[linux-block.git] / block / blk.h

/* SPDX-License-Identifier: GPL-2.0 */
#ifndef BLK_INTERNAL_H
#define BLK_INTERNAL_H

#include <linux/blk-crypto.h>
#include <linux/memblock.h>	/* for max_pfn/max_low_pfn */
#include <xen/xen.h>
#include "blk-crypto-internal.h"

struct elevator_type;

/* Max future timer expiry for timeouts */
#define BLK_MAX_TIMEOUT		(5 * HZ)

extern struct dentry *blk_debugfs_root;

struct blk_flush_queue {
	unsigned int		flush_pending_idx:1;
	unsigned int		flush_running_idx:1;
	blk_status_t 		rq_status;
	unsigned long		flush_pending_since;
	struct list_head	flush_queue[2];
	struct list_head	flush_data_in_flight;
	struct request		*flush_rq;

	spinlock_t		mq_flush_lock;
};

extern struct kmem_cache *blk_requestq_cachep;
extern struct kmem_cache *blk_requestq_srcu_cachep;
extern struct kobj_type blk_queue_ktype;
extern struct ida blk_queue_ida;

bool is_flush_rq(struct request *req);

struct blk_flush_queue *blk_alloc_flush_queue(int node, int cmd_size,
					      gfp_t flags);
void blk_free_flush_queue(struct blk_flush_queue *q);

void blk_freeze_queue(struct request_queue *q);
void __blk_mq_unfreeze_queue(struct request_queue *q, bool force_atomic);
void blk_queue_start_drain(struct request_queue *q);
int __bio_queue_enter(struct request_queue *q, struct bio *bio);
void submit_bio_noacct_nocheck(struct bio *bio);

static inline bool blk_try_enter_queue(struct request_queue *q, bool pm)
{
	rcu_read_lock();
	if (!percpu_ref_tryget_live_rcu(&q->q_usage_counter))
		goto fail;

	/*
	 * The code that increments the pm_only counter must ensure that the
	 * counter is globally visible before the queue is unfrozen.
	 */
	if (blk_queue_pm_only(q) &&
	    (!pm || queue_rpm_status(q) == RPM_SUSPENDED))
		goto fail_put;

	rcu_read_unlock();
	return true;

fail_put:
	blk_queue_exit(q);
fail:
	rcu_read_unlock();
	return false;
}

static inline int bio_queue_enter(struct bio *bio)
{
	struct request_queue *q = bdev_get_queue(bio->bi_bdev);

	if (blk_try_enter_queue(q, false))
		return 0;
	return __bio_queue_enter(q, bio);
}

#define BIO_INLINE_VECS 4
struct bio_vec *bvec_alloc(mempool_t *pool, unsigned short *nr_vecs,
		gfp_t gfp_mask);
void bvec_free(mempool_t *pool, struct bio_vec *bv, unsigned short nr_vecs);

static inline bool biovec_phys_mergeable(struct request_queue *q,
		struct bio_vec *vec1, struct bio_vec *vec2)
{
	unsigned long mask = queue_segment_boundary(q);
	phys_addr_t addr1 = page_to_phys(vec1->bv_page) + vec1->bv_offset;
	phys_addr_t addr2 = page_to_phys(vec2->bv_page) + vec2->bv_offset;

	if (addr1 + vec1->bv_len != addr2)
		return false;
	if (xen_domain() && !xen_biovec_phys_mergeable(vec1, vec2->bv_page))
		return false;
	if ((addr1 | mask) != ((addr2 + vec2->bv_len - 1) | mask))
		return false;
	return true;
}

static inline bool __bvec_gap_to_prev(struct queue_limits *lim,
		struct bio_vec *bprv, unsigned int offset)
{
	return (offset & lim->virt_boundary_mask) ||
		((bprv->bv_offset + bprv->bv_len) & lim->virt_boundary_mask);
}

/*
 * Check if adding a bio_vec after bprv with offset would create a gap in
 * the SG list. Most drivers don't care about this, but some do.
 */
static inline bool bvec_gap_to_prev(struct queue_limits *lim,
		struct bio_vec *bprv, unsigned int offset)
{
	if (!lim->virt_boundary_mask)
		return false;
	return __bvec_gap_to_prev(lim, bprv, offset);
}

static inline bool rq_mergeable(struct request *rq)
{
	if (blk_rq_is_passthrough(rq))
		return false;

	if (req_op(rq) == REQ_OP_FLUSH)
		return false;

	if (req_op(rq) == REQ_OP_WRITE_ZEROES)
		return false;

	if (req_op(rq) == REQ_OP_ZONE_APPEND)
		return false;

	if (rq->cmd_flags & REQ_NOMERGE_FLAGS)
		return false;
	if (rq->rq_flags & RQF_NOMERGE_FLAGS)
		return false;

	return true;
}

/*
 * There are two different ways to handle DISCARD merges:
 *  1) If max_discard_segments > 1, the driver treats every bio as a range and
 *     send the bios to controller together. The ranges don't need to be
 *     contiguous.
 *  2) Otherwise, the request will be normal read/write requests.  The ranges
 *     need to be contiguous.
 */
static inline bool blk_discard_mergable(struct request *req)
{
	if (req_op(req) == REQ_OP_DISCARD &&
	    queue_max_discard_segments(req->q) > 1)
		return true;
	return false;
}

static inline unsigned int blk_queue_get_max_sectors(struct request_queue *q,
						     enum req_op op)
{
	if (unlikely(op == REQ_OP_DISCARD || op == REQ_OP_SECURE_ERASE))
		return min(q->limits.max_discard_sectors,
			   UINT_MAX >> SECTOR_SHIFT);

	if (unlikely(op == REQ_OP_WRITE_ZEROES))
		return q->limits.max_write_zeroes_sectors;

	return q->limits.max_sectors;
}

#ifdef CONFIG_BLK_DEV_INTEGRITY
void blk_flush_integrity(void);
bool __bio_integrity_endio(struct bio *);
void bio_integrity_free(struct bio *bio);
static inline bool bio_integrity_endio(struct bio *bio)
{
	if (bio_integrity(bio))
		return __bio_integrity_endio(bio);
	return true;
}

bool blk_integrity_merge_rq(struct request_queue *, struct request *,
		struct request *);
bool blk_integrity_merge_bio(struct request_queue *, struct request *,
		struct bio *);

static inline bool integrity_req_gap_back_merge(struct request *req,
		struct bio *next)
{
	struct bio_integrity_payload *bip = bio_integrity(req->bio);
	struct bio_integrity_payload *bip_next = bio_integrity(next);

	return bvec_gap_to_prev(&req->q->limits,
				&bip->bip_vec[bip->bip_vcnt - 1],
				bip_next->bip_vec[0].bv_offset);
}

static inline bool integrity_req_gap_front_merge(struct request *req,
		struct bio *bio)
{
	struct bio_integrity_payload *bip = bio_integrity(bio);
	struct bio_integrity_payload *bip_next = bio_integrity(req->bio);

	return bvec_gap_to_prev(&req->q->limits,
				&bip->bip_vec[bip->bip_vcnt - 1],
				bip_next->bip_vec[0].bv_offset);
}

int blk_integrity_add(struct gendisk *disk);
void blk_integrity_del(struct gendisk *);
#else /* CONFIG_BLK_DEV_INTEGRITY */
static inline bool blk_integrity_merge_rq(struct request_queue *rq,
		struct request *r1, struct request *r2)
{
	return true;
}
static inline bool blk_integrity_merge_bio(struct request_queue *rq,
		struct request *r, struct bio *b)
{
	return true;
}
static inline bool integrity_req_gap_back_merge(struct request *req,
		struct bio *next)
{
	return false;
}
static inline bool integrity_req_gap_front_merge(struct request *req,
		struct bio *bio)
{
	return false;
}

static inline void blk_flush_integrity(void)
{
}
static inline bool bio_integrity_endio(struct bio *bio)
{
	return true;
}
static inline void bio_integrity_free(struct bio *bio)
{
}
static inline int blk_integrity_add(struct gendisk *disk)
{
	return 0;
}
static inline void blk_integrity_del(struct gendisk *disk)
{
}
#endif /* CONFIG_BLK_DEV_INTEGRITY */

unsigned long blk_rq_timeout(unsigned long timeout);
void blk_add_timer(struct request *req);
const char *blk_status_to_str(blk_status_t status);

bool blk_attempt_plug_merge(struct request_queue *q, struct bio *bio,
		unsigned int nr_segs);
bool blk_bio_list_merge(struct request_queue *q, struct list_head *list,
			struct bio *bio, unsigned int nr_segs);

/*
 * Plug flush limits
 */
#define BLK_MAX_REQUEST_COUNT	32
#define BLK_PLUG_FLUSH_SIZE	(128 * 1024)

/*
 * Internal elevator interface
 */
#define ELV_ON_HASH(rq) ((rq)->rq_flags & RQF_HASHED)

void blk_insert_flush(struct request *rq);

int elevator_switch(struct request_queue *q, struct elevator_type *new_e);
void elevator_exit(struct request_queue *q);
int elv_register_queue(struct request_queue *q, bool uevent);
void elv_unregister_queue(struct request_queue *q);

ssize_t part_size_show(struct device *dev, struct device_attribute *attr,
		char *buf);
ssize_t part_stat_show(struct device *dev, struct device_attribute *attr,
		char *buf);
ssize_t part_inflight_show(struct device *dev, struct device_attribute *attr,
		char *buf);
ssize_t part_fail_show(struct device *dev, struct device_attribute *attr,
		char *buf);
ssize_t part_fail_store(struct device *dev, struct device_attribute *attr,
		const char *buf, size_t count);
ssize_t part_timeout_show(struct device *, struct device_attribute *, char *);
ssize_t part_timeout_store(struct device *, struct device_attribute *,
				const char *, size_t);

static inline bool bio_may_exceed_limits(struct bio *bio,
		struct queue_limits *lim)
{
	switch (bio_op(bio)) {
	case REQ_OP_DISCARD:
	case REQ_OP_SECURE_ERASE:
	case REQ_OP_WRITE_ZEROES:
		return true; /* non-trivial splitting decisions */
	default:
		break;
	}

	/*
	 * All drivers must accept single-segments bios that are <= PAGE_SIZE.
	 * This is a quick and dirty check that relies on the fact that
	 * bi_io_vec[0] is always valid if a bio has data.  The check might
	 * lead to occasional false negatives when bios are cloned, but compared
	 * to the performance impact of cloned bios themselves the loop below
	 * doesn't matter anyway.
	 */
	return lim->chunk_sectors || bio->bi_vcnt != 1 ||
		bio->bi_io_vec->bv_len + bio->bi_io_vec->bv_offset > PAGE_SIZE;
}

struct bio *__bio_split_to_limits(struct bio *bio, struct queue_limits *lim,
		       unsigned int *nr_segs);
int ll_back_merge_fn(struct request *req, struct bio *bio,
		unsigned int nr_segs);
bool blk_attempt_req_merge(struct request_queue *q, struct request *rq,
				struct request *next);
unsigned int blk_recalc_rq_segments(struct request *rq);
void blk_rq_set_mixed_merge(struct request *rq);
bool blk_rq_merge_ok(struct request *rq, struct bio *bio);
enum elv_merge blk_try_merge(struct request *rq, struct bio *bio);

int blk_dev_init(void);

/*
 * Contribute to IO statistics IFF:
 *
 *	a) it's attached to a gendisk, and
 *	b) the queue had IO stats enabled when this request was started
 */
static inline bool blk_do_io_stat(struct request *rq)
{
	return (rq->rq_flags & RQF_IO_STAT) && !blk_rq_is_passthrough(rq);
}

void update_io_ticks(struct block_device *part, unsigned long now, bool end);

static inline void req_set_nomerge(struct request_queue *q, struct request *req)
{
	req->cmd_flags |= REQ_NOMERGE;
	if (req == q->last_merge)
		q->last_merge = NULL;
}

/*
 * Internal io_context interface
 */
struct io_cq *ioc_find_get_icq(struct request_queue *q);
struct io_cq *ioc_lookup_icq(struct request_queue *q);
#ifdef CONFIG_BLK_ICQ
void ioc_clear_queue(struct request_queue *q);
#else
static inline void ioc_clear_queue(struct request_queue *q)
{
}
#endif /* CONFIG_BLK_ICQ */

#ifdef CONFIG_BLK_DEV_THROTTLING_LOW
extern ssize_t blk_throtl_sample_time_show(struct request_queue *q, char *page);
extern ssize_t blk_throtl_sample_time_store(struct request_queue *q,
	const char *page, size_t count);
extern void blk_throtl_bio_endio(struct bio *bio);
extern void blk_throtl_stat_add(struct request *rq, u64 time);
#else
static inline void blk_throtl_bio_endio(struct bio *bio) { }
static inline void blk_throtl_stat_add(struct request *rq, u64 time) { }
#endif

struct bio *__blk_queue_bounce(struct bio *bio, struct request_queue *q);

static inline bool blk_queue_may_bounce(struct request_queue *q)
{
	return IS_ENABLED(CONFIG_BOUNCE) &&
		q->limits.bounce == BLK_BOUNCE_HIGH &&
		max_low_pfn >= max_pfn;
}

static inline struct bio *blk_queue_bounce(struct bio *bio,
		struct request_queue *q)
{
	if (unlikely(blk_queue_may_bounce(q) && bio_has_data(bio)))
		return __blk_queue_bounce(bio, q);
	return bio;
}

#ifdef CONFIG_BLK_CGROUP_IOLATENCY
int blk_iolatency_init(struct gendisk *disk);
#else
static inline int blk_iolatency_init(struct gendisk *disk) { return 0; };
#endif

#ifdef CONFIG_BLK_DEV_ZONED
void disk_free_zone_bitmaps(struct gendisk *disk);
void disk_clear_zone_settings(struct gendisk *disk);
#else
static inline void disk_free_zone_bitmaps(struct gendisk *disk) {}
static inline void disk_clear_zone_settings(struct gendisk *disk) {}
#endif

int blk_alloc_ext_minor(void);
void blk_free_ext_minor(unsigned int minor);
#define ADDPART_FLAG_NONE	0
#define ADDPART_FLAG_RAID	1
#define ADDPART_FLAG_WHOLEDISK	2
int bdev_add_partition(struct gendisk *disk, int partno, sector_t start,
		sector_t length);
int bdev_del_partition(struct gendisk *disk, int partno);
int bdev_resize_partition(struct gendisk *disk, int partno, sector_t start,
		sector_t length);
void blk_drop_partitions(struct gendisk *disk);

struct gendisk *__alloc_disk_node(struct request_queue *q, int node_id,
		struct lock_class_key *lkclass);

int bio_add_hw_page(struct request_queue *q, struct bio *bio,
		struct page *page, unsigned int len, unsigned int offset,
		unsigned int max_sectors, bool *same_page);

static inline struct kmem_cache *blk_get_queue_kmem_cache(bool srcu)
{
	if (srcu)
		return blk_requestq_srcu_cachep;
	return blk_requestq_cachep;
}
struct request_queue *blk_alloc_queue(int node_id, bool alloc_srcu);

int disk_scan_partitions(struct gendisk *disk, fmode_t mode);

int disk_alloc_events(struct gendisk *disk);
void disk_add_events(struct gendisk *disk);
void disk_del_events(struct gendisk *disk);
void disk_release_events(struct gendisk *disk);
void disk_block_events(struct gendisk *disk);
void disk_unblock_events(struct gendisk *disk);
void disk_flush_events(struct gendisk *disk, unsigned int mask);
extern struct device_attribute dev_attr_events;
extern struct device_attribute dev_attr_events_async;
extern struct device_attribute dev_attr_events_poll_msecs;

extern struct attribute_group blk_trace_attr_group;

long blkdev_ioctl(struct file *file, unsigned cmd, unsigned long arg);
long compat_blkdev_ioctl(struct file *file, unsigned cmd, unsigned long arg);

extern const struct address_space_operations def_blk_aops;

int disk_register_independent_access_ranges(struct gendisk *disk);
void disk_unregister_independent_access_ranges(struct gendisk *disk);

#ifdef CONFIG_FAIL_MAKE_REQUEST
bool should_fail_request(struct block_device *part, unsigned int bytes);
#else /* CONFIG_FAIL_MAKE_REQUEST */
static inline bool should_fail_request(struct block_device *part,
					unsigned int bytes)
{
	return false;
}
#endif /* CONFIG_FAIL_MAKE_REQUEST */

/*
 * Optimized request reference counting. Ideally we'd make timeouts be more
 * clever, as that's the only reason we need references at all... But until
 * this happens, this is faster than using refcount_t. Also see:
 *
 * abc54d634334 ("io_uring: switch to atomic_t for io_kiocb reference count")
 */
#define req_ref_zero_or_close_to_overflow(req)	\
	((unsigned int) atomic_read(&(req->ref)) + 127u <= 127u)

static inline bool req_ref_inc_not_zero(struct request *req)
{
	return atomic_inc_not_zero(&req->ref);
}

static inline bool req_ref_put_and_test(struct request *req)
{
	WARN_ON_ONCE(req_ref_zero_or_close_to_overflow(req));
	return atomic_dec_and_test(&req->ref);
}

static inline void req_ref_set(struct request *req, int value)
{
	atomic_set(&req->ref, value);
}

static inline int req_ref_read(struct request *req)
{
	return atomic_read(&req->ref);
}

#endif /* BLK_INTERNAL_H */
Commit	Line	Data
b2441318	1	/* SPDX-License-Identifier: GPL-2.0 */
8324aa91 JA	2	#ifndef BLK_INTERNAL_H
	3	#define BLK_INTERNAL_H
	4
a892c8d5	5	#include <linux/blk-crypto.h>
9bb33f24	6	#include <linux/memblock.h> /* for max_pfn/max_low_pfn */
c39ae60d	7	#include <xen/xen.h>
a892c8d5	8	#include "blk-crypto-internal.h"
a73f730d	9
2e9bc346 CH	10	struct elevator_type;
2e9bc346 CH	11
0d2602ca JA	12	/* Max future timer expiry for timeouts */
	13	#define BLK_MAX_TIMEOUT (5 * HZ)
	14
18fbda91	15	extern struct dentry *blk_debugfs_root;
18fbda91	16
7c94e1c1	17	struct blk_flush_queue {
7c94e1c1 ML	18	unsigned int flush_pending_idx:1;
7c94e1c1 ML	19	unsigned int flush_running_idx:1;
8d699663	20	blk_status_t rq_status;
7c94e1c1 ML	21	unsigned long flush_pending_since;
	22	struct list_head flush_queue[2];
	23	struct list_head flush_data_in_flight;
	24	struct request *flush_rq;
0048b483	25
7c94e1c1 ML	26	spinlock_t mq_flush_lock;
	27	};
	28
8324aa91	29	extern struct kmem_cache *blk_requestq_cachep;
704b914f	30	extern struct kmem_cache *blk_requestq_srcu_cachep;
8324aa91	31	extern struct kobj_type blk_queue_ktype;
a73f730d	32	extern struct ida blk_queue_ida;
8324aa91	33
a9ed27a7	34	bool is_flush_rq(struct request *req);
8d699663	35
754a1572 GJ	36	struct blk_flush_queue *blk_alloc_flush_queue(int node, int cmd_size,
754a1572 GJ	37	gfp_t flags);
f70ced09	38	void blk_free_flush_queue(struct blk_flush_queue *q);
f3552655	39
3ef28e83	40	void blk_freeze_queue(struct request_queue *q);
aec89dc5	41	void __blk_mq_unfreeze_queue(struct request_queue *q, bool force_atomic);
8e141f9e	42	void blk_queue_start_drain(struct request_queue *q);
c98cb5bb	43	int __bio_queue_enter(struct request_queue q, struct bio bio);
3f98c753	44	void submit_bio_noacct_nocheck(struct bio *bio);
c98cb5bb JA	45
	46	static inline bool blk_try_enter_queue(struct request_queue *q, bool pm)
	47	{
	48	rcu_read_lock();
	49	if (!percpu_ref_tryget_live_rcu(&q->q_usage_counter))
	50	goto fail;
	51
	52	/*
	53	* The code that increments the pm_only counter must ensure that the
	54	* counter is globally visible before the queue is unfrozen.
	55	*/
	56	if (blk_queue_pm_only(q) &&
	57	(!pm \|\| queue_rpm_status(q) == RPM_SUSPENDED))
	58	goto fail_put;
	59
	60	rcu_read_unlock();
	61	return true;
	62
	63	fail_put:
	64	blk_queue_exit(q);
	65	fail:
	66	rcu_read_unlock();
	67	return false;
	68	}
	69
	70	static inline int bio_queue_enter(struct bio *bio)
	71	{
	72	struct request_queue *q = bdev_get_queue(bio->bi_bdev);
	73
	74	if (blk_try_enter_queue(q, false))
	75	return 0;
	76	return __bio_queue_enter(q, bio);
	77	}
3ef28e83	78
dc0b8a57	79	#define BIO_INLINE_VECS 4
7a800a20 CH	80	struct bio_vec bvec_alloc(mempool_t pool, unsigned short *nr_vecs,
	81	gfp_t gfp_mask);
	82	void bvec_free(mempool_t pool, struct bio_vec bv, unsigned short nr_vecs);
eec716a1	83
3dccdae5 CH	84	static inline bool biovec_phys_mergeable(struct request_queue *q,
3dccdae5 CH	85	struct bio_vec vec1, struct bio_vec vec2)
6a9f5f24	86	{
3dccdae5	87	unsigned long mask = queue_segment_boundary(q);
6e768461 CH	88	phys_addr_t addr1 = page_to_phys(vec1->bv_page) + vec1->bv_offset;
6e768461 CH	89	phys_addr_t addr2 = page_to_phys(vec2->bv_page) + vec2->bv_offset;
3dccdae5 CH	90
3dccdae5 CH	91	if (addr1 + vec1->bv_len != addr2)
6a9f5f24	92	return false;
0383ad43	93	if (xen_domain() && !xen_biovec_phys_mergeable(vec1, vec2->bv_page))
6a9f5f24	94	return false;
3dccdae5 CH	95	if ((addr1 \| mask) != ((addr2 + vec2->bv_len - 1) \| mask))
3dccdae5 CH	96	return false;
6a9f5f24 CH	97	return true;
	98	}
	99
c55ddd90	100	static inline bool __bvec_gap_to_prev(struct queue_limits *lim,
27ca1d4e CH	101	struct bio_vec *bprv, unsigned int offset)
27ca1d4e CH	102	{
c55ddd90 CH	103	return (offset & lim->virt_boundary_mask) \|\|
c55ddd90 CH	104	((bprv->bv_offset + bprv->bv_len) & lim->virt_boundary_mask);
27ca1d4e CH	105	}
	106
	107	/*
	108	* Check if adding a bio_vec after bprv with offset would create a gap in
	109	* the SG list. Most drivers don't care about this, but some do.
	110	*/
c55ddd90	111	static inline bool bvec_gap_to_prev(struct queue_limits *lim,
27ca1d4e CH	112	struct bio_vec *bprv, unsigned int offset)
27ca1d4e CH	113	{
c55ddd90	114	if (!lim->virt_boundary_mask)
27ca1d4e	115	return false;
c55ddd90	116	return __bvec_gap_to_prev(lim, bprv, offset);
27ca1d4e CH	117	}
27ca1d4e CH	118
badf7f64 CH	119	static inline bool rq_mergeable(struct request *rq)
	120	{
	121	if (blk_rq_is_passthrough(rq))
	122	return false;
	123
	124	if (req_op(rq) == REQ_OP_FLUSH)
	125	return false;
	126
	127	if (req_op(rq) == REQ_OP_WRITE_ZEROES)
	128	return false;
	129
	130	if (req_op(rq) == REQ_OP_ZONE_APPEND)
	131	return false;
	132
	133	if (rq->cmd_flags & REQ_NOMERGE_FLAGS)
	134	return false;
	135	if (rq->rq_flags & RQF_NOMERGE_FLAGS)
	136	return false;
	137
	138	return true;
	139	}
	140
	141	/*
	142	* There are two different ways to handle DISCARD merges:
	143	* 1) If max_discard_segments > 1, the driver treats every bio as a range and
	144	* send the bios to controller together. The ranges don't need to be
	145	* contiguous.
	146	* 2) Otherwise, the request will be normal read/write requests. The ranges
	147	* need to be contiguous.
	148	*/
	149	static inline bool blk_discard_mergable(struct request *req)
	150	{
	151	if (req_op(req) == REQ_OP_DISCARD &&
	152	queue_max_discard_segments(req->q) > 1)
	153	return true;
	154	return false;
	155	}
	156
2a9336c4	157	static inline unsigned int blk_queue_get_max_sectors(struct request_queue *q,
77e7ffd7	158	enum req_op op)
2a9336c4 CH	159	{
	160	if (unlikely(op == REQ_OP_DISCARD \|\| op == REQ_OP_SECURE_ERASE))
	161	return min(q->limits.max_discard_sectors,
	162	UINT_MAX >> SECTOR_SHIFT);
	163
	164	if (unlikely(op == REQ_OP_WRITE_ZEROES))
	165	return q->limits.max_write_zeroes_sectors;
	166
	167	return q->limits.max_sectors;
	168	}
	169
5a48fc14 DW	170	#ifdef CONFIG_BLK_DEV_INTEGRITY
5a48fc14 DW	171	void blk_flush_integrity(void);
7c20f116	172	bool __bio_integrity_endio(struct bio *);
ece841ab	173	void bio_integrity_free(struct bio *bio);
7c20f116 CH	174	static inline bool bio_integrity_endio(struct bio *bio)
	175	{
	176	if (bio_integrity(bio))
	177	return __bio_integrity_endio(bio);
	178	return true;
	179	}
43b729bf	180
92cf2fd1 CH	181	bool blk_integrity_merge_rq(struct request_queue , struct request ,
92cf2fd1 CH	182	struct request *);
d59da419 CH	183	bool blk_integrity_merge_bio(struct request_queue , struct request ,
d59da419 CH	184	struct bio *);
92cf2fd1	185
43b729bf CH	186	static inline bool integrity_req_gap_back_merge(struct request *req,
	187	struct bio *next)
	188	{
	189	struct bio_integrity_payload *bip = bio_integrity(req->bio);
	190	struct bio_integrity_payload *bip_next = bio_integrity(next);
	191
c55ddd90 CH	192	return bvec_gap_to_prev(&req->q->limits,
c55ddd90 CH	193	&bip->bip_vec[bip->bip_vcnt - 1],
43b729bf CH	194	bip_next->bip_vec[0].bv_offset);
	195	}
	196
	197	static inline bool integrity_req_gap_front_merge(struct request *req,
	198	struct bio *bio)
	199	{
	200	struct bio_integrity_payload *bip = bio_integrity(bio);
	201	struct bio_integrity_payload *bip_next = bio_integrity(req->bio);
	202
c55ddd90 CH	203	return bvec_gap_to_prev(&req->q->limits,
c55ddd90 CH	204	&bip->bip_vec[bip->bip_vcnt - 1],
43b729bf CH	205	bip_next->bip_vec[0].bv_offset);
43b729bf CH	206	}
581e2600	207
614310c9	208	int blk_integrity_add(struct gendisk *disk);
581e2600	209	void blk_integrity_del(struct gendisk *);
43b729bf	210	#else /* CONFIG_BLK_DEV_INTEGRITY */
92cf2fd1 CH	211	static inline bool blk_integrity_merge_rq(struct request_queue *rq,
	212	struct request r1, struct request r2)
	213	{
	214	return true;
	215	}
d59da419 CH	216	static inline bool blk_integrity_merge_bio(struct request_queue *rq,
	217	struct request r, struct bio b)
	218	{
	219	return true;
	220	}
43b729bf CH	221	static inline bool integrity_req_gap_back_merge(struct request *req,
	222	struct bio *next)
	223	{
	224	return false;
	225	}
	226	static inline bool integrity_req_gap_front_merge(struct request *req,
	227	struct bio *bio)
	228	{
	229	return false;
	230	}
	231
5a48fc14 DW	232	static inline void blk_flush_integrity(void)
	233	{
	234	}
7c20f116 CH	235	static inline bool bio_integrity_endio(struct bio *bio)
	236	{
	237	return true;
	238	}
ece841ab JT	239	static inline void bio_integrity_free(struct bio *bio)
	240	{
	241	}
614310c9	242	static inline int blk_integrity_add(struct gendisk *disk)
581e2600	243	{
614310c9	244	return 0;
581e2600 CH	245	}
	246	static inline void blk_integrity_del(struct gendisk *disk)
	247	{
	248	}
43b729bf	249	#endif /* CONFIG_BLK_DEV_INTEGRITY */
8324aa91	250
0d2602ca	251	unsigned long blk_rq_timeout(unsigned long timeout);
87ee7b11	252	void blk_add_timer(struct request *req);
0d7a29a2	253	const char *blk_status_to_str(blk_status_t status);
320ae51f	254
320ae51f	255	bool blk_attempt_plug_merge(struct request_queue q, struct bio bio,
0c5bcc92	256	unsigned int nr_segs);
bdc6a287 BW	257	bool blk_bio_list_merge(struct request_queue q, struct list_head list,
bdc6a287 BW	258	struct bio *bio, unsigned int nr_segs);
320ae51f	259
ba0ffdd8 JA	260	/*
	261	* Plug flush limits
	262	*/
	263	#define BLK_MAX_REQUEST_COUNT 32
	264	#define BLK_PLUG_FLUSH_SIZE (128 * 1024)
	265
158dbda0 TH	266	/*
	267	* Internal elevator interface
	268	*/
e8064021	269	#define ELV_ON_HASH(rq) ((rq)->rq_flags & RQF_HASHED)
158dbda0	270
2b504bd4	271	void blk_insert_flush(struct request *rq);
dd831006	272
8237c01f	273	int elevator_switch(struct request_queue q, struct elevator_type new_e);
0c6cb3a2	274	void elevator_exit(struct request_queue *q);
cecf5d87	275	int elv_register_queue(struct request_queue *q, bool uevent);
83d016ac BVA	276	void elv_unregister_queue(struct request_queue *q);
83d016ac BVA	277
3ad5cee5 CH	278	ssize_t part_size_show(struct device dev, struct device_attribute attr,
	279	char *buf);
	280	ssize_t part_stat_show(struct device dev, struct device_attribute attr,
	281	char *buf);
	282	ssize_t part_inflight_show(struct device dev, struct device_attribute attr,
	283	char *buf);
	284	ssize_t part_fail_show(struct device dev, struct device_attribute attr,
	285	char *buf);
	286	ssize_t part_fail_store(struct device dev, struct device_attribute attr,
	287	const char *buf, size_t count);
581d4e28 JA	288	ssize_t part_timeout_show(struct device , struct device_attribute , char *);
	289	ssize_t part_timeout_store(struct device , struct device_attribute ,
	290	const char *, size_t);
581d4e28	291
c55ddd90 CH	292	static inline bool bio_may_exceed_limits(struct bio *bio,
c55ddd90 CH	293	struct queue_limits *lim)
abd45c15 JA	294	{
	295	switch (bio_op(bio)) {
	296	case REQ_OP_DISCARD:
	297	case REQ_OP_SECURE_ERASE:
	298	case REQ_OP_WRITE_ZEROES:
abd45c15 JA	299	return true; /* non-trivial splitting decisions */
	300	default:
	301	break;
	302	}
	303
	304	/*
	305	* All drivers must accept single-segments bios that are <= PAGE_SIZE.
	306	* This is a quick and dirty check that relies on the fact that
	307	* bi_io_vec[0] is always valid if a bio has data. The check might
	308	* lead to occasional false negatives when bios are cloned, but compared
	309	* to the performance impact of cloned bios themselves the loop below
	310	* doesn't matter anyway.
	311	*/
c55ddd90	312	return lim->chunk_sectors \|\| bio->bi_vcnt != 1 \|\|
abd45c15 JA	313	bio->bi_io_vec->bv_len + bio->bi_io_vec->bv_offset > PAGE_SIZE;
	314	}
	315
c55ddd90	316	struct bio __bio_split_to_limits(struct bio bio, struct queue_limits *lim,
5a97806f	317	unsigned int *nr_segs);
14ccb66b CH	318	int ll_back_merge_fn(struct request req, struct bio bio,
14ccb66b CH	319	unsigned int nr_segs);
fd2ef39c	320	bool blk_attempt_req_merge(struct request_queue q, struct request rq,
5e84ea3a	321	struct request *next);
e9cd19c0	322	unsigned int blk_recalc_rq_segments(struct request *rq);
80a761fd	323	void blk_rq_set_mixed_merge(struct request *rq);
050c8ea8	324	bool blk_rq_merge_ok(struct request rq, struct bio bio);
34fe7c05	325	enum elv_merge blk_try_merge(struct request rq, struct bio bio);
d6d48196	326
ff88972c AB	327	int blk_dev_init(void);
ff88972c AB	328
c2553b58 JA	329	/*
	330	* Contribute to IO statistics IFF:
	331	*
	332	* a) it's attached to a gendisk, and
48d9b0d4	333	* b) the queue had IO stats enabled when this request was started
c2553b58	334	*/
599d067d	335	static inline bool blk_do_io_stat(struct request *rq)
fb8ec18c	336	{
41fa7222	337	return (rq->rq_flags & RQF_IO_STAT) && !blk_rq_is_passthrough(rq);
be6bfe36 PB	338	}
be6bfe36 PB	339
450b7879	340	void update_io_ticks(struct block_device *part, unsigned long now, bool end);
fb8ec18c	341
6cf7677f CH	342	static inline void req_set_nomerge(struct request_queue q, struct request req)
	343	{
	344	req->cmd_flags \|= REQ_NOMERGE;
	345	if (req == q->last_merge)
	346	q->last_merge = NULL;
	347	}
	348
f2dbd76a TH	349	/*
	350	* Internal io_context interface
	351	*/
87dd1d63	352	struct io_cq ioc_find_get_icq(struct request_queue q);
eca5892a	353	struct io_cq ioc_lookup_icq(struct request_queue q);
5ef16305	354	#ifdef CONFIG_BLK_ICQ
7e5a8794	355	void ioc_clear_queue(struct request_queue *q);
5ef16305 CH	356	#else
	357	static inline void ioc_clear_queue(struct request_queue *q)
	358	{
	359	}
	360	#endif /* CONFIG_BLK_ICQ */
f2dbd76a	361
297e3d85 SL	362	#ifdef CONFIG_BLK_DEV_THROTTLING_LOW
	363	extern ssize_t blk_throtl_sample_time_show(struct request_queue q, char page);
	364	extern ssize_t blk_throtl_sample_time_store(struct request_queue *q,
	365	const char *page, size_t count);
9e234eea	366	extern void blk_throtl_bio_endio(struct bio *bio);
b9147dd1	367	extern void blk_throtl_stat_add(struct request *rq, u64 time);
9e234eea SL	368	#else
9e234eea SL	369	static inline void blk_throtl_bio_endio(struct bio *bio) { }
b9147dd1	370	static inline void blk_throtl_stat_add(struct request *rq, u64 time) { }
297e3d85	371	#endif
bc9fcbf9	372
51d798cd	373	struct bio __blk_queue_bounce(struct bio bio, struct request_queue *q);
9bb33f24 CH	374
	375	static inline bool blk_queue_may_bounce(struct request_queue *q)
	376	{
	377	return IS_ENABLED(CONFIG_BOUNCE) &&
	378	q->limits.bounce == BLK_BOUNCE_HIGH &&
	379	max_low_pfn >= max_pfn;
	380	}
	381
51d798cd CH	382	static inline struct bio blk_queue_bounce(struct bio bio,
51d798cd CH	383	struct request_queue *q)
3bce016a	384	{
51d798cd CH	385	if (unlikely(blk_queue_may_bounce(q) && bio_has_data(bio)))
	386	return __blk_queue_bounce(bio, q);
	387	return bio;
3bce016a	388	}
3bce016a	389
d7067512	390	#ifdef CONFIG_BLK_CGROUP_IOLATENCY
16fac1b5	391	int blk_iolatency_init(struct gendisk *disk);
d7067512	392	#else
16fac1b5	393	static inline int blk_iolatency_init(struct gendisk *disk) { return 0; };
d7067512 JB	394	#endif
d7067512 JB	395
bf505456	396	#ifdef CONFIG_BLK_DEV_ZONED
5d400665	397	void disk_free_zone_bitmaps(struct gendisk *disk);
b3c72f81	398	void disk_clear_zone_settings(struct gendisk *disk);
bf505456	399	#else
5d400665	400	static inline void disk_free_zone_bitmaps(struct gendisk *disk) {}
b3c72f81	401	static inline void disk_clear_zone_settings(struct gendisk *disk) {}
bf505456 DLM	402	#endif
bf505456 DLM	403
7c3f828b CH	404	int blk_alloc_ext_minor(void);
7c3f828b CH	405	void blk_free_ext_minor(unsigned int minor);
581e2600 CH	406	#define ADDPART_FLAG_NONE 0
	407	#define ADDPART_FLAG_RAID 1
	408	#define ADDPART_FLAG_WHOLEDISK 2
7f6be376 CH	409	int bdev_add_partition(struct gendisk *disk, int partno, sector_t start,
7f6be376 CH	410	sector_t length);
926fbb16	411	int bdev_del_partition(struct gendisk *disk, int partno);
3d2e7989 CH	412	int bdev_resize_partition(struct gendisk *disk, int partno, sector_t start,
3d2e7989 CH	413	sector_t length);
e7243285	414	void blk_drop_partitions(struct gendisk *disk);
581e2600	415
6f8191fd CH	416	struct gendisk __alloc_disk_node(struct request_queue q, int node_id,
	417	struct lock_class_key *lkclass);
	418
e4581105	419	int bio_add_hw_page(struct request_queue q, struct bio bio,
130879f1	420	struct page *page, unsigned int len, unsigned int offset,
e4581105	421	unsigned int max_sectors, bool *same_page);
130879f1	422
704b914f ML	423	static inline struct kmem_cache *blk_get_queue_kmem_cache(bool srcu)
	424	{
	425	if (srcu)
	426	return blk_requestq_srcu_cachep;
	427	return blk_requestq_cachep;
	428	}
	429	struct request_queue *blk_alloc_queue(int node_id, bool alloc_srcu);
	430
e16e506c	431	int disk_scan_partitions(struct gendisk *disk, fmode_t mode);
da7ba729	432
92e7755e	433	int disk_alloc_events(struct gendisk *disk);
d5870edf CH	434	void disk_add_events(struct gendisk *disk);
	435	void disk_del_events(struct gendisk *disk);
	436	void disk_release_events(struct gendisk *disk);
926597ff CH	437	void disk_block_events(struct gendisk *disk);
	438	void disk_unblock_events(struct gendisk *disk);
	439	void disk_flush_events(struct gendisk *disk, unsigned int mask);
2bc8cda5 CH	440	extern struct device_attribute dev_attr_events;
	441	extern struct device_attribute dev_attr_events_async;
	442	extern struct device_attribute dev_attr_events_poll_msecs;
d5870edf	443
cc5c516d CH	444	extern struct attribute_group blk_trace_attr_group;
cc5c516d CH	445
8a709512	446	long blkdev_ioctl(struct file *file, unsigned cmd, unsigned long arg);
84b8514b CH	447	long compat_blkdev_ioctl(struct file *file, unsigned cmd, unsigned long arg);
84b8514b CH	448
cd82cca7 CH	449	extern const struct address_space_operations def_blk_aops;
cd82cca7 CH	450
22d0c408	451	int disk_register_independent_access_ranges(struct gendisk *disk);
a2247f19 DLM	452	void disk_unregister_independent_access_ranges(struct gendisk *disk);
a2247f19 DLM	453
06c8c691 CH	454	#ifdef CONFIG_FAIL_MAKE_REQUEST
	455	bool should_fail_request(struct block_device *part, unsigned int bytes);
	456	#else /* CONFIG_FAIL_MAKE_REQUEST */
	457	static inline bool should_fail_request(struct block_device *part,
	458	unsigned int bytes)
	459	{
	460	return false;
	461	}
	462	#endif /* CONFIG_FAIL_MAKE_REQUEST */
	463
0a467d0f JA	464	/*
	465	* Optimized request reference counting. Ideally we'd make timeouts be more
	466	* clever, as that's the only reason we need references at all... But until
	467	* this happens, this is faster than using refcount_t. Also see:
	468	*
	469	* abc54d634334 ("io_uring: switch to atomic_t for io_kiocb reference count")
	470	*/
	471	#define req_ref_zero_or_close_to_overflow(req) \
	472	((unsigned int) atomic_read(&(req->ref)) + 127u <= 127u)
	473
	474	static inline bool req_ref_inc_not_zero(struct request *req)
	475	{
	476	return atomic_inc_not_zero(&req->ref);
	477	}
	478
	479	static inline bool req_ref_put_and_test(struct request *req)
	480	{
	481	WARN_ON_ONCE(req_ref_zero_or_close_to_overflow(req));
	482	return atomic_dec_and_test(&req->ref);
	483	}
	484
	485	static inline void req_ref_set(struct request *req, int value)
	486	{
	487	atomic_set(&req->ref, value);
	488	}
	489
	490	static inline int req_ref_read(struct request *req)
	491	{
	492	return atomic_read(&req->ref);
	493	}
	494
bc9fcbf9	495	#endif /* BLK_INTERNAL_H */