[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;

static inline void __blk_get_queue(struct request_queue *q)
{
	kobject_get(&q->kobj);
}

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 request_queue *q,
		struct bio_vec *bprv, unsigned int offset)
{
	return (offset & queue_virt_boundary(q)) ||
		((bprv->bv_offset + bprv->bv_len) & queue_virt_boundary(q));
}

/*
 * 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 request_queue *q,
		struct bio_vec *bprv, unsigned int offset)
{
	if (!queue_virt_boundary(q))
		return false;
	return __bvec_gap_to_prev(q, 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,
						     int 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, &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, &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_mq(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 blk_may_split(struct request_queue *q, struct bio *bio)
{
	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 q->limits.chunk_sectors || bio->bi_vcnt != 1 ||
		bio->bi_io_vec->bv_len + bio->bi_io_vec->bv_offset > PAGE_SIZE;
}

void __blk_queue_split(struct request_queue *q, struct bio **bio,
			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;
}

/*
 * The max size one bio can handle is UINT_MAX becasue bvec_iter.bi_size
 * is defined as 'unsigned int', meantime it has to aligned to with logical
 * block size which is the minimum accepted unit by hardware.
 */
static inline unsigned int bio_allowed_max_sectors(struct request_queue *q)
{
	return round_down(UINT_MAX, queue_logical_block_size(q)) >> 9;
}

/*
 * 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

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

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

#ifdef CONFIG_BLK_CGROUP_IOLATENCY
extern int blk_iolatency_init(struct request_queue *q);
#else
static inline int blk_iolatency_init(struct request_queue *q) { 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
09ac46c4 TH	34	static inline void __blk_get_queue(struct request_queue *q)
	35	{
	36	kobject_get(&q->kobj);
	37	}
	38
a9ed27a7	39	bool is_flush_rq(struct request *req);
8d699663	40
754a1572 GJ	41	struct blk_flush_queue *blk_alloc_flush_queue(int node, int cmd_size,
754a1572 GJ	42	gfp_t flags);
f70ced09	43	void blk_free_flush_queue(struct blk_flush_queue *q);
f3552655	44
3ef28e83	45	void blk_freeze_queue(struct request_queue *q);
aec89dc5	46	void __blk_mq_unfreeze_queue(struct request_queue *q, bool force_atomic);
8e141f9e	47	void blk_queue_start_drain(struct request_queue *q);
c98cb5bb	48	int __bio_queue_enter(struct request_queue q, struct bio bio);
3f98c753	49	void submit_bio_noacct_nocheck(struct bio *bio);
c98cb5bb JA	50
	51	static inline bool blk_try_enter_queue(struct request_queue *q, bool pm)
	52	{
	53	rcu_read_lock();
	54	if (!percpu_ref_tryget_live_rcu(&q->q_usage_counter))
	55	goto fail;
	56
	57	/*
	58	* The code that increments the pm_only counter must ensure that the
	59	* counter is globally visible before the queue is unfrozen.
	60	*/
	61	if (blk_queue_pm_only(q) &&
	62	(!pm \|\| queue_rpm_status(q) == RPM_SUSPENDED))
	63	goto fail_put;
	64
	65	rcu_read_unlock();
	66	return true;
	67
	68	fail_put:
	69	blk_queue_exit(q);
	70	fail:
	71	rcu_read_unlock();
	72	return false;
	73	}
	74
	75	static inline int bio_queue_enter(struct bio *bio)
	76	{
	77	struct request_queue *q = bdev_get_queue(bio->bi_bdev);
	78
	79	if (blk_try_enter_queue(q, false))
	80	return 0;
	81	return __bio_queue_enter(q, bio);
	82	}
3ef28e83	83
dc0b8a57	84	#define BIO_INLINE_VECS 4
7a800a20 CH	85	struct bio_vec bvec_alloc(mempool_t pool, unsigned short *nr_vecs,
	86	gfp_t gfp_mask);
	87	void bvec_free(mempool_t pool, struct bio_vec bv, unsigned short nr_vecs);
eec716a1	88
3dccdae5 CH	89	static inline bool biovec_phys_mergeable(struct request_queue *q,
3dccdae5 CH	90	struct bio_vec vec1, struct bio_vec vec2)
6a9f5f24	91	{
3dccdae5	92	unsigned long mask = queue_segment_boundary(q);
6e768461 CH	93	phys_addr_t addr1 = page_to_phys(vec1->bv_page) + vec1->bv_offset;
6e768461 CH	94	phys_addr_t addr2 = page_to_phys(vec2->bv_page) + vec2->bv_offset;
3dccdae5 CH	95
3dccdae5 CH	96	if (addr1 + vec1->bv_len != addr2)
6a9f5f24	97	return false;
0383ad43	98	if (xen_domain() && !xen_biovec_phys_mergeable(vec1, vec2->bv_page))
6a9f5f24	99	return false;
3dccdae5 CH	100	if ((addr1 \| mask) != ((addr2 + vec2->bv_len - 1) \| mask))
3dccdae5 CH	101	return false;
6a9f5f24 CH	102	return true;
	103	}
	104
27ca1d4e CH	105	static inline bool __bvec_gap_to_prev(struct request_queue *q,
	106	struct bio_vec *bprv, unsigned int offset)
	107	{
df376b2e	108	return (offset & queue_virt_boundary(q)) \|\|
27ca1d4e CH	109	((bprv->bv_offset + bprv->bv_len) & queue_virt_boundary(q));
	110	}
	111
	112	/*
	113	* Check if adding a bio_vec after bprv with offset would create a gap in
	114	* the SG list. Most drivers don't care about this, but some do.
	115	*/
	116	static inline bool bvec_gap_to_prev(struct request_queue *q,
	117	struct bio_vec *bprv, unsigned int offset)
	118	{
	119	if (!queue_virt_boundary(q))
	120	return false;
	121	return __bvec_gap_to_prev(q, bprv, offset);
	122	}
	123
badf7f64 CH	124	static inline bool rq_mergeable(struct request *rq)
	125	{
	126	if (blk_rq_is_passthrough(rq))
	127	return false;
	128
	129	if (req_op(rq) == REQ_OP_FLUSH)
	130	return false;
	131
	132	if (req_op(rq) == REQ_OP_WRITE_ZEROES)
	133	return false;
	134
	135	if (req_op(rq) == REQ_OP_ZONE_APPEND)
	136	return false;
	137
	138	if (rq->cmd_flags & REQ_NOMERGE_FLAGS)
	139	return false;
	140	if (rq->rq_flags & RQF_NOMERGE_FLAGS)
	141	return false;
	142
	143	return true;
	144	}
	145
	146	/*
	147	* There are two different ways to handle DISCARD merges:
	148	* 1) If max_discard_segments > 1, the driver treats every bio as a range and
	149	* send the bios to controller together. The ranges don't need to be
	150	* contiguous.
	151	* 2) Otherwise, the request will be normal read/write requests. The ranges
	152	* need to be contiguous.
	153	*/
	154	static inline bool blk_discard_mergable(struct request *req)
	155	{
	156	if (req_op(req) == REQ_OP_DISCARD &&
	157	queue_max_discard_segments(req->q) > 1)
	158	return true;
	159	return false;
	160	}
	161
2a9336c4 CH	162	static inline unsigned int blk_queue_get_max_sectors(struct request_queue *q,
	163	int op)
	164	{
	165	if (unlikely(op == REQ_OP_DISCARD \|\| op == REQ_OP_SECURE_ERASE))
	166	return min(q->limits.max_discard_sectors,
	167	UINT_MAX >> SECTOR_SHIFT);
	168
	169	if (unlikely(op == REQ_OP_WRITE_ZEROES))
	170	return q->limits.max_write_zeroes_sectors;
	171
	172	return q->limits.max_sectors;
	173	}
	174
5a48fc14 DW	175	#ifdef CONFIG_BLK_DEV_INTEGRITY
5a48fc14 DW	176	void blk_flush_integrity(void);
7c20f116	177	bool __bio_integrity_endio(struct bio *);
ece841ab	178	void bio_integrity_free(struct bio *bio);
7c20f116 CH	179	static inline bool bio_integrity_endio(struct bio *bio)
	180	{
	181	if (bio_integrity(bio))
	182	return __bio_integrity_endio(bio);
	183	return true;
	184	}
43b729bf	185
92cf2fd1 CH	186	bool blk_integrity_merge_rq(struct request_queue , struct request ,
92cf2fd1 CH	187	struct request *);
d59da419 CH	188	bool blk_integrity_merge_bio(struct request_queue , struct request ,
d59da419 CH	189	struct bio *);
92cf2fd1	190
43b729bf CH	191	static inline bool integrity_req_gap_back_merge(struct request *req,
	192	struct bio *next)
	193	{
	194	struct bio_integrity_payload *bip = bio_integrity(req->bio);
	195	struct bio_integrity_payload *bip_next = bio_integrity(next);
	196
	197	return bvec_gap_to_prev(req->q, &bip->bip_vec[bip->bip_vcnt - 1],
	198	bip_next->bip_vec[0].bv_offset);
	199	}
	200
	201	static inline bool integrity_req_gap_front_merge(struct request *req,
	202	struct bio *bio)
	203	{
	204	struct bio_integrity_payload *bip = bio_integrity(bio);
	205	struct bio_integrity_payload *bip_next = bio_integrity(req->bio);
	206
	207	return bvec_gap_to_prev(req->q, &bip->bip_vec[bip->bip_vcnt - 1],
	208	bip_next->bip_vec[0].bv_offset);
	209	}
581e2600	210
614310c9	211	int blk_integrity_add(struct gendisk *disk);
581e2600	212	void blk_integrity_del(struct gendisk *);
43b729bf	213	#else /* CONFIG_BLK_DEV_INTEGRITY */
92cf2fd1 CH	214	static inline bool blk_integrity_merge_rq(struct request_queue *rq,
	215	struct request r1, struct request r2)
	216	{
	217	return true;
	218	}
d59da419 CH	219	static inline bool blk_integrity_merge_bio(struct request_queue *rq,
	220	struct request r, struct bio b)
	221	{
	222	return true;
	223	}
43b729bf CH	224	static inline bool integrity_req_gap_back_merge(struct request *req,
	225	struct bio *next)
	226	{
	227	return false;
	228	}
	229	static inline bool integrity_req_gap_front_merge(struct request *req,
	230	struct bio *bio)
	231	{
	232	return false;
	233	}
	234
5a48fc14 DW	235	static inline void blk_flush_integrity(void)
	236	{
	237	}
7c20f116 CH	238	static inline bool bio_integrity_endio(struct bio *bio)
	239	{
	240	return true;
	241	}
ece841ab JT	242	static inline void bio_integrity_free(struct bio *bio)
	243	{
	244	}
614310c9	245	static inline int blk_integrity_add(struct gendisk *disk)
581e2600	246	{
614310c9	247	return 0;
581e2600 CH	248	}
	249	static inline void blk_integrity_del(struct gendisk *disk)
	250	{
	251	}
43b729bf	252	#endif /* CONFIG_BLK_DEV_INTEGRITY */
8324aa91	253
0d2602ca	254	unsigned long blk_rq_timeout(unsigned long timeout);
87ee7b11	255	void blk_add_timer(struct request *req);
0d7a29a2	256	const char *blk_status_to_str(blk_status_t status);
320ae51f	257
320ae51f	258	bool blk_attempt_plug_merge(struct request_queue q, struct bio bio,
0c5bcc92	259	unsigned int nr_segs);
bdc6a287 BW	260	bool blk_bio_list_merge(struct request_queue q, struct list_head list,
bdc6a287 BW	261	struct bio *bio, unsigned int nr_segs);
320ae51f	262
ba0ffdd8 JA	263	/*
	264	* Plug flush limits
	265	*/
	266	#define BLK_MAX_REQUEST_COUNT 32
	267	#define BLK_PLUG_FLUSH_SIZE (128 * 1024)
	268
158dbda0 TH	269	/*
	270	* Internal elevator interface
	271	*/
e8064021	272	#define ELV_ON_HASH(rq) ((rq)->rq_flags & RQF_HASHED)
158dbda0	273
2b504bd4	274	void blk_insert_flush(struct request *rq);
dd831006	275
d48ece20 JW	276	int elevator_switch_mq(struct request_queue *q,
d48ece20 JW	277	struct elevator_type *new_e);
0c6cb3a2	278	void elevator_exit(struct request_queue *q);
cecf5d87	279	int elv_register_queue(struct request_queue *q, bool uevent);
83d016ac BVA	280	void elv_unregister_queue(struct request_queue *q);
83d016ac BVA	281
3ad5cee5 CH	282	ssize_t part_size_show(struct device dev, struct device_attribute attr,
	283	char *buf);
	284	ssize_t part_stat_show(struct device dev, struct device_attribute attr,
	285	char *buf);
	286	ssize_t part_inflight_show(struct device dev, struct device_attribute attr,
	287	char *buf);
	288	ssize_t part_fail_show(struct device dev, struct device_attribute attr,
	289	char *buf);
	290	ssize_t part_fail_store(struct device dev, struct device_attribute attr,
	291	const char *buf, size_t count);
581d4e28 JA	292	ssize_t part_timeout_show(struct device , struct device_attribute , char *);
	293	ssize_t part_timeout_store(struct device , struct device_attribute ,
	294	const char *, size_t);
581d4e28	295
abd45c15 JA	296	static inline bool blk_may_split(struct request_queue q, struct bio bio)
	297	{
	298	switch (bio_op(bio)) {
	299	case REQ_OP_DISCARD:
	300	case REQ_OP_SECURE_ERASE:
	301	case REQ_OP_WRITE_ZEROES:
abd45c15 JA	302	return true; /* non-trivial splitting decisions */
	303	default:
	304	break;
	305	}
	306
	307	/*
	308	* All drivers must accept single-segments bios that are <= PAGE_SIZE.
	309	* This is a quick and dirty check that relies on the fact that
	310	* bi_io_vec[0] is always valid if a bio has data. The check might
	311	* lead to occasional false negatives when bios are cloned, but compared
	312	* to the performance impact of cloned bios themselves the loop below
	313	* doesn't matter anyway.
	314	*/
	315	return q->limits.chunk_sectors \|\| bio->bi_vcnt != 1 \|\|
	316	bio->bi_io_vec->bv_len + bio->bi_io_vec->bv_offset > PAGE_SIZE;
	317	}
	318
	319	void __blk_queue_split(struct request_queue q, struct bio *bio,
	320	unsigned int *nr_segs);
14ccb66b CH	321	int ll_back_merge_fn(struct request req, struct bio bio,
14ccb66b CH	322	unsigned int nr_segs);
fd2ef39c	323	bool blk_attempt_req_merge(struct request_queue q, struct request rq,
5e84ea3a	324	struct request *next);
e9cd19c0	325	unsigned int blk_recalc_rq_segments(struct request *rq);
80a761fd	326	void blk_rq_set_mixed_merge(struct request *rq);
050c8ea8	327	bool blk_rq_merge_ok(struct request rq, struct bio bio);
34fe7c05	328	enum elv_merge blk_try_merge(struct request rq, struct bio bio);
d6d48196	329
ff88972c AB	330	int blk_dev_init(void);
ff88972c AB	331
c2553b58 JA	332	/*
	333	* Contribute to IO statistics IFF:
	334	*
	335	* a) it's attached to a gendisk, and
48d9b0d4	336	* b) the queue had IO stats enabled when this request was started
c2553b58	337	*/
599d067d	338	static inline bool blk_do_io_stat(struct request *rq)
fb8ec18c	339	{
41fa7222	340	return (rq->rq_flags & RQF_IO_STAT) && !blk_rq_is_passthrough(rq);
be6bfe36 PB	341	}
be6bfe36 PB	342
450b7879	343	void update_io_ticks(struct block_device *part, unsigned long now, bool end);
fb8ec18c	344
6cf7677f CH	345	static inline void req_set_nomerge(struct request_queue q, struct request req)
	346	{
	347	req->cmd_flags \|= REQ_NOMERGE;
	348	if (req == q->last_merge)
	349	q->last_merge = NULL;
	350	}
	351
1adfc5e4 ML	352	/*
	353	* The max size one bio can handle is UINT_MAX becasue bvec_iter.bi_size
	354	* is defined as 'unsigned int', meantime it has to aligned to with logical
	355	* block size which is the minimum accepted unit by hardware.
	356	*/
	357	static inline unsigned int bio_allowed_max_sectors(struct request_queue *q)
	358	{
	359	return round_down(UINT_MAX, queue_logical_block_size(q)) >> 9;
	360	}
	361
f2dbd76a TH	362	/*
	363	* Internal io_context interface
	364	*/
87dd1d63	365	struct io_cq ioc_find_get_icq(struct request_queue q);
eca5892a	366	struct io_cq ioc_lookup_icq(struct request_queue q);
5ef16305	367	#ifdef CONFIG_BLK_ICQ
7e5a8794	368	void ioc_clear_queue(struct request_queue *q);
5ef16305 CH	369	#else
	370	static inline void ioc_clear_queue(struct request_queue *q)
	371	{
	372	}
	373	#endif /* CONFIG_BLK_ICQ */
f2dbd76a	374
297e3d85 SL	375	#ifdef CONFIG_BLK_DEV_THROTTLING_LOW
	376	extern ssize_t blk_throtl_sample_time_show(struct request_queue q, char page);
	377	extern ssize_t blk_throtl_sample_time_store(struct request_queue *q,
	378	const char *page, size_t count);
9e234eea	379	extern void blk_throtl_bio_endio(struct bio *bio);
b9147dd1	380	extern void blk_throtl_stat_add(struct request *rq, u64 time);
9e234eea SL	381	#else
9e234eea SL	382	static inline void blk_throtl_bio_endio(struct bio *bio) { }
b9147dd1	383	static inline void blk_throtl_stat_add(struct request *rq, u64 time) { }
297e3d85	384	#endif
bc9fcbf9	385
9bb33f24 CH	386	void __blk_queue_bounce(struct request_queue q, struct bio *bio);
	387
	388	static inline bool blk_queue_may_bounce(struct request_queue *q)
	389	{
	390	return IS_ENABLED(CONFIG_BOUNCE) &&
	391	q->limits.bounce == BLK_BOUNCE_HIGH &&
	392	max_low_pfn >= max_pfn;
	393	}
	394
3bce016a CH	395	static inline void blk_queue_bounce(struct request_queue q, struct bio *bio)
3bce016a CH	396	{
9bb33f24 CH	397	if (unlikely(blk_queue_may_bounce(q) && bio_has_data(*bio)))
9bb33f24 CH	398	__blk_queue_bounce(q, bio);
3bce016a	399	}
3bce016a	400
d7067512 JB	401	#ifdef CONFIG_BLK_CGROUP_IOLATENCY
	402	extern int blk_iolatency_init(struct request_queue *q);
	403	#else
	404	static inline int blk_iolatency_init(struct request_queue *q) { return 0; }
	405	#endif
	406
bf505456	407	#ifdef CONFIG_BLK_DEV_ZONED
5d400665	408	void disk_free_zone_bitmaps(struct gendisk *disk);
b3c72f81	409	void disk_clear_zone_settings(struct gendisk *disk);
bf505456	410	#else
5d400665	411	static inline void disk_free_zone_bitmaps(struct gendisk *disk) {}
b3c72f81	412	static inline void disk_clear_zone_settings(struct gendisk *disk) {}
bf505456 DLM	413	#endif
bf505456 DLM	414
7c3f828b CH	415	int blk_alloc_ext_minor(void);
7c3f828b CH	416	void blk_free_ext_minor(unsigned int minor);
581e2600 CH	417	#define ADDPART_FLAG_NONE 0
	418	#define ADDPART_FLAG_RAID 1
	419	#define ADDPART_FLAG_WHOLEDISK 2
7f6be376 CH	420	int bdev_add_partition(struct gendisk *disk, int partno, sector_t start,
7f6be376 CH	421	sector_t length);
926fbb16	422	int bdev_del_partition(struct gendisk *disk, int partno);
3d2e7989 CH	423	int bdev_resize_partition(struct gendisk *disk, int partno, sector_t start,
3d2e7989 CH	424	sector_t length);
e7243285	425	void blk_drop_partitions(struct gendisk *disk);
581e2600	426
6f8191fd CH	427	struct gendisk __alloc_disk_node(struct request_queue q, int node_id,
	428	struct lock_class_key *lkclass);
	429
e4581105	430	int bio_add_hw_page(struct request_queue q, struct bio bio,
130879f1	431	struct page *page, unsigned int len, unsigned int offset,
e4581105	432	unsigned int max_sectors, bool *same_page);
130879f1	433
704b914f ML	434	static inline struct kmem_cache *blk_get_queue_kmem_cache(bool srcu)
	435	{
	436	if (srcu)
	437	return blk_requestq_srcu_cachep;
	438	return blk_requestq_cachep;
	439	}
	440	struct request_queue *blk_alloc_queue(int node_id, bool alloc_srcu);
	441
e16e506c	442	int disk_scan_partitions(struct gendisk *disk, fmode_t mode);
da7ba729	443
92e7755e	444	int disk_alloc_events(struct gendisk *disk);
d5870edf CH	445	void disk_add_events(struct gendisk *disk);
	446	void disk_del_events(struct gendisk *disk);
	447	void disk_release_events(struct gendisk *disk);
926597ff CH	448	void disk_block_events(struct gendisk *disk);
	449	void disk_unblock_events(struct gendisk *disk);
	450	void disk_flush_events(struct gendisk *disk, unsigned int mask);
2bc8cda5 CH	451	extern struct device_attribute dev_attr_events;
	452	extern struct device_attribute dev_attr_events_async;
	453	extern struct device_attribute dev_attr_events_poll_msecs;
d5870edf	454
cc5c516d CH	455	extern struct attribute_group blk_trace_attr_group;
cc5c516d CH	456
8a709512	457	long blkdev_ioctl(struct file *file, unsigned cmd, unsigned long arg);
84b8514b CH	458	long compat_blkdev_ioctl(struct file *file, unsigned cmd, unsigned long arg);
84b8514b CH	459
cd82cca7 CH	460	extern const struct address_space_operations def_blk_aops;
cd82cca7 CH	461
22d0c408	462	int disk_register_independent_access_ranges(struct gendisk *disk);
a2247f19 DLM	463	void disk_unregister_independent_access_ranges(struct gendisk *disk);
a2247f19 DLM	464
06c8c691 CH	465	#ifdef CONFIG_FAIL_MAKE_REQUEST
	466	bool should_fail_request(struct block_device *part, unsigned int bytes);
	467	#else /* CONFIG_FAIL_MAKE_REQUEST */
	468	static inline bool should_fail_request(struct block_device *part,
	469	unsigned int bytes)
	470	{
	471	return false;
	472	}
	473	#endif /* CONFIG_FAIL_MAKE_REQUEST */
	474
0a467d0f JA	475	/*
	476	* Optimized request reference counting. Ideally we'd make timeouts be more
	477	* clever, as that's the only reason we need references at all... But until
	478	* this happens, this is faster than using refcount_t. Also see:
	479	*
	480	* abc54d634334 ("io_uring: switch to atomic_t for io_kiocb reference count")
	481	*/
	482	#define req_ref_zero_or_close_to_overflow(req) \
	483	((unsigned int) atomic_read(&(req->ref)) + 127u <= 127u)
	484
	485	static inline bool req_ref_inc_not_zero(struct request *req)
	486	{
	487	return atomic_inc_not_zero(&req->ref);
	488	}
	489
	490	static inline bool req_ref_put_and_test(struct request *req)
	491	{
	492	WARN_ON_ONCE(req_ref_zero_or_close_to_overflow(req));
	493	return atomic_dec_and_test(&req->ref);
	494	}
	495
	496	static inline void req_ref_set(struct request *req, int value)
	497	{
	498	atomic_set(&req->ref, value);
	499	}
	500
	501	static inline int req_ref_read(struct request *req)
	502	{
	503	return atomic_read(&req->ref);
	504	}
	505
bc9fcbf9	506	#endif /* BLK_INTERNAL_H */