super;
	struct mutex		bd_mutex;	/* open/close mutex */
	void *			bd_claiming;
	void *			bd_holder;
	int			bd_holders;
	bool			bd_write_holder;
#ifdef CONFIG_SYSFS
	struct list_head	bd_holder_disks;
#endif
	struct block_device *	bd_contains;
	u8			bd_partno;
	struct hd_struct *	bd_part;
	/* number of times partitions within this device have been opened. */
	unsigned		bd_part_count;

	spinlock_t		bd_size_lock; /* for bd_inode->i_size updates */
	struct gendisk *	bd_disk;
	struct backing_dev_info *bd_bdi;

	/* The counter of freeze processes */
	int			bd_fsfreeze_count;
	/* Mutex for freeze */
	struct mutex		bd_fsfreeze_mutex;
	struct super_block	*bd_fsfreeze_sb;
} __randomize_layout;

/*
 * 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_STS_ZONE_RESOURCE is returned from the driver to the block layer if zone
 * related resources are unavailable, but the driver can guarantee the queue
 * will be rerun in the future once the resources become available again.
 *
 * This is different from BLK_STS_DEV_RESOURCE in that it explicitly references
 * a zone specific resource and IO to a different zone on the same device could
 * still be served. Examples of that are zones that are write-locked, but a read
 * to the same zone could be served.
 */
#define BLK_STS_ZONE_RESOURCE	((__force blk_status_t)14)

/*
 * BLK_STS_ZONE_OPEN_RESOURCE is returned from the driver in the completion
 * path if the device returns a status indicating that too many zone resources
 * are currently open. The same command should be successful if resubmitted
 * after the number of open zones decreases below the device's limits, which is
 * reported in the request_queue's max_open_zones.
 */
#define BLK_STS_ZONE_OPEN_RESOURCE	((__force blk_status_t)15)

/*
 * BLK_STS_ZONE_ACTIVE_RESOURCE is returned from the driver in the completion
 * path if the device returns a status indicating that too many zone resources
 * are currently active. The same command should be successful if resubmitted
 * after the number of active zones decreases below the device's limits, which
 * is reported in the request_queue's max_active_zones.
 */
#define BLK_STS_ZONE_ACTIVE_RESOURCE	((__force blk_status_t)16)

/**
 * 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;
	atomic_t		__bi_remaining;

	struct bvec_iter	bi_iter;

	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

#ifdef CONFIG_BLK_INLINE_ENCRYPTION
	struct bio_crypt_ctx	*bi_crypt_context;
#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[];
};

#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_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_CGROUP_ACCT,	/* has been accounted to a cgroup */
	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,
	/* write the same sector many times */
	REQ_OP_WRITE_SAME	= 7,
	/* write the zero filled sector many times */
	REQ_OP_WRITE_ZEROES	= 9,
	/* Open a zone */
	REQ_OP_ZONE_OPEN	= 10,
	/* Close a zone */
	REQ_OP_ZONE_CLOSE	= 11,
	/* Transition a zone to full */
	REQ_OP_ZONE_FINISH	= 12,
	/* write data at the current zone write pointer */
	REQ_OP_ZONE_APPEND	= 13,
	/* reset a zone write pointer */
	REQ_OP_ZONE_RESET	= 15,
	/* reset all the zone present on the device */
	REQ_OP_ZONE_RESET_ALL	= 17,

	/* 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 */
	/*
	 * 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_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,
	STAT_FLUSH,

	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;
}

/*
 * Check if a bio or request operation is a zone management operation, with
 * the exception of REQ_OP_ZONE_RESET_ALL which is treated as a special case
 * due to its different handling in the block layer and device response in
 * case of command failure.
 */
static inline bool op_is_zone_mgmt(enum req_opf op)
{
	switch (op & REQ_OP_MASK) {
	case REQ_OP_ZONE_RESET:
	case REQ_OP_ZONE_OPEN:
	case REQ_OP_ZONE_CLOSE:
	case REQ_OP_ZONE_FINISH:
		return true;
	default:
		return false;
	}
}

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_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;
}

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;
852c788f TH	17	struct io_context;
852c788f TH	18	struct cgroup_subsys_state;
4246a0b6	19	typedef void (bio_end_io_t) (struct bio *);
a892c8d5	20	struct bio_crypt_ctx;
7cc01581	21
621c1f42	22	struct block_device {
46d40cfa	23	dev_t bd_dev;
621c1f42 CH	24	int bd_openers;
	25	struct inode * bd_inode; /* will die */
	26	struct super_block * bd_super;
	27	struct mutex bd_mutex; /* open/close mutex */
	28	void * bd_claiming;
	29	void * bd_holder;
	30	int bd_holders;
	31	bool bd_write_holder;
	32	#ifdef CONFIG_SYSFS
	33	struct list_head bd_holder_disks;
	34	#endif
	35	struct block_device * bd_contains;
621c1f42 CH	36	u8 bd_partno;
	37	struct hd_struct * bd_part;
	38	/* number of times partitions within this device have been opened. */
	39	unsigned bd_part_count;
38430f08	40
c2b4bb8c	41	spinlock_t bd_size_lock; /* for bd_inode->i_size updates */
621c1f42	42	struct gendisk * bd_disk;
621c1f42	43	struct backing_dev_info *bd_bdi;
621c1f42 CH	44
	45	/* The counter of freeze processes */
	46	int bd_fsfreeze_count;
	47	/* Mutex for freeze */
	48	struct mutex bd_fsfreeze_mutex;
040f04bd	49	struct super_block *bd_fsfreeze_sb;
621c1f42 CH	50	} __randomize_layout;
621c1f42 CH	51
2a842aca CH	52	/*
2a842aca CH	53	* Block error status values. See block/blk-core:blk_errors for the details.
6e2fb221	54	* Alpha cannot write a byte atomically, so we need to use 32-bit value.
2a842aca	55	*/
6e2fb221 MP	56	#if defined(CONFIG_ALPHA) && !defined(__alpha_bwx__)
	57	typedef u32 __bitwise blk_status_t;
	58	#else
2a842aca	59	typedef u8 __bitwise blk_status_t;
6e2fb221	60	#endif
2a842aca CH	61	#define BLK_STS_OK 0
	62	#define BLK_STS_NOTSUPP ((__force blk_status_t)1)
	63	#define BLK_STS_TIMEOUT ((__force blk_status_t)2)
	64	#define BLK_STS_NOSPC ((__force blk_status_t)3)
	65	#define BLK_STS_TRANSPORT ((__force blk_status_t)4)
	66	#define BLK_STS_TARGET ((__force blk_status_t)5)
	67	#define BLK_STS_NEXUS ((__force blk_status_t)6)
	68	#define BLK_STS_MEDIUM ((__force blk_status_t)7)
	69	#define BLK_STS_PROTECTION ((__force blk_status_t)8)
	70	#define BLK_STS_RESOURCE ((__force blk_status_t)9)
	71	#define BLK_STS_IOERR ((__force blk_status_t)10)
	72
4e4cbee9 CH	73	/* hack for device mapper, don't use elsewhere: */
	74	#define BLK_STS_DM_REQUEUE ((__force blk_status_t)11)
	75
03a07c92 GR	76	#define BLK_STS_AGAIN ((__force blk_status_t)12)
03a07c92 GR	77
86ff7c2a ML	78	/*
	79	* BLK_STS_DEV_RESOURCE is returned from the driver to the block layer if
	80	* device related resources are unavailable, but the driver can guarantee
	81	* that the queue will be rerun in the future once resources become
	82	* available again. This is typically the case for device specific
	83	* resources that are consumed for IO. If the driver fails allocating these
	84	* resources, we know that inflight (or pending) IO will free these
	85	* resource upon completion.
	86	*
	87	* This is different from BLK_STS_RESOURCE in that it explicitly references
	88	* a device specific resource. For resources of wider scope, allocation
	89	* failure can happen without having pending IO. This means that we can't
	90	* rely on request completions freeing these resources, as IO may not be in
	91	* flight. Examples of that are kernel memory allocations, DMA mappings, or
	92	* any other system wide resources.
	93	*/
	94	#define BLK_STS_DEV_RESOURCE ((__force blk_status_t)13)
	95
0512a75b KB	96	/*
	97	* BLK_STS_ZONE_RESOURCE is returned from the driver to the block layer if zone
	98	* related resources are unavailable, but the driver can guarantee the queue
	99	* will be rerun in the future once the resources become available again.
	100	*
	101	* This is different from BLK_STS_DEV_RESOURCE in that it explicitly references
	102	* a zone specific resource and IO to a different zone on the same device could
	103	* still be served. Examples of that are zones that are write-locked, but a read
	104	* to the same zone could be served.
	105	*/
	106	#define BLK_STS_ZONE_RESOURCE ((__force blk_status_t)14)
	107
3b481d91 KB	108	/*
	109	* BLK_STS_ZONE_OPEN_RESOURCE is returned from the driver in the completion
	110	* path if the device returns a status indicating that too many zone resources
	111	* are currently open. The same command should be successful if resubmitted
	112	* after the number of open zones decreases below the device's limits, which is
	113	* reported in the request_queue's max_open_zones.
	114	*/
	115	#define BLK_STS_ZONE_OPEN_RESOURCE ((__force blk_status_t)15)
	116
	117	/*
	118	* BLK_STS_ZONE_ACTIVE_RESOURCE is returned from the driver in the completion
	119	* path if the device returns a status indicating that too many zone resources
	120	* are currently active. The same command should be successful if resubmitted
	121	* after the number of active zones decreases below the device's limits, which
	122	* is reported in the request_queue's max_active_zones.
	123	*/
	124	#define BLK_STS_ZONE_ACTIVE_RESOURCE ((__force blk_status_t)16)
	125
9111e568 KB	126	/**
	127	* blk_path_error - returns true if error may be path related
	128	* @error: status the request was completed with
	129	*
	130	* Description:
	131	* This classifies block error status into non-retryable errors and ones
	132	* that may be successful if retried on a failover path.
	133	*
	134	* Return:
	135	* %false - retrying failover path will not help
	136	* %true - may succeed if retried
	137	*/
	138	static inline bool blk_path_error(blk_status_t error)
	139	{
	140	switch (error) {
	141	case BLK_STS_NOTSUPP:
	142	case BLK_STS_NOSPC:
	143	case BLK_STS_TARGET:
	144	case BLK_STS_NEXUS:
	145	case BLK_STS_MEDIUM:
	146	case BLK_STS_PROTECTION:
	147	return false;
	148	}
	149
	150	/* Anything else could be a path failure, so should be retried */
	151	return true;
	152	}
	153
5238dcf4 OS	154	/*
	155	* From most significant bit:
	156	* 1 bit: reserved for other usage, see below
	157	* 12 bits: original size of bio
	158	* 51 bits: issue time of bio
	159	*/
	160	#define BIO_ISSUE_RES_BITS 1
	161	#define BIO_ISSUE_SIZE_BITS 12
	162	#define BIO_ISSUE_RES_SHIFT (64 - BIO_ISSUE_RES_BITS)
	163	#define BIO_ISSUE_SIZE_SHIFT (BIO_ISSUE_RES_SHIFT - BIO_ISSUE_SIZE_BITS)
	164	#define BIO_ISSUE_TIME_MASK ((1ULL << BIO_ISSUE_SIZE_SHIFT) - 1)
	165	#define BIO_ISSUE_SIZE_MASK \
	166	(((1ULL << BIO_ISSUE_SIZE_BITS) - 1) << BIO_ISSUE_SIZE_SHIFT)
	167	#define BIO_ISSUE_RES_MASK (~((1ULL << BIO_ISSUE_RES_SHIFT) - 1))
	168
	169	/* Reserved bit for blk-throtl */
	170	#define BIO_ISSUE_THROTL_SKIP_LATENCY (1ULL << 63)
	171
	172	struct bio_issue {
	173	u64 value;
	174	};
	175
	176	static inline u64 __bio_issue_time(u64 time)
	177	{
	178	return time & BIO_ISSUE_TIME_MASK;
	179	}
	180
	181	static inline u64 bio_issue_time(struct bio_issue *issue)
	182	{
	183	return __bio_issue_time(issue->value);
	184	}
	185
	186	static inline sector_t bio_issue_size(struct bio_issue *issue)
	187	{
	188	return ((issue->value & BIO_ISSUE_SIZE_MASK) >> BIO_ISSUE_SIZE_SHIFT);
	189	}
	190
	191	static inline void bio_issue_init(struct bio_issue *issue,
	192	sector_t size)
	193	{
	194	size &= (1ULL << BIO_ISSUE_SIZE_BITS) - 1;
	195	issue->value = ((issue->value & BIO_ISSUE_RES_MASK) \|
	196	(ktime_get_ns() & BIO_ISSUE_TIME_MASK) \|
	197	((u64)size << BIO_ISSUE_SIZE_SHIFT));
	198	}
	199
7cc01581 TH	200	/*
	201	* main unit of I/O for the block layer and lower layers (ie drivers and
	202	* stacking drivers)
	203	*/
	204	struct bio {
7cc01581	205	struct bio bi_next; / request queue link */
74d46992	206	struct gendisk *bi_disk;
1eff9d32 JA	207	unsigned int bi_opf; /* bottom bits req flags,
	208	* top bits REQ_OP. Use
	209	* accessors.
4e1b2d52	210	*/
dbde775c	211	unsigned short bi_flags; /* status, etc and bvec pool number */
43b62ce3	212	unsigned short bi_ioprio;
cb6934f8	213	unsigned short bi_write_hint;
111be883 SL	214	blk_status_t bi_status;
111be883 SL	215	u8 bi_partno;
993e4cde	216	atomic_t __bi_remaining;
7cc01581	217
111be883	218	struct bvec_iter bi_iter;
196d38bc	219
7cc01581 TH	220	bio_end_io_t *bi_end_io;
	221
	222	void *bi_private;
852c788f TH	223	#ifdef CONFIG_BLK_CGROUP
852c788f TH	224	/*
db6638d7 DZ	225	* Represents the association of the css and request_queue for the bio.
	226	* If a bio goes direct to device, it will not have a blkg as it will
	227	* not have a request_queue associated with it. The reference is put
	228	* on release of the bio.
852c788f	229	*/
08e18eab	230	struct blkcg_gq *bi_blkg;
5238dcf4	231	struct bio_issue bi_issue;
7caa4715 TH	232	#ifdef CONFIG_BLK_CGROUP_IOCOST
	233	u64 bi_iocost_cost;
	234	#endif
852c788f	235	#endif
a892c8d5 ST	236
	237	#ifdef CONFIG_BLK_INLINE_ENCRYPTION
	238	struct bio_crypt_ctx *bi_crypt_context;
	239	#endif
	240
180b2f95	241	union {
7cc01581	242	#if defined(CONFIG_BLK_DEV_INTEGRITY)
180b2f95	243	struct bio_integrity_payload bi_integrity; / data integrity */
7cc01581	244	#endif
180b2f95	245	};
7cc01581	246
4f024f37 KO	247	unsigned short bi_vcnt; /* how many bio_vec's */
4f024f37 KO	248
f44b48c7 KO	249	/*
	250	* Everything starting with bi_max_vecs will be preserved by bio_reset()
	251	*/
	252
4f024f37	253	unsigned short bi_max_vecs; /* max bvl_vecs we can hold */
f44b48c7	254
dac56212	255	atomic_t __bi_cnt; /* pin count */
f44b48c7 KO	256
	257	struct bio_vec bi_io_vec; / the actual vec list */
	258
395c72a7 KO	259	struct bio_set *bi_pool;
395c72a7 KO	260
7cc01581 TH	261	/*
	262	* We can inline a number of vecs at the end of the bio, to avoid
	263	* double allocations for a small number of bio_vecs. This member
	264	* MUST obviously be kept at the very end of the bio.
	265	*/
5a58ec8c	266	struct bio_vec bi_inline_vecs[];
7cc01581 TH	267	};
7cc01581 TH	268
f44b48c7 KO	269	#define BIO_RESET_BYTES offsetof(struct bio, bi_max_vecs)
f44b48c7 KO	270
7cc01581 TH	271	/*
	272	* bio flags
	273	*/
2b24e6f6 JT	274	enum {
2b24e6f6 JT	275	BIO_NO_PAGE_REF, /* don't put release vec pages */
2b24e6f6 JT	276	BIO_CLONED, /* doesn't own data */
2b24e6f6 JT	277	BIO_BOUNCED, /* bio is a bounce bio */
b8e24a93	278	BIO_WORKINGSET, /* contains userspace workingset pages */
2b24e6f6 JT	279	BIO_QUIET, /* Make BIO Quiet */
	280	BIO_CHAIN, /* chained bio, ->bi_remaining in effect */
	281	BIO_REFFED, /* bio has elevated ->bi_cnt */
	282	BIO_THROTTLED, /* This bio has already been subjected to
8d2bbd4c	283	* throttling rules. Don't do it again. */
2b24e6f6	284	BIO_TRACE_COMPLETION, /* bio_endio() should trace the final completion
fbbaf700	285	* of this bio. */
0376e9ef	286	BIO_CGROUP_ACCT, /* has been accounted to a cgroup */
2b24e6f6 JT	287	BIO_TRACKED, /* set if bio goes through the rq_qos path */
	288	BIO_FLAG_LAST
	289	};
cd4a4ae4	290
dbde775c	291	/* See BVEC_POOL_OFFSET below before adding new flags */
f44b48c7	292
7cc01581	293	/*
ed996a52 CH	294	* We support 6 different bvec pools, the last one is magic in that it
ed996a52 CH	295	* is backed by a mempool.
7cc01581	296	*/
ed996a52 CH	297	#define BVEC_POOL_NR 6
	298	#define BVEC_POOL_MAX (BVEC_POOL_NR - 1)
	299
	300	/*
dbde775c	301	* Top 3 bits of bio flags indicate the pool the bvecs came from. We add
ed996a52 CH	302	* 1 to the actual index so that 0 indicates that there are no bvecs to be
	303	* freed.
	304	*/
dbde775c	305	#define BVEC_POOL_BITS (3)
c0acf12a	306	#define BVEC_POOL_OFFSET (16 - BVEC_POOL_BITS)
ed996a52	307	#define BVEC_POOL_IDX(bio) ((bio)->bi_flags >> BVEC_POOL_OFFSET)
dbde775c N	308	#if (1<< BVEC_POOL_BITS) < (BVEC_POOL_NR+1)
	309	# error "BVEC_POOL_BITS is too small"
	310	#endif
	311
	312	/*
	313	* Flags starting here get preserved by bio_reset() - this includes
	314	* only BVEC_POOL_IDX()
	315	*/
	316	#define BIO_RESET_BITS BVEC_POOL_OFFSET
7cc01581	317
9a95e4ef BVA	318	typedef __u32 __bitwise blk_mq_req_flags_t;
9a95e4ef BVA	319
7cc01581	320	/*
ef295ecf CH	321	* Operations and flags common to the bio and request structures.
ef295ecf CH	322	* We use 8 bits for encoding the operation, and the remaining 24 for flags.
87374179 CH	323	*
	324	* The least significant bit of the operation number indicates the data
	325	* transfer direction:
	326	*
	327	* - if the least significant bit is set transfers are TO the device
	328	* - if the least significant bit is not set transfers are FROM the device
	329	*
	330	* If a operation does not transfer data the least significant bit has no
	331	* meaning.
7cc01581	332	*/
ef295ecf CH	333	#define REQ_OP_BITS 8
	334	#define REQ_OP_MASK ((1 << REQ_OP_BITS) - 1)
	335	#define REQ_FLAG_BITS 24
	336
	337	enum req_opf {
87374179 CH	338	/* read sectors from the device */
	339	REQ_OP_READ = 0,
	340	/* write sectors to the device */
	341	REQ_OP_WRITE = 1,
	342	/* flush the volatile write cache */
	343	REQ_OP_FLUSH = 2,
	344	/* discard sectors */
	345	REQ_OP_DISCARD = 3,
87374179 CH	346	/* securely erase sectors */
87374179 CH	347	REQ_OP_SECURE_ERASE = 5,
87374179 CH	348	/* write the same sector many times */
87374179 CH	349	REQ_OP_WRITE_SAME = 7,
a6f0788e	350	/* write the zero filled sector many times */
1d62ac13	351	REQ_OP_WRITE_ZEROES = 9,
6c1b1da5 AJ	352	/* Open a zone */
	353	REQ_OP_ZONE_OPEN = 10,
	354	/* Close a zone */
	355	REQ_OP_ZONE_CLOSE = 11,
	356	/* Transition a zone to full */
	357	REQ_OP_ZONE_FINISH = 12,
0512a75b KB	358	/* write data at the current zone write pointer */
0512a75b KB	359	REQ_OP_ZONE_APPEND = 13,
ecdef9f4 CL	360	/* reset a zone write pointer */
	361	REQ_OP_ZONE_RESET = 15,
	362	/* reset all the zone present on the device */
	363	REQ_OP_ZONE_RESET_ALL = 17,
ef295ecf	364
aebf526b CH	365	/* SCSI passthrough using struct scsi_request */
	366	REQ_OP_SCSI_IN = 32,
	367	REQ_OP_SCSI_OUT = 33,
	368	/* Driver private requests */
	369	REQ_OP_DRV_IN = 34,
	370	REQ_OP_DRV_OUT = 35,
	371
ef295ecf CH	372	REQ_OP_LAST,
	373	};
	374
	375	enum req_flag_bits {
	376	__REQ_FAILFAST_DEV = /* no driver retries of device errors */
	377	REQ_OP_BITS,
7cc01581 TH	378	__REQ_FAILFAST_TRANSPORT, /* no driver retries of transport errors */
7cc01581 TH	379	__REQ_FAILFAST_DRIVER, /* no driver retries of driver errors */
7cc01581 TH	380	__REQ_SYNC, /* request is sync (sync write or read) */
7cc01581 TH	381	__REQ_META, /* metadata io request */
65299a3b	382	__REQ_PRIO, /* boost priority in cfq */
bd1c1c21	383	__REQ_NOMERGE, /* don't touch this for merging */
a2b80967	384	__REQ_IDLE, /* anticipate more IO after this one */
180b2f95	385	__REQ_INTEGRITY, /* I/O includes block integrity payload */
8e4bf844	386	__REQ_FUA, /* forced unit access */
28a8f0d3	387	__REQ_PREFLUSH, /* request for cache flush */
188bd2b1	388	__REQ_RAHEAD, /* read ahead, can fail anytime */
1d796d6a	389	__REQ_BACKGROUND, /* background IO */
8977f563	390	__REQ_NOWAIT, /* Don't wait if request will block */
d3f77dfd TH	391	/*
	392	* When a shared kthread needs to issue a bio for a cgroup, doing
	393	* so synchronously can lead to priority inversions as the kthread
	394	* can be trapped waiting for that cgroup. CGROUP_PUNT flag makes
	395	* submit_bio() punt the actual issuing to a dedicated per-blkcg
	396	* work item to avoid such priority inversions.
	397	*/
	398	__REQ_CGROUP_PUNT,
d928be9f CH	399
	400	/* command specific flags for REQ_OP_WRITE_ZEROES: */
	401	__REQ_NOUNMAP, /* do not free blocks when zeroing */
	402
d1e36282 JA	403	__REQ_HIPRI,
d1e36282 JA	404
96222bcc CH	405	/* for driver use */
96222bcc CH	406	__REQ_DRV,
0d1e0c7c	407	__REQ_SWAP, /* swapping request. */
7cc01581 TH	408	__REQ_NR_BITS, /* stops here */
	409	};
	410
5953316d JA	411	#define REQ_FAILFAST_DEV (1ULL << __REQ_FAILFAST_DEV)
	412	#define REQ_FAILFAST_TRANSPORT (1ULL << __REQ_FAILFAST_TRANSPORT)
	413	#define REQ_FAILFAST_DRIVER (1ULL << __REQ_FAILFAST_DRIVER)
	414	#define REQ_SYNC (1ULL << __REQ_SYNC)
	415	#define REQ_META (1ULL << __REQ_META)
	416	#define REQ_PRIO (1ULL << __REQ_PRIO)
ef295ecf	417	#define REQ_NOMERGE (1ULL << __REQ_NOMERGE)
a2b80967	418	#define REQ_IDLE (1ULL << __REQ_IDLE)
180b2f95	419	#define REQ_INTEGRITY (1ULL << __REQ_INTEGRITY)
ef295ecf CH	420	#define REQ_FUA (1ULL << __REQ_FUA)
	421	#define REQ_PREFLUSH (1ULL << __REQ_PREFLUSH)
	422	#define REQ_RAHEAD (1ULL << __REQ_RAHEAD)
1d796d6a	423	#define REQ_BACKGROUND (1ULL << __REQ_BACKGROUND)
8977f563	424	#define REQ_NOWAIT (1ULL << __REQ_NOWAIT)
d3f77dfd TH	425	#define REQ_CGROUP_PUNT (1ULL << __REQ_CGROUP_PUNT)
d3f77dfd TH	426
d928be9f	427	#define REQ_NOUNMAP (1ULL << __REQ_NOUNMAP)
d1e36282	428	#define REQ_HIPRI (1ULL << __REQ_HIPRI)
d928be9f	429
96222bcc	430	#define REQ_DRV (1ULL << __REQ_DRV)
0d1e0c7c	431	#define REQ_SWAP (1ULL << __REQ_SWAP)
d928be9f	432
7cc01581 TH	433	#define REQ_FAILFAST_MASK \
7cc01581 TH	434	(REQ_FAILFAST_DEV \| REQ_FAILFAST_TRANSPORT \| REQ_FAILFAST_DRIVER)
7cc01581	435
e2a60da7	436	#define REQ_NOMERGE_FLAGS \
e8064021	437	(REQ_NOMERGE \| REQ_PREFLUSH \| REQ_FUA)
e2a60da7	438
dbae2c55 MC	439	enum stat_group {
	440	STAT_READ,
	441	STAT_WRITE,
bdca3c87	442	STAT_DISCARD,
b6866318	443	STAT_FLUSH,
dbae2c55 MC	444
	445	NR_STAT_GROUPS
	446	};
	447
ef295ecf CH	448	#define bio_op(bio) \
	449	((bio)->bi_opf & REQ_OP_MASK)
	450	#define req_op(req) \
	451	((req)->cmd_flags & REQ_OP_MASK)
7cc01581	452
ef295ecf	453	/* obsolete, don't use in new code */
93c5bdf7 CH	454	static inline void bio_set_op_attrs(struct bio *bio, unsigned op,
	455	unsigned op_flags)
	456	{
	457	bio->bi_opf = op \| op_flags;
	458	}
c11f0c0b	459
87374179 CH	460	static inline bool op_is_write(unsigned int op)
	461	{
	462	return (op & 1);
	463	}
	464
f73f44eb CH	465	/*
	466	* Check if the bio or request is one that needs special treatment in the
	467	* flush state machine.
	468	*/
	469	static inline bool op_is_flush(unsigned int op)
	470	{
	471	return op & (REQ_FUA \| REQ_PREFLUSH);
	472	}
	473
b685d3d6 CH	474	/*
	475	* Reads are always treated as synchronous, as are requests with the FUA or
	476	* PREFLUSH flag. Other operations may be marked as synchronous using the
	477	* REQ_SYNC flag.
	478	*/
ef295ecf CH	479	static inline bool op_is_sync(unsigned int op)
ef295ecf CH	480	{
b685d3d6 CH	481	return (op & REQ_OP_MASK) == REQ_OP_READ \|\|
b685d3d6 CH	482	(op & (REQ_SYNC \| REQ_FUA \| REQ_PREFLUSH));
ef295ecf	483	}
c11f0c0b	484
bdca3c87 MC	485	static inline bool op_is_discard(unsigned int op)
	486	{
	487	return (op & REQ_OP_MASK) == REQ_OP_DISCARD;
	488	}
	489
6c1b1da5 AJ	490	/*
	491	* Check if a bio or request operation is a zone management operation, with
	492	* the exception of REQ_OP_ZONE_RESET_ALL which is treated as a special case
	493	* due to its different handling in the block layer and device response in
	494	* case of command failure.
	495	*/
	496	static inline bool op_is_zone_mgmt(enum req_opf op)
	497	{
	498	switch (op & REQ_OP_MASK) {
	499	case REQ_OP_ZONE_RESET:
	500	case REQ_OP_ZONE_OPEN:
	501	case REQ_OP_ZONE_CLOSE:
	502	case REQ_OP_ZONE_FINISH:
	503	return true;
	504	default:
	505	return false;
	506	}
	507	}
	508
ddcf35d3 MC	509	static inline int op_stat_group(unsigned int op)
ddcf35d3 MC	510	{
bdca3c87 MC	511	if (op_is_discard(op))
bdca3c87 MC	512	return STAT_DISCARD;
ddcf35d3 MC	513	return op_is_write(op);
	514	}
	515
dece1635	516	typedef unsigned int blk_qc_t;
fd2d3326 JA	517	#define BLK_QC_T_NONE -1U
	518	#define BLK_QC_T_SHIFT 16
	519	#define BLK_QC_T_INTERNAL (1U << 31)
dece1635 JA	520
	521	static inline bool blk_qc_t_valid(blk_qc_t cookie)
	522	{
3aab9177	523	return cookie != BLK_QC_T_NONE;
dece1635 JA	524	}
dece1635 JA	525
dece1635 JA	526	static inline unsigned int blk_qc_t_to_queue_num(blk_qc_t cookie)
dece1635 JA	527	{
fd2d3326	528	return (cookie & ~BLK_QC_T_INTERNAL) >> BLK_QC_T_SHIFT;
dece1635 JA	529	}
	530
	531	static inline unsigned int blk_qc_t_to_tag(blk_qc_t cookie)
	532	{
e3a7a3bf	533	return cookie & ((1u << BLK_QC_T_SHIFT) - 1);
dece1635 JA	534	}
dece1635 JA	535
fd2d3326 JA	536	static inline bool blk_qc_t_is_internal(blk_qc_t cookie)
	537	{
	538	return (cookie & BLK_QC_T_INTERNAL) != 0;
	539	}
	540
cf43e6be	541	struct blk_rq_stat {
eca8b53a	542	u64 mean;
cf43e6be JA	543	u64 min;
cf43e6be JA	544	u64 max;
eca8b53a	545	u32 nr_samples;
cf43e6be	546	u64 batch;
cf43e6be JA	547	};
cf43e6be JA	548
7cc01581	549	#endif /* __LINUX_BLK_TYPES_H */