blkdev.h 62.2 KB
Newer Older
1
/* SPDX-License-Identifier: GPL-2.0 */
L
Linus Torvalds 已提交
2 3 4
#ifndef _LINUX_BLKDEV_H
#define _LINUX_BLKDEV_H

5
#include <linux/sched.h>
6
#include <linux/sched/clock.h>
7

J
Jens Axboe 已提交
8 9
#ifdef CONFIG_BLOCK

L
Linus Torvalds 已提交
10 11 12
#include <linux/major.h>
#include <linux/genhd.h>
#include <linux/list.h>
13
#include <linux/llist.h>
L
Linus Torvalds 已提交
14 15 16
#include <linux/timer.h>
#include <linux/workqueue.h>
#include <linux/pagemap.h>
17
#include <linux/backing-dev-defs.h>
L
Linus Torvalds 已提交
18 19
#include <linux/wait.h>
#include <linux/mempool.h>
D
Dan Williams 已提交
20
#include <linux/pfn.h>
L
Linus Torvalds 已提交
21 22
#include <linux/bio.h>
#include <linux/stringify.h>
23
#include <linux/gfp.h>
24
#include <linux/bsg.h>
25
#include <linux/smp.h>
T
Tejun Heo 已提交
26
#include <linux/rcupdate.h>
27
#include <linux/percpu-refcount.h>
C
Christoph Hellwig 已提交
28
#include <linux/scatterlist.h>
29
#include <linux/blkzoned.h>
L
Linus Torvalds 已提交
30

31
struct module;
32 33
struct scsi_ioctl_command;

L
Linus Torvalds 已提交
34 35
struct request_queue;
struct elevator_queue;
36
struct blk_trace;
37 38
struct request;
struct sg_io_hdr;
M
Mike Christie 已提交
39
struct bsg_job;
T
Tejun Heo 已提交
40
struct blkcg_gq;
41
struct blk_flush_queue;
42
struct pr_ops;
43
struct rq_qos;
44 45
struct blk_queue_stats;
struct blk_stat_callback;
L
Linus Torvalds 已提交
46 47 48 49

#define BLKDEV_MIN_RQ	4
#define BLKDEV_MAX_RQ	128	/* Default maximum */

50
/* Must be consistent with blk_mq_poll_stats_bkt() */
51 52
#define BLK_MQ_POLL_STATS_BKTS 16

T
Tejun Heo 已提交
53 54 55 56
/*
 * Maximum number of blkcg policies allowed to be registered concurrently.
 * Defined here to simplify include dependency.
 */
57
#define BLKCG_MAX_POLS		5
T
Tejun Heo 已提交
58

59
typedef void (rq_end_io_fn)(struct request *, blk_status_t);
L
Linus Torvalds 已提交
60

61 62 63
#define BLK_RL_SYNCFULL		(1U << 0)
#define BLK_RL_ASYNCFULL	(1U << 1)

L
Linus Torvalds 已提交
64
struct request_list {
65
	struct request_queue	*q;	/* the queue this rl belongs to */
66 67 68
#ifdef CONFIG_BLK_CGROUP
	struct blkcg_gq		*blkg;	/* blkg this request pool belongs to */
#endif
69 70 71 72
	/*
	 * count[], starved[], and wait[] are indexed by
	 * BLK_RW_SYNC/BLK_RW_ASYNC
	 */
73 74 75 76
	int			count[2];
	int			starved[2];
	mempool_t		*rq_pool;
	wait_queue_head_t	wait[2];
77
	unsigned int		flags;
L
Linus Torvalds 已提交
78 79
};

80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118
/*
 * request flags */
typedef __u32 __bitwise req_flags_t;

/* elevator knows about this request */
#define RQF_SORTED		((__force req_flags_t)(1 << 0))
/* drive already may have started this one */
#define RQF_STARTED		((__force req_flags_t)(1 << 1))
/* uses tagged queueing */
#define RQF_QUEUED		((__force req_flags_t)(1 << 2))
/* may not be passed by ioscheduler */
#define RQF_SOFTBARRIER		((__force req_flags_t)(1 << 3))
/* request for flush sequence */
#define RQF_FLUSH_SEQ		((__force req_flags_t)(1 << 4))
/* merge of different types, fail separately */
#define RQF_MIXED_MERGE		((__force req_flags_t)(1 << 5))
/* track inflight for MQ */
#define RQF_MQ_INFLIGHT		((__force req_flags_t)(1 << 6))
/* don't call prep for this one */
#define RQF_DONTPREP		((__force req_flags_t)(1 << 7))
/* set for "ide_preempt" requests and also for requests for which the SCSI
   "quiesce" state must be ignored. */
#define RQF_PREEMPT		((__force req_flags_t)(1 << 8))
/* contains copies of user pages */
#define RQF_COPY_USER		((__force req_flags_t)(1 << 9))
/* vaguely specified driver internal error.  Ignored by the block layer */
#define RQF_FAILED		((__force req_flags_t)(1 << 10))
/* don't warn about errors */
#define RQF_QUIET		((__force req_flags_t)(1 << 11))
/* elevator private data attached */
#define RQF_ELVPRIV		((__force req_flags_t)(1 << 12))
/* account I/O stat */
#define RQF_IO_STAT		((__force req_flags_t)(1 << 13))
/* request came from our alloc pool */
#define RQF_ALLOCED		((__force req_flags_t)(1 << 14))
/* runtime pm request */
#define RQF_PM			((__force req_flags_t)(1 << 15))
/* on IO scheduler merge hash */
#define RQF_HASHED		((__force req_flags_t)(1 << 16))
119 120
/* IO stats tracking on */
#define RQF_STATS		((__force req_flags_t)(1 << 17))
121 122 123
/* Look at ->special_vec for the actual data payload instead of the
   bio chain. */
#define RQF_SPECIAL_PAYLOAD	((__force req_flags_t)(1 << 18))
124 125
/* The per-zone write lock is held for this request */
#define RQF_ZONE_WRITE_LOCKED	((__force req_flags_t)(1 << 19))
J
Jens Axboe 已提交
126
/* already slept for hybrid poll */
K
Keith Busch 已提交
127
#define RQF_MQ_POLL_SLEPT	((__force req_flags_t)(1 << 20))
128 129
/* ->timeout has been called, don't expire again */
#define RQF_TIMED_OUT		((__force req_flags_t)(1 << 21))
130 131 132

/* flags that prevent us from merging requests: */
#define RQF_NOMERGE_FLAGS \
133
	(RQF_STARTED | RQF_SOFTBARRIER | RQF_FLUSH_SEQ | RQF_SPECIAL_PAYLOAD)
134

K
Keith Busch 已提交
135 136 137 138 139 140 141 142 143
/*
 * Request state for blk-mq.
 */
enum mq_rq_state {
	MQ_RQ_IDLE		= 0,
	MQ_RQ_IN_FLIGHT		= 1,
	MQ_RQ_COMPLETE		= 2,
};

L
Linus Torvalds 已提交
144
/*
145 146 147 148
 * Try to put the fields that are referenced together in the same cacheline.
 *
 * If you modify this structure, make sure to update blk_rq_init() and
 * especially blk_mq_rq_ctx_init() to take care of the added fields.
L
Linus Torvalds 已提交
149 150
 */
struct request {
151
	struct request_queue *q;
152
	struct blk_mq_ctx *mq_ctx;
153
	struct blk_mq_hw_ctx *mq_hctx;
154

155
	int cpu;
156
	unsigned int cmd_flags;		/* op and common flags */
157
	req_flags_t rq_flags;
158 159 160

	int internal_tag;

161 162
	/* the following two fields are internal, NEVER access directly */
	unsigned int __data_len;	/* total data len */
163
	int tag;
164
	sector_t __sector;		/* sector cursor */
L
Linus Torvalds 已提交
165 166 167 168

	struct bio *bio;
	struct bio *biotail;

169 170
	struct list_head queuelist;

171 172 173 174 175 176 177 178 179 180 181 182
	/*
	 * The hash is used inside the scheduler, and killed once the
	 * request reaches the dispatch list. The ipi_list is only used
	 * to queue the request for softirq completion, which is long
	 * after the request has been unhashed (and even removed from
	 * the dispatch list).
	 */
	union {
		struct hlist_node hash;	/* merge hash */
		struct list_head ipi_list;
	};

183 184 185
	/*
	 * The rb_node is only used inside the io scheduler, requests
	 * are pruned when moved to the dispatch queue. So let the
186
	 * completion_data share space with the rb_node.
187 188 189
	 */
	union {
		struct rb_node rb_node;	/* sort/lookup */
190
		struct bio_vec special_vec;
191
		void *completion_data;
192
		int error_count; /* for legacy drivers, don't use */
193
	};
194

195
	/*
196
	 * Three pointers are available for the IO schedulers, if they need
197 198
	 * more they have to dynamically allocate it.  Flush requests are
	 * never put on the IO scheduler. So let the flush fields share
199
	 * space with the elevator data.
200
	 */
201
	union {
202 203 204 205 206
		struct {
			struct io_cq		*icq;
			void			*priv[2];
		} elv;

207 208 209
		struct {
			unsigned int		seq;
			struct list_head	list;
210
			rq_end_io_fn		*saved_end_io;
211 212
		} flush;
	};
213

214
	struct gendisk *rq_disk;
215
	struct hd_struct *part;
216 217 218 219 220
#ifdef CONFIG_BLK_RQ_ALLOC_TIME
	/* Time that the first bio started allocating this request. */
	u64 alloc_time_ns;
#endif
	/* Time that this request was allocated for this IO. */
221
	u64 start_time_ns;
222 223 224 225 226 227
	/* Time that I/O was submitted to the device. */
	u64 io_start_time_ns;

#ifdef CONFIG_BLK_WBT
	unsigned short wbt_flags;
#endif
228 229 230 231 232 233
	/*
	 * rq sectors used for blk stats. It has the same value
	 * with blk_rq_sectors(rq), except that it never be zeroed
	 * by completion.
	 */
	unsigned short stats_sectors;
234 235 236

	/*
	 * Number of scatter-gather DMA addr+len pairs after
L
Linus Torvalds 已提交
237 238 239
	 * physical address coalescing is performed.
	 */
	unsigned short nr_phys_segments;
240

241 242 243
#if defined(CONFIG_BLK_DEV_INTEGRITY)
	unsigned short nr_integrity_segments;
#endif
L
Linus Torvalds 已提交
244

245
	unsigned short write_hint;
246 247
	unsigned short ioprio;

T
Tejun Heo 已提交
248
	void *special;		/* opaque pointer available for LLD use */
249

250
	unsigned int extra_len;	/* length of alignment and padding */
L
Linus Torvalds 已提交
251

K
Keith Busch 已提交
252 253
	enum mq_rq_state state;
	refcount_t ref;
254

255 256
	unsigned int timeout;

257 258
	/* access through blk_rq_set_deadline, blk_rq_deadline */
	unsigned long __deadline;
259

J
Jens Axboe 已提交
260
	struct list_head timeout_list;
L
Linus Torvalds 已提交
261

262
	union {
263
		struct __call_single_data csd;
264 265 266
		u64 fifo_time;
	};

L
Linus Torvalds 已提交
267
	/*
268
	 * completion callback.
L
Linus Torvalds 已提交
269 270 271
	 */
	rq_end_io_fn *end_io;
	void *end_io_data;
272 273 274

	/* for bidi */
	struct request *next_rq;
275 276 277 278

#ifdef CONFIG_BLK_CGROUP
	struct request_list *rl;		/* rl this rq is alloced from */
#endif
L
Linus Torvalds 已提交
279 280
};

281 282 283 284 285 286 287 288 289 290
static inline bool blk_op_is_scsi(unsigned int op)
{
	return op == REQ_OP_SCSI_IN || op == REQ_OP_SCSI_OUT;
}

static inline bool blk_op_is_private(unsigned int op)
{
	return op == REQ_OP_DRV_IN || op == REQ_OP_DRV_OUT;
}

291 292
static inline bool blk_rq_is_scsi(struct request *rq)
{
293
	return blk_op_is_scsi(req_op(rq));
294 295 296 297
}

static inline bool blk_rq_is_private(struct request *rq)
{
298
	return blk_op_is_private(req_op(rq));
299 300
}

301 302
static inline bool blk_rq_is_passthrough(struct request *rq)
{
303
	return blk_rq_is_scsi(rq) || blk_rq_is_private(rq);
304 305
}

306 307 308 309 310 311 312
static inline bool bio_is_passthrough(struct bio *bio)
{
	unsigned op = bio_op(bio);

	return blk_op_is_scsi(op) || blk_op_is_private(op);
}

313 314 315 316 317
static inline unsigned short req_get_ioprio(struct request *req)
{
	return req->ioprio;
}

L
Linus Torvalds 已提交
318 319
#include <linux/elevator.h>

320 321
struct blk_queue_ctx;

322
typedef void (request_fn_proc) (struct request_queue *q);
323
typedef blk_qc_t (make_request_fn) (struct request_queue *q, struct bio *bio);
324
typedef int (prep_rq_fn) (struct request_queue *, struct request *);
325
typedef void (unprep_rq_fn) (struct request_queue *, struct request *);
L
Linus Torvalds 已提交
326 327

struct bio_vec;
328
typedef void (softirq_done_fn)(struct request *);
329
typedef int (dma_drain_needed_fn)(struct request *);
330
typedef int (lld_busy_fn) (struct request_queue *q);
M
Mike Christie 已提交
331
typedef int (bsg_job_fn) (struct bsg_job *);
332 333
typedef int (init_rq_fn)(struct request_queue *, struct request *, gfp_t);
typedef void (exit_rq_fn)(struct request_queue *, struct request *);
L
Linus Torvalds 已提交
334

J
Jens Axboe 已提交
335
enum blk_eh_timer_return {
336 337
	BLK_EH_DONE,		/* drivers has completed the command */
	BLK_EH_RESET_TIMER,	/* reset timer and try again */
J
Jens Axboe 已提交
338 339 340 341
};

typedef enum blk_eh_timer_return (rq_timed_out_fn)(struct request *);

L
Linus Torvalds 已提交
342 343 344 345 346 347 348 349 350
enum blk_queue_state {
	Queue_down,
	Queue_up,
};

struct blk_queue_tag {
	struct request **tag_index;	/* map of busy tags */
	unsigned long *tag_map;		/* bit map of free/busy tags */
	int max_depth;			/* what we will send to device */
351
	int real_max_depth;		/* what the array can hold */
L
Linus Torvalds 已提交
352
	atomic_t refcnt;		/* map can be shared */
353 354
	int alloc_policy;		/* tag allocation policy */
	int next_tag;			/* next tag */
L
Linus Torvalds 已提交
355
};
356 357
#define BLK_TAG_ALLOC_FIFO 0 /* allocate starting from 0 */
#define BLK_TAG_ALLOC_RR 1 /* allocate starting from last allocated tag */
L
Linus Torvalds 已提交
358

359 360 361
#define BLK_SCSI_MAX_CMDS	(256)
#define BLK_SCSI_CMD_PER_LONG	(BLK_SCSI_MAX_CMDS / (sizeof(long) * 8))

D
Damien Le Moal 已提交
362 363 364 365 366 367 368 369 370
/*
 * Zoned block device models (zoned limit).
 */
enum blk_zoned_model {
	BLK_ZONED_NONE,	/* Regular block device */
	BLK_ZONED_HA,	/* Host-aware zoned block device */
	BLK_ZONED_HM,	/* Host-managed zoned block device */
};

371 372 373
struct queue_limits {
	unsigned long		bounce_pfn;
	unsigned long		seg_boundary_mask;
374
	unsigned long		virt_boundary_mask;
375 376

	unsigned int		max_hw_sectors;
377
	unsigned int		max_dev_sectors;
378
	unsigned int		chunk_sectors;
379 380
	unsigned int		max_sectors;
	unsigned int		max_segment_size;
381
	unsigned int		physical_block_size;
382
	unsigned int		logical_block_size;
383 384 385
	unsigned int		alignment_offset;
	unsigned int		io_min;
	unsigned int		io_opt;
386
	unsigned int		max_discard_sectors;
387
	unsigned int		max_hw_discard_sectors;
388
	unsigned int		max_write_same_sectors;
389
	unsigned int		max_write_zeroes_sectors;
390 391
	unsigned int		discard_granularity;
	unsigned int		discard_alignment;
392

393
	unsigned short		max_segments;
394
	unsigned short		max_integrity_segments;
395
	unsigned short		max_discard_segments;
396

397
	unsigned char		misaligned;
398
	unsigned char		discard_misaligned;
399
	unsigned char		cluster;
400
	unsigned char		raid_partial_stripes_expensive;
D
Damien Le Moal 已提交
401
	enum blk_zoned_model	zoned;
402 403 404

	ALI_HOTFIX_RESERVE(1)
	ALI_HOTFIX_RESERVE(2)
405 406
};

407 408 409 410 411 412 413 414 415 416 417 418 419
#ifdef CONFIG_BLK_DEV_ZONED

struct blk_zone_report_hdr {
	unsigned int	nr_zones;
	u8		padding[60];
};

extern int blkdev_report_zones(struct block_device *bdev,
			       sector_t sector, struct blk_zone *zones,
			       unsigned int *nr_zones, gfp_t gfp_mask);
extern int blkdev_reset_zones(struct block_device *bdev, sector_t sectors,
			      sector_t nr_sectors, gfp_t gfp_mask);

S
Shaun Tancheff 已提交
420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440
extern int blkdev_report_zones_ioctl(struct block_device *bdev, fmode_t mode,
				     unsigned int cmd, unsigned long arg);
extern int blkdev_reset_zones_ioctl(struct block_device *bdev, fmode_t mode,
				    unsigned int cmd, unsigned long arg);

#else /* CONFIG_BLK_DEV_ZONED */

static inline int blkdev_report_zones_ioctl(struct block_device *bdev,
					    fmode_t mode, unsigned int cmd,
					    unsigned long arg)
{
	return -ENOTTY;
}

static inline int blkdev_reset_zones_ioctl(struct block_device *bdev,
					   fmode_t mode, unsigned int cmd,
					   unsigned long arg)
{
	return -ENOTTY;
}

441 442
#endif /* CONFIG_BLK_DEV_ZONED */

443 444 445 446 447 448
/*
 * default request hang threshold, unit is millisecond. If one request does
 * not complete in this threashold time, consider this request as hang.
 */
#define BLK_REQ_HANG_THRESHOLD  5000

449
struct request_queue {
L
Linus Torvalds 已提交
450 451 452 453 454
	/*
	 * Together with queue_head for cacheline sharing
	 */
	struct list_head	queue_head;
	struct request		*last_merge;
J
Jens Axboe 已提交
455
	struct elevator_queue	*elevator;
456 457
	int			nr_rqs[2];	/* # allocated [a]sync rqs */
	int			nr_rqs_elvpriv;	/* # allocated rqs w/ elvpriv */
L
Linus Torvalds 已提交
458

459
	struct blk_queue_stats	*stats;
460
	struct rq_qos		*rq_qos;
J
Jens Axboe 已提交
461

L
Linus Torvalds 已提交
462
	/*
463 464 465 466
	 * If blkcg is not used, @q->root_rl serves all requests.  If blkcg
	 * is used, root blkg allocates from @q->root_rl and all other
	 * blkgs from their own blkg->rl.  Which one to use should be
	 * determined using bio_request_list().
L
Linus Torvalds 已提交
467
	 */
468
	struct request_list	root_rl;
L
Linus Torvalds 已提交
469 470 471 472

	request_fn_proc		*request_fn;
	make_request_fn		*make_request_fn;
	prep_rq_fn		*prep_rq_fn;
473
	unprep_rq_fn		*unprep_rq_fn;
474
	softirq_done_fn		*softirq_done_fn;
J
Jens Axboe 已提交
475
	rq_timed_out_fn		*rq_timed_out_fn;
476
	dma_drain_needed_fn	*dma_drain_needed;
477
	lld_busy_fn		*lld_busy_fn;
478
	/* Called just after a request is allocated */
479
	init_rq_fn		*init_rq_fn;
480
	/* Called just before a request is freed */
481
	exit_rq_fn		*exit_rq_fn;
482 483
	/* Called from inside blk_get_request() */
	void (*initialize_rq_fn)(struct request *rq);
L
Linus Torvalds 已提交
484

J
Jens Axboe 已提交
485
	const struct blk_mq_ops	*mq_ops;
486 487

	/* sw queues */
488
	struct blk_mq_ctx __percpu	*queue_ctx;
489 490
	unsigned int		nr_queues;

491 492
	unsigned int		queue_depth;

493 494 495 496
	/* hw dispatch queues */
	struct blk_mq_hw_ctx	**queue_hw_ctx;
	unsigned int		nr_hw_queues;

497 498 499
	/*
	 * Dispatch queue sorting
	 */
J
Jens Axboe 已提交
500
	sector_t		end_sector;
501 502
	struct request		*boundary_rq;

L
Linus Torvalds 已提交
503
	/*
504
	 * Delayed queue handling
L
Linus Torvalds 已提交
505
	 */
506
	struct delayed_work	delay_work;
L
Linus Torvalds 已提交
507

508
	struct backing_dev_info	*backing_dev_info;
L
Linus Torvalds 已提交
509 510 511 512 513 514 515 516

	/*
	 * The queue owner gets to use this for whatever they like.
	 * ll_rw_blk doesn't touch it.
	 */
	void			*queuedata;

	/*
517
	 * various queue flags, see QUEUE_* below
L
Linus Torvalds 已提交
518
	 */
519
	unsigned long		queue_flags;
520 521 522 523 524 525
	/*
	 * Number of contexts that have called blk_set_pm_only(). If this
	 * counter is above zero then only RQF_PM and RQF_PREEMPT requests are
	 * processed.
	 */
	atomic_t		pm_only;
L
Linus Torvalds 已提交
526

527 528 529 530 531 532
	/*
	 * ida allocated id for this queue.  Used to index queues from
	 * ioctx.
	 */
	int			id;

L
Linus Torvalds 已提交
533
	/*
534
	 * queue needs bounce pages for pages above this limit
L
Linus Torvalds 已提交
535
	 */
536
	gfp_t			bounce_gfp;
L
Linus Torvalds 已提交
537 538

	/*
已提交
539 540 541
	 * protects queue structures from reentrancy. ->__queue_lock should
	 * _never_ be used directly, it is queue private. always use
	 * ->queue_lock.
L
Linus Torvalds 已提交
542
	 */
已提交
543
	spinlock_t		__queue_lock;
L
Linus Torvalds 已提交
544 545 546 547 548 549 550
	spinlock_t		*queue_lock;

	/*
	 * queue kobject
	 */
	struct kobject kobj;

551 552 553
	/*
	 * mq queue kobject
	 */
554
	struct kobject *mq_kobj;
555

556 557 558 559
#ifdef  CONFIG_BLK_DEV_INTEGRITY
	struct blk_integrity integrity;
#endif	/* CONFIG_BLK_DEV_INTEGRITY */

560
#ifdef CONFIG_PM
L
Lin Ming 已提交
561 562 563 564 565
	struct device		*dev;
	int			rpm_status;
	unsigned int		nr_pending;
#endif

L
Linus Torvalds 已提交
566 567 568 569 570 571 572 573
	/*
	 * queue settings
	 */
	unsigned long		nr_requests;	/* Max # of requests */
	unsigned int		nr_congestion_on;
	unsigned int		nr_congestion_off;
	unsigned int		nr_batching;

J
James Bottomley 已提交
574
	unsigned int		dma_drain_size;
575
	void			*dma_drain_buffer;
576
	unsigned int		dma_pad_mask;
L
Linus Torvalds 已提交
577 578 579 580
	unsigned int		dma_alignment;

	struct blk_queue_tag	*queue_tags;

581
	unsigned int		nr_sorted;
582
	unsigned int		in_flight[2];
583

584 585 586 587 588 589
	/*
	 * Number of active block driver functions for which blk_drain_queue()
	 * must wait. Must be incremented around functions that unlock the
	 * queue_lock internally, e.g. scsi_request_fn().
	 */
	unsigned int		request_fn_active;
L
Linus Torvalds 已提交
590

J
Jens Axboe 已提交
591
	unsigned int		rq_timeout;
592
	unsigned int		rq_hang_threshold;
593
	int			poll_nsec;
594 595

	struct blk_stat_callback	*poll_cb;
596
	struct blk_rq_stat	poll_stat[BLK_MQ_POLL_STATS_BKTS];
597

J
Jens Axboe 已提交
598
	struct timer_list	timeout;
599
	struct work_struct	timeout_work;
J
Jens Axboe 已提交
600 601
	struct list_head	timeout_list;

602
	struct list_head	icq_list;
603
#ifdef CONFIG_BLK_CGROUP
604
	DECLARE_BITMAP		(blkcg_pols, BLKCG_MAX_POLS);
T
Tejun Heo 已提交
605
	struct blkcg_gq		*root_blkg;
606
	struct list_head	blkg_list;
607
#endif
608

609 610
	struct queue_limits	limits;

611
#ifdef CONFIG_BLK_DEV_ZONED
612 613 614 615 616 617 618 619 620 621 622
	/*
	 * Zoned block device information for request dispatch control.
	 * nr_zones is the total number of zones of the device. This is always
	 * 0 for regular block devices. seq_zones_bitmap is a bitmap of nr_zones
	 * bits which indicates if a zone is conventional (bit clear) or
	 * sequential (bit set). seq_zones_wlock is a bitmap of nr_zones
	 * bits which indicates if a zone is write locked, that is, if a write
	 * request targeting the zone was dispatched. All three fields are
	 * initialized by the low level device driver (e.g. scsi/sd.c).
	 * Stacking drivers (device mappers) may or may not initialize
	 * these fields.
623 624 625 626 627
	 *
	 * Reads of this information must be protected with blk_queue_enter() /
	 * blk_queue_exit(). Modifying this information is only allowed while
	 * no requests are being processed. See also blk_mq_freeze_queue() and
	 * blk_mq_unfreeze_queue().
628 629 630 631
	 */
	unsigned int		nr_zones;
	unsigned long		*seq_zones_bitmap;
	unsigned long		*seq_zones_wlock;
632
#endif /* CONFIG_BLK_DEV_ZONED */
633

L
Linus Torvalds 已提交
634 635 636 637 638
	/*
	 * sg stuff
	 */
	unsigned int		sg_timeout;
	unsigned int		sg_reserved_size;
639
	int			node;
640
#ifdef CONFIG_BLK_DEV_IO_TRACE
641
	struct blk_trace __rcu	*blk_trace;
642
	struct mutex		blk_trace_mutex;
643
#endif
L
Linus Torvalds 已提交
644
	/*
645
	 * for flush operations
L
Linus Torvalds 已提交
646
	 */
647
	struct blk_flush_queue	*fq;
648

649 650
	struct list_head	requeue_list;
	spinlock_t		requeue_lock;
651
	struct delayed_work	requeue_work;
652

653
	struct mutex		sysfs_lock;
654

655
	int			bypass_depth;
656
	atomic_t		mq_freeze_depth;
657

658
#if defined(CONFIG_BLK_DEV_BSG)
M
Mike Christie 已提交
659
	bsg_job_fn		*bsg_job_fn;
660 661
	struct bsg_class_device bsg_dev;
#endif
662 663 664 665 666

#ifdef CONFIG_BLK_DEV_THROTTLING
	/* Throttle data */
	struct throtl_data *td;
#endif
T
Tejun Heo 已提交
667
	struct rcu_head		rcu_head;
668
	wait_queue_head_t	mq_freeze_wq;
669
	struct percpu_ref	q_usage_counter;
670
	struct list_head	all_q_node;
671 672 673

	struct blk_mq_tag_set	*tag_set;
	struct list_head	tag_set_list;
674
	struct bio_set		bio_split;
675

676
#ifdef CONFIG_BLK_DEBUG_FS
677
	struct dentry		*debugfs_dir;
678
	struct dentry		*sched_debugfs_dir;
679 680
#endif

681
	bool			mq_sysfs_init_done;
682 683 684

	size_t			cmd_size;
	void			*rq_alloc_data;
685 686

	struct work_struct	release_work;
687 688 689

#define BLK_MAX_WRITE_HINTS	5
	u64			write_hints[BLK_MAX_WRITE_HINTS];
690 691 692 693 694

	ALI_HOTFIX_RESERVE(1)
	ALI_HOTFIX_RESERVE(2)
	ALI_HOTFIX_RESERVE(3)
	ALI_HOTFIX_RESERVE(4)
L
Linus Torvalds 已提交
695 696
};

697 698 699 700 701 702 703 704 705
#define QUEUE_FLAG_QUEUED	0	/* uses generic tag queueing */
#define QUEUE_FLAG_STOPPED	1	/* queue is stopped */
#define QUEUE_FLAG_DYING	2	/* queue being torn down */
#define QUEUE_FLAG_BYPASS	3	/* act as dumb FIFO queue */
#define QUEUE_FLAG_BIDI		4	/* queue supports bidi requests */
#define QUEUE_FLAG_NOMERGES     5	/* disable merge attempts */
#define QUEUE_FLAG_SAME_COMP	6	/* complete on same CPU-group */
#define QUEUE_FLAG_FAIL_IO	7	/* fake timeout */
#define QUEUE_FLAG_NONROT	9	/* non-rotational device (SSD) */
706
#define QUEUE_FLAG_VIRT        QUEUE_FLAG_NONROT /* paravirt device */
707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725
#define QUEUE_FLAG_IO_STAT     10	/* do IO stats */
#define QUEUE_FLAG_DISCARD     11	/* supports DISCARD */
#define QUEUE_FLAG_NOXMERGES   12	/* No extended merges */
#define QUEUE_FLAG_ADD_RANDOM  13	/* Contributes to random pool */
#define QUEUE_FLAG_SECERASE    14	/* supports secure erase */
#define QUEUE_FLAG_SAME_FORCE  15	/* force complete on same CPU */
#define QUEUE_FLAG_DEAD        16	/* queue tear-down finished */
#define QUEUE_FLAG_INIT_DONE   17	/* queue is initialized */
#define QUEUE_FLAG_NO_SG_MERGE 18	/* don't attempt to merge SG segments*/
#define QUEUE_FLAG_POLL	       19	/* IO polling enabled if set */
#define QUEUE_FLAG_WC	       20	/* Write back caching */
#define QUEUE_FLAG_FUA	       21	/* device supports FUA writes */
#define QUEUE_FLAG_FLUSH_NQ    22	/* flush not queueuable */
#define QUEUE_FLAG_DAX         23	/* device supports DAX */
#define QUEUE_FLAG_STATS       24	/* track rq completion times */
#define QUEUE_FLAG_POLL_STATS  25	/* collecting stats for hybrid polling */
#define QUEUE_FLAG_REGISTERED  26	/* queue has been registered to a disk */
#define QUEUE_FLAG_SCSI_PASSTHROUGH 27	/* queue supports SCSI commands */
#define QUEUE_FLAG_QUIESCED    28	/* queue has been quiesced */
726
#define QUEUE_FLAG_RQ_ALLOC_TIME 29	/* record rq->alloc_time_ns */
727 728

#define QUEUE_FLAG_DEFAULT	((1 << QUEUE_FLAG_IO_STAT) |		\
729 730
				 (1 << QUEUE_FLAG_SAME_COMP)	|	\
				 (1 << QUEUE_FLAG_ADD_RANDOM))
731

732
#define QUEUE_FLAG_MQ_DEFAULT	((1 << QUEUE_FLAG_IO_STAT) |		\
733
				 (1 << QUEUE_FLAG_SAME_COMP))
734

735 736 737 738 739
void blk_queue_flag_set(unsigned int flag, struct request_queue *q);
void blk_queue_flag_clear(unsigned int flag, struct request_queue *q);
bool blk_queue_flag_test_and_set(unsigned int flag, struct request_queue *q);
bool blk_queue_flag_test_and_clear(unsigned int flag, struct request_queue *q);

L
Linus Torvalds 已提交
740 741
#define blk_queue_tagged(q)	test_bit(QUEUE_FLAG_QUEUED, &(q)->queue_flags)
#define blk_queue_stopped(q)	test_bit(QUEUE_FLAG_STOPPED, &(q)->queue_flags)
B
Bart Van Assche 已提交
742
#define blk_queue_dying(q)	test_bit(QUEUE_FLAG_DYING, &(q)->queue_flags)
743
#define blk_queue_dead(q)	test_bit(QUEUE_FLAG_DEAD, &(q)->queue_flags)
744
#define blk_queue_bypass(q)	test_bit(QUEUE_FLAG_BYPASS, &(q)->queue_flags)
745
#define blk_queue_init_done(q)	test_bit(QUEUE_FLAG_INIT_DONE, &(q)->queue_flags)
746
#define blk_queue_nomerges(q)	test_bit(QUEUE_FLAG_NOMERGES, &(q)->queue_flags)
747 748
#define blk_queue_noxmerges(q)	\
	test_bit(QUEUE_FLAG_NOXMERGES, &(q)->queue_flags)
749
#define blk_queue_nonrot(q)	test_bit(QUEUE_FLAG_NONROT, &(q)->queue_flags)
750
#define blk_queue_io_stat(q)	test_bit(QUEUE_FLAG_IO_STAT, &(q)->queue_flags)
751
#define blk_queue_add_random(q)	test_bit(QUEUE_FLAG_ADD_RANDOM, &(q)->queue_flags)
752
#define blk_queue_discard(q)	test_bit(QUEUE_FLAG_DISCARD, &(q)->queue_flags)
753 754
#define blk_queue_secure_erase(q) \
	(test_bit(QUEUE_FLAG_SECERASE, &(q)->queue_flags))
755
#define blk_queue_dax(q)	test_bit(QUEUE_FLAG_DAX, &(q)->queue_flags)
756 757
#define blk_queue_scsi_passthrough(q)	\
	test_bit(QUEUE_FLAG_SCSI_PASSTHROUGH, &(q)->queue_flags)
758 759 760 761 762 763
#ifdef CONFIG_BLK_RQ_ALLOC_TIME
#define blk_queue_rq_alloc_time(q)	\
	test_bit(QUEUE_FLAG_RQ_ALLOC_TIME, &(q)->queue_flags)
#else
#define blk_queue_rq_alloc_time(q)	false
#endif
L
Linus Torvalds 已提交
764

765 766 767
#define blk_noretry_request(rq) \
	((rq)->cmd_flags & (REQ_FAILFAST_DEV|REQ_FAILFAST_TRANSPORT| \
			     REQ_FAILFAST_DRIVER))
768
#define blk_queue_quiesced(q)	test_bit(QUEUE_FLAG_QUIESCED, &(q)->queue_flags)
769
#define blk_queue_pm_only(q)	atomic_read(&(q)->pm_only)
770
#define blk_queue_fua(q)	test_bit(QUEUE_FLAG_FUA, &(q)->queue_flags)
771

772 773
extern void blk_set_pm_only(struct request_queue *q);
extern void blk_clear_pm_only(struct request_queue *q);
774

775 776 777 778 779
static inline int queue_in_flight(struct request_queue *q)
{
	return q->in_flight[0] + q->in_flight[1];
}

780 781 782 783
static inline bool blk_account_rq(struct request *rq)
{
	return (rq->rq_flags & RQF_STARTED) && !blk_rq_is_passthrough(rq);
}
784

785
#define blk_rq_cpu_valid(rq)	((rq)->cpu != -1)
786
#define blk_bidi_rq(rq)		((rq)->next_rq != NULL)
787 788
/* rq->queuelist of dequeued request must be list_empty() */
#define blk_queued_rq(rq)	(!list_empty(&(rq)->queuelist))
L
Linus Torvalds 已提交
789 790 791

#define list_entry_rq(ptr)	list_entry((ptr), struct request, queuelist)

792
#define rq_data_dir(rq)		(op_is_write(req_op(rq)) ? WRITE : READ)
L
Linus Torvalds 已提交
793

794 795 796 797 798 799 800 801 802
/*
 * Driver can handle struct request, if it either has an old style
 * request_fn defined, or is blk-mq based.
 */
static inline bool queue_is_rq_based(struct request_queue *q)
{
	return q->request_fn || q->mq_ops;
}

803 804 805 806 807
static inline unsigned int blk_queue_cluster(struct request_queue *q)
{
	return q->limits.cluster;
}

D
Damien Le Moal 已提交
808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824
static inline enum blk_zoned_model
blk_queue_zoned_model(struct request_queue *q)
{
	return q->limits.zoned;
}

static inline bool blk_queue_is_zoned(struct request_queue *q)
{
	switch (blk_queue_zoned_model(q)) {
	case BLK_ZONED_HA:
	case BLK_ZONED_HM:
		return true;
	default:
		return false;
	}
}

825
static inline unsigned int blk_queue_zone_sectors(struct request_queue *q)
826 827 828 829
{
	return blk_queue_is_zoned(q) ? q->limits.chunk_sectors : 0;
}

830
#ifdef CONFIG_BLK_DEV_ZONED
831 832 833 834 835 836 837 838 839 840 841 842 843 844 845
static inline unsigned int blk_queue_zone_no(struct request_queue *q,
					     sector_t sector)
{
	if (!blk_queue_is_zoned(q))
		return 0;
	return sector >> ilog2(q->limits.chunk_sectors);
}

static inline bool blk_queue_zone_is_seq(struct request_queue *q,
					 sector_t sector)
{
	if (!blk_queue_is_zoned(q) || !q->seq_zones_bitmap)
		return false;
	return test_bit(blk_queue_zone_no(q, sector), q->seq_zones_bitmap);
}
846
#endif /* CONFIG_BLK_DEV_ZONED */
847

848 849
static inline bool rq_is_sync(struct request *rq)
{
850
	return op_is_sync(rq->cmd_flags);
851 852
}

853
static inline bool blk_rl_full(struct request_list *rl, bool sync)
L
Linus Torvalds 已提交
854
{
855 856 857
	unsigned int flag = sync ? BLK_RL_SYNCFULL : BLK_RL_ASYNCFULL;

	return rl->flags & flag;
L
Linus Torvalds 已提交
858 859
}

860
static inline void blk_set_rl_full(struct request_list *rl, bool sync)
L
Linus Torvalds 已提交
861
{
862 863 864
	unsigned int flag = sync ? BLK_RL_SYNCFULL : BLK_RL_ASYNCFULL;

	rl->flags |= flag;
L
Linus Torvalds 已提交
865 866
}

867
static inline void blk_clear_rl_full(struct request_list *rl, bool sync)
L
Linus Torvalds 已提交
868
{
869 870 871
	unsigned int flag = sync ? BLK_RL_SYNCFULL : BLK_RL_ASYNCFULL;

	rl->flags &= ~flag;
L
Linus Torvalds 已提交
872 873
}

874 875
static inline bool rq_mergeable(struct request *rq)
{
876
	if (blk_rq_is_passthrough(rq))
877
		return false;
L
Linus Torvalds 已提交
878

879 880 881
	if (req_op(rq) == REQ_OP_FLUSH)
		return false;

882 883 884
	if (req_op(rq) == REQ_OP_WRITE_ZEROES)
		return false;

885
	if (rq->cmd_flags & REQ_NOMERGE_FLAGS)
886 887
		return false;
	if (rq->rq_flags & RQF_NOMERGE_FLAGS)
888 889 890 891
		return false;

	return true;
}
L
Linus Torvalds 已提交
892

893 894
static inline bool blk_write_same_mergeable(struct bio *a, struct bio *b)
{
895 896
	if (bio_page(a) == bio_page(b) &&
	    bio_offset(a) == bio_offset(b))
897 898 899 900 901
		return true;

	return false;
}

902 903 904 905 906 907 908 909
static inline unsigned int blk_queue_depth(struct request_queue *q)
{
	if (q->queue_depth)
		return q->queue_depth;

	return q->nr_requests;
}

L
Linus Torvalds 已提交
910 911 912
/*
 * q->prep_rq_fn return values
 */
913 914 915 916 917 918
enum {
	BLKPREP_OK,		/* serve it */
	BLKPREP_KILL,		/* fatal error, kill, return -EIO */
	BLKPREP_DEFER,		/* leave on queue */
	BLKPREP_INVALID,	/* invalid command, kill, return -EREMOTEIO */
};
L
Linus Torvalds 已提交
919 920 921 922 923 924 925 926 927 928

extern unsigned long blk_max_low_pfn, blk_max_pfn;

/*
 * standard bounce addresses:
 *
 * BLK_BOUNCE_HIGH	: bounce all highmem pages
 * BLK_BOUNCE_ANY	: don't bounce anything
 * BLK_BOUNCE_ISA	: bounce pages above ISA DMA boundary
 */
929 930

#if BITS_PER_LONG == 32
L
Linus Torvalds 已提交
931
#define BLK_BOUNCE_HIGH		((u64)blk_max_low_pfn << PAGE_SHIFT)
932 933 934 935
#else
#define BLK_BOUNCE_HIGH		-1ULL
#endif
#define BLK_BOUNCE_ANY		(-1ULL)
936
#define BLK_BOUNCE_ISA		(DMA_BIT_MASK(24))
L
Linus Torvalds 已提交
937

938 939 940 941
/*
 * default timeout for SG_IO if none specified
 */
#define BLK_DEFAULT_SG_TIMEOUT	(60 * HZ)
L
Linus Torvalds 已提交
942
#define BLK_MIN_SG_TIMEOUT	(7 * HZ)
943

944 945 946 947
struct rq_map_data {
	struct page **pages;
	int page_order;
	int nr_entries;
948
	unsigned long offset;
949
	int null_mapped;
950
	int from_user;
951 952
};

953
struct req_iterator {
954
	struct bvec_iter iter;
955 956 957 958
	struct bio *bio;
};

/* This should not be used directly - use rq_for_each_segment */
959 960
#define for_each_bio(_bio)		\
	for (; _bio; _bio = _bio->bi_next)
961
#define __rq_for_each_bio(_bio, rq)	\
L
Linus Torvalds 已提交
962 963 964
	if ((rq->bio))			\
		for (_bio = (rq)->bio; _bio; _bio = _bio->bi_next)

965 966
#define rq_for_each_segment(bvl, _rq, _iter)			\
	__rq_for_each_bio(_iter.bio, _rq)			\
967
		bio_for_each_segment(bvl, _iter.bio, _iter.iter)
968

K
Kent Overstreet 已提交
969
#define rq_iter_last(bvec, _iter)				\
970
		(_iter.bio->bi_next == NULL &&			\
K
Kent Overstreet 已提交
971
		 bio_iter_last(bvec, _iter.iter))
972

973 974 975 976 977 978 979 980 981 982 983
#ifndef ARCH_IMPLEMENTS_FLUSH_DCACHE_PAGE
# error	"You should define ARCH_IMPLEMENTS_FLUSH_DCACHE_PAGE for your platform"
#endif
#if ARCH_IMPLEMENTS_FLUSH_DCACHE_PAGE
extern void rq_flush_dcache_pages(struct request *rq);
#else
static inline void rq_flush_dcache_pages(struct request *rq)
{
}
#endif

L
Linus Torvalds 已提交
984 985
extern int blk_register_queue(struct gendisk *disk);
extern void blk_unregister_queue(struct gendisk *disk);
986
extern blk_qc_t generic_make_request(struct bio *bio);
987
extern blk_qc_t direct_make_request(struct bio *bio);
988
extern void blk_rq_init(struct request_queue *q, struct request *rq);
989
extern void blk_init_request_from_bio(struct request *req, struct bio *bio);
L
Linus Torvalds 已提交
990
extern void blk_put_request(struct request *);
991
extern void __blk_put_request(struct request_queue *, struct request *);
992
extern struct request *blk_get_request(struct request_queue *, unsigned int op,
993
				       blk_mq_req_flags_t flags);
994
extern void blk_requeue_request(struct request_queue *, struct request *);
995
extern int blk_lld_busy(struct request_queue *q);
996 997 998 999 1000
extern int blk_rq_prep_clone(struct request *rq, struct request *rq_src,
			     struct bio_set *bs, gfp_t gfp_mask,
			     int (*bio_ctr)(struct bio *, struct bio *, void *),
			     void *data);
extern void blk_rq_unprep_clone(struct request *rq);
1001
extern blk_status_t blk_insert_cloned_request(struct request_queue *q,
1002
				     struct request *rq);
J
Jens Axboe 已提交
1003
extern int blk_rq_append_bio(struct request *rq, struct bio **bio);
1004
extern void blk_delay_queue(struct request_queue *, unsigned long);
1005
extern void blk_queue_split(struct request_queue *, struct bio **);
1006
extern void blk_recount_segments(struct request_queue *, struct bio *);
1007
extern int scsi_verify_blk_ioctl(struct block_device *, unsigned int);
1008 1009
extern int scsi_cmd_blk_ioctl(struct block_device *, fmode_t,
			      unsigned int, void __user *);
1010 1011
extern int scsi_cmd_ioctl(struct request_queue *, struct gendisk *, fmode_t,
			  unsigned int, void __user *);
1012 1013
extern int sg_scsi_ioctl(struct request_queue *, struct gendisk *, fmode_t,
			 struct scsi_ioctl_command __user *);
1014

1015
extern int blk_queue_enter(struct request_queue *q, blk_mq_req_flags_t flags);
1016
extern void blk_queue_exit(struct request_queue *q);
1017
extern void blk_start_queue(struct request_queue *q);
J
Jens Axboe 已提交
1018
extern void blk_start_queue_async(struct request_queue *q);
1019
extern void blk_stop_queue(struct request_queue *q);
L
Linus Torvalds 已提交
1020
extern void blk_sync_queue(struct request_queue *q);
1021
extern void __blk_stop_queue(struct request_queue *q);
1022
extern void __blk_run_queue(struct request_queue *q);
1023
extern void __blk_run_queue_uncond(struct request_queue *q);
1024
extern void blk_run_queue(struct request_queue *);
1025
extern void blk_run_queue_async(struct request_queue *q);
1026
extern int blk_rq_map_user(struct request_queue *, struct request *,
1027 1028
			   struct rq_map_data *, void __user *, unsigned long,
			   gfp_t);
J
Jens Axboe 已提交
1029
extern int blk_rq_unmap_user(struct bio *);
1030 1031
extern int blk_rq_map_kern(struct request_queue *, struct request *, void *, unsigned int, gfp_t);
extern int blk_rq_map_user_iov(struct request_queue *, struct request *,
1032 1033
			       struct rq_map_data *, const struct iov_iter *,
			       gfp_t);
1034
extern void blk_execute_rq(struct request_queue *, struct gendisk *,
1035
			  struct request *, int);
1036
extern void blk_execute_rq_nowait(struct request_queue *, struct gendisk *,
1037
				  struct request *, int, rq_end_io_fn *);
1038

1039 1040 1041
int blk_status_to_errno(blk_status_t status);
blk_status_t errno_to_blk_status(int errno);

1042
int blk_poll(struct request_queue *q, blk_qc_t cookie, bool spin);
J
Jens Axboe 已提交
1043

1044
static inline struct request_queue *bdev_get_queue(struct block_device *bdev)
L
Linus Torvalds 已提交
1045
{
1046
	return bdev->bd_disk->queue;	/* this is never NULL */
L
Linus Torvalds 已提交
1047 1048
}

1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061
/*
 * The basic unit of block I/O is a sector. It is used in a number of contexts
 * in Linux (blk, bio, genhd). The size of one sector is 512 = 2**9
 * bytes. Variables of type sector_t represent an offset or size that is a
 * multiple of 512 bytes. Hence these two constants.
 */
#ifndef SECTOR_SHIFT
#define SECTOR_SHIFT 9
#endif
#ifndef SECTOR_SIZE
#define SECTOR_SIZE (1 << SECTOR_SHIFT)
#endif

1062
/*
1063 1064 1065 1066 1067 1068
 * blk_rq_pos()			: the current sector
 * blk_rq_bytes()		: bytes left in the entire request
 * blk_rq_cur_bytes()		: bytes left in the current segment
 * blk_rq_err_bytes()		: bytes left till the next error boundary
 * blk_rq_sectors()		: sectors left in the entire request
 * blk_rq_cur_sectors()		: sectors left in the current segment
1069
 * blk_rq_stats_sectors()	: sectors of the entire request used for stats
1070
 */
1071 1072
static inline sector_t blk_rq_pos(const struct request *rq)
{
1073
	return rq->__sector;
1074 1075 1076 1077
}

static inline unsigned int blk_rq_bytes(const struct request *rq)
{
1078
	return rq->__data_len;
1079 1080
}

1081 1082 1083 1084
static inline int blk_rq_cur_bytes(const struct request *rq)
{
	return rq->bio ? bio_cur_bytes(rq->bio) : 0;
}
1085

1086 1087
extern unsigned int blk_rq_err_bytes(const struct request *rq);

1088 1089
static inline unsigned int blk_rq_sectors(const struct request *rq)
{
1090
	return blk_rq_bytes(rq) >> SECTOR_SHIFT;
1091 1092 1093 1094
}

static inline unsigned int blk_rq_cur_sectors(const struct request *rq)
{
1095
	return blk_rq_cur_bytes(rq) >> SECTOR_SHIFT;
1096 1097
}

1098 1099 1100 1101 1102
static inline unsigned int blk_rq_stats_sectors(const struct request *rq)
{
	return rq->stats_sectors;
}

1103
#ifdef CONFIG_BLK_DEV_ZONED
1104 1105 1106 1107 1108 1109 1110 1111 1112
static inline unsigned int blk_rq_zone_no(struct request *rq)
{
	return blk_queue_zone_no(rq->q, blk_rq_pos(rq));
}

static inline unsigned int blk_rq_zone_is_seq(struct request *rq)
{
	return blk_queue_zone_is_seq(rq->q, blk_rq_pos(rq));
}
1113
#endif /* CONFIG_BLK_DEV_ZONED */
1114

1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127
/*
 * Some commands like WRITE SAME have a payload or data transfer size which
 * is different from the size of the request.  Any driver that supports such
 * commands using the RQF_SPECIAL_PAYLOAD flag needs to use this helper to
 * calculate the data transfer size.
 */
static inline unsigned int blk_rq_payload_bytes(struct request *rq)
{
	if (rq->rq_flags & RQF_SPECIAL_PAYLOAD)
		return rq->special_vec.bv_len;
	return blk_rq_bytes(rq);
}

C
Christoph Hellwig 已提交
1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138
/*
 * Return the first full biovec in the request.  The caller needs to check that
 * there are any bvecs before calling this helper.
 */
static inline struct bio_vec req_bvec(struct request *rq)
{
	if (rq->rq_flags & RQF_SPECIAL_PAYLOAD)
		return rq->special_vec;
	return bvec_iter_bvec(rq->bio->bi_io_vec, rq->bio->bi_iter);
}

1139
static inline unsigned int blk_queue_get_max_sectors(struct request_queue *q,
1140
						     int op)
1141
{
A
Adrian Hunter 已提交
1142
	if (unlikely(op == REQ_OP_DISCARD || op == REQ_OP_SECURE_ERASE))
1143 1144
		return min(q->limits.max_discard_sectors,
			   UINT_MAX >> SECTOR_SHIFT);
1145

1146
	if (unlikely(op == REQ_OP_WRITE_SAME))
1147 1148
		return q->limits.max_write_same_sectors;

1149 1150 1151
	if (unlikely(op == REQ_OP_WRITE_ZEROES))
		return q->limits.max_write_zeroes_sectors;

1152 1153 1154
	return q->limits.max_sectors;
}

1155 1156 1157 1158 1159 1160 1161 1162
/*
 * Return maximum size of a request at given offset. Only valid for
 * file system requests.
 */
static inline unsigned int blk_max_size_offset(struct request_queue *q,
					       sector_t offset)
{
	if (!q->limits.chunk_sectors)
1163
		return q->limits.max_sectors;
1164

1165 1166
	return min(q->limits.max_sectors, (unsigned int)(q->limits.chunk_sectors -
			(offset & (q->limits.chunk_sectors - 1))));
1167 1168
}

D
Damien Le Moal 已提交
1169 1170
static inline unsigned int blk_rq_get_max_sectors(struct request *rq,
						  sector_t offset)
1171 1172 1173
{
	struct request_queue *q = rq->q;

1174
	if (blk_rq_is_passthrough(rq))
1175 1176
		return q->limits.max_hw_sectors;

A
Adrian Hunter 已提交
1177 1178 1179
	if (!q->limits.chunk_sectors ||
	    req_op(rq) == REQ_OP_DISCARD ||
	    req_op(rq) == REQ_OP_SECURE_ERASE)
1180
		return blk_queue_get_max_sectors(q, req_op(rq));
1181

D
Damien Le Moal 已提交
1182
	return min(blk_max_size_offset(q, offset),
1183
			blk_queue_get_max_sectors(q, req_op(rq)));
1184 1185
}

1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196
static inline unsigned int blk_rq_count_bios(struct request *rq)
{
	unsigned int nr_bios = 0;
	struct bio *bio;

	__rq_for_each_bio(bio, rq)
		nr_bios++;

	return nr_bios;
}

1197 1198 1199 1200 1201 1202 1203
/*
 * Request issue related functions.
 */
extern struct request *blk_peek_request(struct request_queue *q);
extern void blk_start_request(struct request *rq);
extern struct request *blk_fetch_request(struct request_queue *q);

1204 1205
void blk_steal_bios(struct bio_list *list, struct request *rq);

L
Linus Torvalds 已提交
1206
/*
1207 1208 1209 1210 1211
 * Request completion related functions.
 *
 * blk_update_request() completes given number of bytes and updates
 * the request without completing it.
 *
1212 1213
 * blk_end_request() and friends.  __blk_end_request() must be called
 * with the request queue spinlock acquired.
L
Linus Torvalds 已提交
1214 1215
 *
 * Several drivers define their own end_request and call
1216 1217
 * blk_end_request() for parts of the original function.
 * This prevents code duplication in drivers.
L
Linus Torvalds 已提交
1218
 */
1219
extern bool blk_update_request(struct request *rq, blk_status_t error,
1220
			       unsigned int nr_bytes);
1221 1222
extern void blk_finish_request(struct request *rq, blk_status_t error);
extern bool blk_end_request(struct request *rq, blk_status_t error,
1223
			    unsigned int nr_bytes);
1224 1225
extern void blk_end_request_all(struct request *rq, blk_status_t error);
extern bool __blk_end_request(struct request *rq, blk_status_t error,
1226
			      unsigned int nr_bytes);
1227 1228
extern void __blk_end_request_all(struct request *rq, blk_status_t error);
extern bool __blk_end_request_cur(struct request *rq, blk_status_t error);
1229

1230
extern void blk_complete_request(struct request *);
J
Jens Axboe 已提交
1231 1232
extern void __blk_complete_request(struct request *);
extern void blk_abort_request(struct request *);
1233
extern void blk_unprep_request(struct request *);
1234

L
Linus Torvalds 已提交
1235 1236 1237
/*
 * Access functions for manipulating queue properties
 */
1238
extern struct request_queue *blk_init_queue_node(request_fn_proc *rfn,
1239
					spinlock_t *lock, int node_id);
1240
extern struct request_queue *blk_init_queue(request_fn_proc *, spinlock_t *);
1241
extern int blk_init_allocated_queue(struct request_queue *);
1242 1243 1244
extern void blk_cleanup_queue(struct request_queue *);
extern void blk_queue_make_request(struct request_queue *, make_request_fn *);
extern void blk_queue_bounce_limit(struct request_queue *, u64);
1245
extern void blk_queue_max_hw_sectors(struct request_queue *, unsigned int);
1246
extern void blk_queue_chunk_sectors(struct request_queue *, unsigned int);
1247
extern void blk_queue_max_segments(struct request_queue *, unsigned short);
1248 1249
extern void blk_queue_max_discard_segments(struct request_queue *,
		unsigned short);
1250
extern void blk_queue_max_segment_size(struct request_queue *, unsigned int);
1251 1252
extern void blk_queue_max_discard_sectors(struct request_queue *q,
		unsigned int max_discard_sectors);
1253 1254
extern void blk_queue_max_write_same_sectors(struct request_queue *q,
		unsigned int max_write_same_sectors);
1255 1256
extern void blk_queue_max_write_zeroes_sectors(struct request_queue *q,
		unsigned int max_write_same_sectors);
1257
extern void blk_queue_logical_block_size(struct request_queue *, unsigned int);
1258
extern void blk_queue_physical_block_size(struct request_queue *, unsigned int);
1259 1260
extern void blk_queue_alignment_offset(struct request_queue *q,
				       unsigned int alignment);
1261
extern void blk_limits_io_min(struct queue_limits *limits, unsigned int min);
1262
extern void blk_queue_io_min(struct request_queue *q, unsigned int min);
1263
extern void blk_limits_io_opt(struct queue_limits *limits, unsigned int opt);
1264
extern void blk_queue_io_opt(struct request_queue *q, unsigned int opt);
1265
extern void blk_set_queue_depth(struct request_queue *q, unsigned int depth);
1266
extern void blk_set_default_limits(struct queue_limits *lim);
1267
extern void blk_set_stacking_limits(struct queue_limits *lim);
1268 1269
extern int blk_stack_limits(struct queue_limits *t, struct queue_limits *b,
			    sector_t offset);
1270 1271
extern int bdev_stack_limits(struct queue_limits *t, struct block_device *bdev,
			    sector_t offset);
1272 1273
extern void disk_stack_limits(struct gendisk *disk, struct block_device *bdev,
			      sector_t offset);
1274
extern void blk_queue_stack_limits(struct request_queue *t, struct request_queue *b);
1275
extern void blk_queue_dma_pad(struct request_queue *, unsigned int);
1276
extern void blk_queue_update_dma_pad(struct request_queue *, unsigned int);
1277 1278 1279
extern int blk_queue_dma_drain(struct request_queue *q,
			       dma_drain_needed_fn *dma_drain_needed,
			       void *buf, unsigned int size);
1280
extern void blk_queue_lld_busy(struct request_queue *q, lld_busy_fn *fn);
1281
extern void blk_queue_segment_boundary(struct request_queue *, unsigned long);
1282
extern void blk_queue_virt_boundary(struct request_queue *, unsigned long);
1283
extern void blk_queue_prep_rq(struct request_queue *, prep_rq_fn *pfn);
1284
extern void blk_queue_unprep_rq(struct request_queue *, unprep_rq_fn *ufn);
1285
extern void blk_queue_dma_alignment(struct request_queue *, int);
1286
extern void blk_queue_update_dma_alignment(struct request_queue *, int);
1287
extern void blk_queue_softirq_done(struct request_queue *, softirq_done_fn *);
J
Jens Axboe 已提交
1288 1289
extern void blk_queue_rq_timed_out(struct request_queue *, rq_timed_out_fn *);
extern void blk_queue_rq_timeout(struct request_queue *, unsigned int);
1290
extern void blk_queue_flush_queueable(struct request_queue *q, bool queueable);
1291
extern void blk_queue_write_cache(struct request_queue *q, bool enabled, bool fua);
1292 1293
extern void blk_queue_rq_hang_threshold(struct request_queue *q,
					unsigned int threshold);
L
Linus Torvalds 已提交
1294

1295 1296 1297 1298 1299 1300 1301 1302 1303
/*
 * Number of physical segments as sent to the device.
 *
 * Normally this is the number of discontiguous data segments sent by the
 * submitter.  But for data-less command like discard we might have no
 * actual data segments submitted, but the driver might have to add it's
 * own special payload.  In that case we still return 1 here so that this
 * special payload will be mapped.
 */
1304 1305 1306 1307 1308 1309 1310
static inline unsigned short blk_rq_nr_phys_segments(struct request *rq)
{
	if (rq->rq_flags & RQF_SPECIAL_PAYLOAD)
		return 1;
	return rq->nr_phys_segments;
}

1311 1312 1313 1314 1315 1316 1317 1318 1319
/*
 * Number of discard segments (or ranges) the driver needs to fill in.
 * Each discard bio merged into a request is counted as one segment.
 */
static inline unsigned short blk_rq_nr_discard_segments(struct request *rq)
{
	return max_t(unsigned short, rq->nr_phys_segments, 1);
}

1320
extern int blk_rq_map_sg(struct request_queue *, struct request *, struct scatterlist *);
L
Linus Torvalds 已提交
1321 1322 1323
extern void blk_dump_rq_flags(struct request *, char *);
extern long nr_blockdev_pages(void);

T
Tejun Heo 已提交
1324
bool __must_check blk_get_queue(struct request_queue *);
1325
struct request_queue *blk_alloc_queue(gfp_t);
1326 1327
struct request_queue *blk_alloc_queue_node(gfp_t gfp_mask, int node_id,
					   spinlock_t *lock);
1328
extern void blk_put_queue(struct request_queue *);
1329
extern void blk_set_queue_dying(struct request_queue *);
L
Linus Torvalds 已提交
1330

L
Lin Ming 已提交
1331 1332 1333
/*
 * block layer runtime pm functions
 */
1334
#ifdef CONFIG_PM
L
Lin Ming 已提交
1335 1336 1337 1338 1339
extern void blk_pm_runtime_init(struct request_queue *q, struct device *dev);
extern int blk_pre_runtime_suspend(struct request_queue *q);
extern void blk_post_runtime_suspend(struct request_queue *q, int err);
extern void blk_pre_runtime_resume(struct request_queue *q);
extern void blk_post_runtime_resume(struct request_queue *q, int err);
1340
extern void blk_set_runtime_active(struct request_queue *q);
L
Lin Ming 已提交
1341 1342 1343 1344 1345 1346 1347 1348 1349 1350
#else
static inline void blk_pm_runtime_init(struct request_queue *q,
	struct device *dev) {}
static inline int blk_pre_runtime_suspend(struct request_queue *q)
{
	return -ENOSYS;
}
static inline void blk_post_runtime_suspend(struct request_queue *q, int err) {}
static inline void blk_pre_runtime_resume(struct request_queue *q) {}
static inline void blk_post_runtime_resume(struct request_queue *q, int err) {}
1351
static inline void blk_set_runtime_active(struct request_queue *q) {}
L
Lin Ming 已提交
1352 1353
#endif

1354
/*
S
Suresh Jayaraman 已提交
1355 1356 1357 1358 1359 1360 1361 1362 1363 1364
 * blk_plug permits building a queue of related requests by holding the I/O
 * fragments for a short period. This allows merging of sequential requests
 * into single larger request. As the requests are moved from a per-task list to
 * the device's request_queue in a batch, this results in improved scalability
 * as the lock contention for request_queue lock is reduced.
 *
 * It is ok not to disable preemption when adding the request to the plug list
 * or when attempting a merge, because blk_schedule_flush_list() will only flush
 * the plug list when the task sleeps by itself. For details, please see
 * schedule() where blk_schedule_flush_plug() is called.
1365
 */
1366
struct blk_plug {
S
Suresh Jayaraman 已提交
1367
	struct list_head list; /* requests */
1368
	struct list_head mq_list; /* blk-mq requests */
S
Suresh Jayaraman 已提交
1369
	struct list_head cb_list; /* md requires an unplug callback */
1370
	bool multiple_queues;
1371
};
1372
#define BLK_MAX_REQUEST_COUNT 16
1373
#define BLK_PLUG_FLUSH_SIZE (128 * 1024)
1374

1375
struct blk_plug_cb;
1376
typedef void (*blk_plug_cb_fn)(struct blk_plug_cb *, bool);
1377 1378
struct blk_plug_cb {
	struct list_head list;
1379 1380
	blk_plug_cb_fn callback;
	void *data;
1381
};
1382 1383
extern struct blk_plug_cb *blk_check_plugged(blk_plug_cb_fn unplug,
					     void *data, int size);
1384 1385
extern void blk_start_plug(struct blk_plug *);
extern void blk_finish_plug(struct blk_plug *);
1386
extern void blk_flush_plug_list(struct blk_plug *, bool);
1387 1388 1389 1390 1391

static inline void blk_flush_plug(struct task_struct *tsk)
{
	struct blk_plug *plug = tsk->plug;

1392 1393 1394 1395 1396 1397 1398 1399
	if (plug)
		blk_flush_plug_list(plug, false);
}

static inline void blk_schedule_flush_plug(struct task_struct *tsk)
{
	struct blk_plug *plug = tsk->plug;

1400
	if (plug)
1401
		blk_flush_plug_list(plug, true);
1402 1403 1404 1405 1406 1407
}

static inline bool blk_needs_flush_plug(struct task_struct *tsk)
{
	struct blk_plug *plug = tsk->plug;

1408 1409 1410 1411
	return plug &&
		(!list_empty(&plug->list) ||
		 !list_empty(&plug->mq_list) ||
		 !list_empty(&plug->cb_list));
1412 1413
}

L
Linus Torvalds 已提交
1414 1415 1416
/*
 * tag stuff
 */
1417 1418 1419
extern int blk_queue_start_tag(struct request_queue *, struct request *);
extern struct request *blk_queue_find_tag(struct request_queue *, int);
extern void blk_queue_end_tag(struct request_queue *, struct request *);
1420
extern int blk_queue_init_tags(struct request_queue *, int, struct blk_queue_tag *, int);
1421 1422
extern void blk_queue_free_tags(struct request_queue *);
extern int blk_queue_resize_tags(struct request_queue *, int);
1423
extern struct blk_queue_tag *blk_init_tags(int, int);
1424
extern void blk_free_tags(struct blk_queue_tag *);
L
Linus Torvalds 已提交
1425

1426 1427 1428 1429 1430 1431 1432
static inline struct request *blk_map_queue_find_tag(struct blk_queue_tag *bqt,
						int tag)
{
	if (unlikely(bqt == NULL || tag >= bqt->real_max_depth))
		return NULL;
	return bqt->tag_index[tag];
}
1433

1434 1435 1436
extern int blkdev_issue_flush(struct block_device *, gfp_t, sector_t *);
extern int blkdev_issue_write_same(struct block_device *bdev, sector_t sector,
		sector_t nr_sects, gfp_t gfp_mask, struct page *page);
1437 1438

#define BLKDEV_DISCARD_SECURE	(1 << 0)	/* issue a secure erase */
1439

1440 1441
extern int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
		sector_t nr_sects, gfp_t gfp_mask, unsigned long flags);
1442
extern int __blkdev_issue_discard(struct block_device *bdev, sector_t sector,
1443
		sector_t nr_sects, gfp_t gfp_mask, int flags,
1444
		struct bio **biop);
1445 1446

#define BLKDEV_ZERO_NOUNMAP	(1 << 0)  /* do not free blocks */
1447
#define BLKDEV_ZERO_NOFALLBACK	(1 << 1)  /* don't write explicit zeroes */
1448

1449 1450
extern int __blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
		sector_t nr_sects, gfp_t gfp_mask, struct bio **biop,
1451
		unsigned flags);
1452
extern int blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
1453 1454
		sector_t nr_sects, gfp_t gfp_mask, unsigned flags);

1455 1456
static inline int sb_issue_discard(struct super_block *sb, sector_t block,
		sector_t nr_blocks, gfp_t gfp_mask, unsigned long flags)
D
David Woodhouse 已提交
1457
{
1458 1459 1460 1461 1462
	return blkdev_issue_discard(sb->s_bdev,
				    block << (sb->s_blocksize_bits -
					      SECTOR_SHIFT),
				    nr_blocks << (sb->s_blocksize_bits -
						  SECTOR_SHIFT),
1463
				    gfp_mask, flags);
D
David Woodhouse 已提交
1464
}
1465
static inline int sb_issue_zeroout(struct super_block *sb, sector_t block,
1466
		sector_t nr_blocks, gfp_t gfp_mask)
1467 1468
{
	return blkdev_issue_zeroout(sb->s_bdev,
1469 1470 1471 1472
				    block << (sb->s_blocksize_bits -
					      SECTOR_SHIFT),
				    nr_blocks << (sb->s_blocksize_bits -
						  SECTOR_SHIFT),
1473
				    gfp_mask, 0);
1474
}
L
Linus Torvalds 已提交
1475

1476
extern int blk_verify_command(unsigned char *cmd, fmode_t mode);
1477

1478 1479 1480
enum blk_default_limits {
	BLK_MAX_SEGMENTS	= 128,
	BLK_SAFE_MAX_SECTORS	= 255,
1481
	BLK_DEF_MAX_SECTORS	= 2560,
1482 1483 1484
	BLK_MAX_SEGMENT_SIZE	= 65536,
	BLK_SEG_BOUNDARY_MASK	= 0xFFFFFFFFUL,
};
1485

1486 1487
static inline unsigned long queue_segment_boundary(struct request_queue *q)
{
1488
	return q->limits.seg_boundary_mask;
1489 1490
}

1491 1492 1493 1494 1495
static inline unsigned long queue_virt_boundary(struct request_queue *q)
{
	return q->limits.virt_boundary_mask;
}

1496 1497
static inline unsigned int queue_max_sectors(struct request_queue *q)
{
1498
	return q->limits.max_sectors;
1499 1500 1501 1502
}

static inline unsigned int queue_max_hw_sectors(struct request_queue *q)
{
1503
	return q->limits.max_hw_sectors;
1504 1505
}

1506
static inline unsigned short queue_max_segments(struct request_queue *q)
1507
{
1508
	return q->limits.max_segments;
1509 1510
}

1511 1512 1513 1514 1515
static inline unsigned short queue_max_discard_segments(struct request_queue *q)
{
	return q->limits.max_discard_segments;
}

1516 1517
static inline unsigned int queue_max_segment_size(struct request_queue *q)
{
1518
	return q->limits.max_segment_size;
1519 1520
}

1521
static inline unsigned queue_logical_block_size(struct request_queue *q)
L
Linus Torvalds 已提交
1522 1523 1524
{
	int retval = 512;

1525 1526
	if (q && q->limits.logical_block_size)
		retval = q->limits.logical_block_size;
L
Linus Torvalds 已提交
1527 1528 1529 1530

	return retval;
}

1531
static inline unsigned int bdev_logical_block_size(struct block_device *bdev)
L
Linus Torvalds 已提交
1532
{
1533
	return queue_logical_block_size(bdev_get_queue(bdev));
L
Linus Torvalds 已提交
1534 1535
}

1536 1537 1538 1539 1540
static inline unsigned int queue_physical_block_size(struct request_queue *q)
{
	return q->limits.physical_block_size;
}

1541
static inline unsigned int bdev_physical_block_size(struct block_device *bdev)
M
Martin K. Petersen 已提交
1542 1543 1544 1545
{
	return queue_physical_block_size(bdev_get_queue(bdev));
}

1546 1547 1548 1549 1550
static inline unsigned int queue_io_min(struct request_queue *q)
{
	return q->limits.io_min;
}

M
Martin K. Petersen 已提交
1551 1552 1553 1554 1555
static inline int bdev_io_min(struct block_device *bdev)
{
	return queue_io_min(bdev_get_queue(bdev));
}

1556 1557 1558 1559 1560
static inline unsigned int queue_io_opt(struct request_queue *q)
{
	return q->limits.io_opt;
}

M
Martin K. Petersen 已提交
1561 1562 1563 1564 1565
static inline int bdev_io_opt(struct block_device *bdev)
{
	return queue_io_opt(bdev_get_queue(bdev));
}

1566 1567
static inline int queue_alignment_offset(struct request_queue *q)
{
M
Martin K. Petersen 已提交
1568
	if (q->limits.misaligned)
1569 1570
		return -1;

M
Martin K. Petersen 已提交
1571
	return q->limits.alignment_offset;
1572 1573
}

1574
static inline int queue_limit_alignment_offset(struct queue_limits *lim, sector_t sector)
1575 1576
{
	unsigned int granularity = max(lim->physical_block_size, lim->io_min);
1577 1578
	unsigned int alignment = sector_div(sector, granularity >> SECTOR_SHIFT)
		<< SECTOR_SHIFT;
1579

1580
	return (granularity + lim->alignment_offset - alignment) % granularity;
1581 1582
}

M
Martin K. Petersen 已提交
1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595
static inline int bdev_alignment_offset(struct block_device *bdev)
{
	struct request_queue *q = bdev_get_queue(bdev);

	if (q->limits.misaligned)
		return -1;

	if (bdev != bdev->bd_contains)
		return bdev->bd_part->alignment_offset;

	return q->limits.alignment_offset;
}

1596 1597 1598 1599 1600 1601 1602 1603
static inline int queue_discard_alignment(struct request_queue *q)
{
	if (q->limits.discard_misaligned)
		return -1;

	return q->limits.discard_alignment;
}

1604
static inline int queue_limit_discard_alignment(struct queue_limits *lim, sector_t sector)
1605
{
1606
	unsigned int alignment, granularity, offset;
1607

1608 1609 1610
	if (!lim->max_discard_sectors)
		return 0;

1611
	/* Why are these in bytes, not sectors? */
1612 1613
	alignment = lim->discard_alignment >> SECTOR_SHIFT;
	granularity = lim->discard_granularity >> SECTOR_SHIFT;
1614 1615 1616 1617 1618 1619 1620 1621 1622 1623
	if (!granularity)
		return 0;

	/* Offset of the partition start in 'granularity' sectors */
	offset = sector_div(sector, granularity);

	/* And why do we do this modulus *again* in blkdev_issue_discard()? */
	offset = (granularity + alignment - offset) % granularity;

	/* Turn it back into bytes, gaah */
1624
	return offset << SECTOR_SHIFT;
1625 1626
}

1627 1628 1629 1630 1631 1632 1633 1634 1635 1636
static inline int bdev_discard_alignment(struct block_device *bdev)
{
	struct request_queue *q = bdev_get_queue(bdev);

	if (bdev != bdev->bd_contains)
		return bdev->bd_part->discard_alignment;

	return q->limits.discard_alignment;
}

1637 1638 1639 1640 1641 1642 1643 1644 1645 1646
static inline unsigned int bdev_write_same(struct block_device *bdev)
{
	struct request_queue *q = bdev_get_queue(bdev);

	if (q)
		return q->limits.max_write_same_sectors;

	return 0;
}

1647 1648 1649 1650 1651 1652 1653 1654 1655 1656
static inline unsigned int bdev_write_zeroes_sectors(struct block_device *bdev)
{
	struct request_queue *q = bdev_get_queue(bdev);

	if (q)
		return q->limits.max_write_zeroes_sectors;

	return 0;
}

D
Damien Le Moal 已提交
1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676
static inline enum blk_zoned_model bdev_zoned_model(struct block_device *bdev)
{
	struct request_queue *q = bdev_get_queue(bdev);

	if (q)
		return blk_queue_zoned_model(q);

	return BLK_ZONED_NONE;
}

static inline bool bdev_is_zoned(struct block_device *bdev)
{
	struct request_queue *q = bdev_get_queue(bdev);

	if (q)
		return blk_queue_is_zoned(q);

	return false;
}

1677
static inline unsigned int bdev_zone_sectors(struct block_device *bdev)
1678 1679 1680 1681
{
	struct request_queue *q = bdev_get_queue(bdev);

	if (q)
1682
		return blk_queue_zone_sectors(q);
1683 1684
	return 0;
}
1685

1686
static inline int queue_dma_alignment(struct request_queue *q)
L
Linus Torvalds 已提交
1687
{
1688
	return q ? q->dma_alignment : 511;
L
Linus Torvalds 已提交
1689 1690
}

1691
static inline int blk_rq_aligned(struct request_queue *q, unsigned long addr,
1692 1693 1694
				 unsigned int len)
{
	unsigned int alignment = queue_dma_alignment(q) | q->dma_pad_mask;
1695
	return !(addr & alignment) && !(len & alignment);
1696 1697
}

L
Linus Torvalds 已提交
1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708
/* assumes size > 256 */
static inline unsigned int blksize_bits(unsigned int size)
{
	unsigned int bits = 8;
	do {
		bits++;
		size >>= 1;
	} while (size > 256);
	return bits;
}

1709
static inline unsigned int block_size(struct block_device *bdev)
L
Linus Torvalds 已提交
1710 1711 1712 1713
{
	return bdev->bd_block_size;
}

1714 1715
static inline bool queue_flush_queueable(struct request_queue *q)
{
J
Jens Axboe 已提交
1716
	return !test_bit(QUEUE_FLAG_FLUSH_NQ, &q->queue_flags);
1717 1718
}

L
Linus Torvalds 已提交
1719 1720 1721 1722 1723 1724
typedef struct {struct page *v;} Sector;

unsigned char *read_dev_sector(struct block_device *, sector_t, Sector *);

static inline void put_dev_sector(Sector p)
{
1725
	put_page(p.v);
L
Linus Torvalds 已提交
1726 1727
}

1728 1729 1730 1731 1732 1733 1734
static inline bool __bvec_gap_to_prev(struct request_queue *q,
				struct bio_vec *bprv, unsigned int offset)
{
	return offset ||
		((bprv->bv_offset + bprv->bv_len) & queue_virt_boundary(q));
}

1735 1736 1737 1738 1739 1740 1741 1742 1743
/*
 * 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;
1744
	return __bvec_gap_to_prev(q, bprv, offset);
1745 1746
}

M
Ming Lei 已提交
1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765
/*
 * Check if the two bvecs from two bios can be merged to one segment.
 * If yes, no need to check gap between the two bios since the 1st bio
 * and the 1st bvec in the 2nd bio can be handled in one segment.
 */
static inline bool bios_segs_mergeable(struct request_queue *q,
		struct bio *prev, struct bio_vec *prev_last_bv,
		struct bio_vec *next_first_bv)
{
	if (!BIOVEC_PHYS_MERGEABLE(prev_last_bv, next_first_bv))
		return false;
	if (!BIOVEC_SEG_BOUNDARY(q, prev_last_bv, next_first_bv))
		return false;
	if (prev->bi_seg_back_size + next_first_bv->bv_len >
			queue_max_segment_size(q))
		return false;
	return true;
}

1766 1767 1768 1769
static inline bool bio_will_gap(struct request_queue *q,
				struct request *prev_rq,
				struct bio *prev,
				struct bio *next)
1770
{
1771 1772 1773
	if (bio_has_data(prev) && queue_virt_boundary(q)) {
		struct bio_vec pb, nb;

1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795
		/*
		 * don't merge if the 1st bio starts with non-zero
		 * offset, otherwise it is quite difficult to respect
		 * sg gap limit. We work hard to merge a huge number of small
		 * single bios in case of mkfs.
		 */
		if (prev_rq)
			bio_get_first_bvec(prev_rq->bio, &pb);
		else
			bio_get_first_bvec(prev, &pb);
		if (pb.bv_offset)
			return true;

		/*
		 * We don't need to worry about the situation that the
		 * merged segment ends in unaligned virt boundary:
		 *
		 * - if 'pb' ends aligned, the merged segment ends aligned
		 * - if 'pb' ends unaligned, the next bio must include
		 *   one single bvec of 'nb', otherwise the 'nb' can't
		 *   merge with 'pb'
		 */
1796 1797
		bio_get_last_bvec(prev, &pb);
		bio_get_first_bvec(next, &nb);
1798

M
Ming Lei 已提交
1799 1800
		if (!bios_segs_mergeable(q, prev, &pb, &nb))
			return __bvec_gap_to_prev(q, &pb, nb.bv_offset);
1801 1802 1803
	}

	return false;
1804 1805 1806 1807
}

static inline bool req_gap_back_merge(struct request *req, struct bio *bio)
{
1808
	return bio_will_gap(req->q, req, req->biotail, bio);
1809 1810 1811 1812
}

static inline bool req_gap_front_merge(struct request *req, struct bio *bio)
{
1813
	return bio_will_gap(req->q, NULL, bio, req->bio);
1814 1815
}

1816
int kblockd_schedule_work(struct work_struct *work);
1817
int kblockd_schedule_work_on(int cpu, struct work_struct *work);
1818
int kblockd_mod_delayed_work_on(int cpu, struct delayed_work *dwork, unsigned long delay);
L
Linus Torvalds 已提交
1819 1820 1821 1822 1823 1824

#define MODULE_ALIAS_BLOCKDEV(major,minor) \
	MODULE_ALIAS("block-major-" __stringify(major) "-" __stringify(minor))
#define MODULE_ALIAS_BLOCKDEV_MAJOR(major) \
	MODULE_ALIAS("block-major-" __stringify(major) "-*")

1825 1826
#if defined(CONFIG_BLK_DEV_INTEGRITY)

1827 1828 1829
enum blk_integrity_flags {
	BLK_INTEGRITY_VERIFY		= 1 << 0,
	BLK_INTEGRITY_GENERATE		= 1 << 1,
1830
	BLK_INTEGRITY_DEVICE_CAPABLE	= 1 << 2,
1831
	BLK_INTEGRITY_IP_CHECKSUM	= 1 << 3,
1832
};
1833

1834
struct blk_integrity_iter {
1835 1836
	void			*prot_buf;
	void			*data_buf;
1837
	sector_t		seed;
1838
	unsigned int		data_size;
1839
	unsigned short		interval;
1840 1841 1842
	const char		*disk_name;
};

1843
typedef blk_status_t (integrity_processing_fn) (struct blk_integrity_iter *);
1844

1845 1846 1847 1848 1849
struct blk_integrity_profile {
	integrity_processing_fn		*generate_fn;
	integrity_processing_fn		*verify_fn;
	const char			*name;
};
1850

1851
extern void blk_integrity_register(struct gendisk *, struct blk_integrity *);
1852
extern void blk_integrity_unregister(struct gendisk *);
1853
extern int blk_integrity_compare(struct gendisk *, struct gendisk *);
1854 1855 1856
extern int blk_rq_map_integrity_sg(struct request_queue *, struct bio *,
				   struct scatterlist *);
extern int blk_rq_count_integrity_sg(struct request_queue *, struct bio *);
1857 1858 1859 1860
extern bool blk_integrity_merge_rq(struct request_queue *, struct request *,
				   struct request *);
extern bool blk_integrity_merge_bio(struct request_queue *, struct request *,
				    struct bio *);
1861

1862
static inline struct blk_integrity *blk_get_integrity(struct gendisk *disk)
1863
{
1864
	struct blk_integrity *bi = &disk->queue->integrity;
1865 1866 1867 1868 1869

	if (!bi->profile)
		return NULL;

	return bi;
1870 1871
}

1872 1873
static inline
struct blk_integrity *bdev_get_integrity(struct block_device *bdev)
1874
{
1875
	return blk_get_integrity(bdev->bd_disk);
1876 1877
}

1878
static inline bool blk_integrity_rq(struct request *rq)
1879
{
1880
	return rq->cmd_flags & REQ_INTEGRITY;
1881 1882
}

1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894
static inline void blk_queue_max_integrity_segments(struct request_queue *q,
						    unsigned int segs)
{
	q->limits.max_integrity_segments = segs;
}

static inline unsigned short
queue_max_integrity_segments(struct request_queue *q)
{
	return q->limits.max_integrity_segments;
}

1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914
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);
}

1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936
/**
 * bio_integrity_intervals - Return number of integrity intervals for a bio
 * @bi:		blk_integrity profile for device
 * @sectors:	Size of the bio in 512-byte sectors
 *
 * Description: The block layer calculates everything in 512 byte
 * sectors but integrity metadata is done in terms of the data integrity
 * interval size of the storage device.  Convert the block layer sectors
 * to the appropriate number of integrity intervals.
 */
static inline unsigned int bio_integrity_intervals(struct blk_integrity *bi,
						   unsigned int sectors)
{
	return sectors >> (bi->interval_exp - 9);
}

static inline unsigned int bio_integrity_bytes(struct blk_integrity *bi,
					       unsigned int sectors)
{
	return bio_integrity_intervals(bi, sectors) * bi->tuple_size;
}

1937 1938
#else /* CONFIG_BLK_DEV_INTEGRITY */

1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960
struct bio;
struct block_device;
struct gendisk;
struct blk_integrity;

static inline int blk_integrity_rq(struct request *rq)
{
	return 0;
}
static inline int blk_rq_count_integrity_sg(struct request_queue *q,
					    struct bio *b)
{
	return 0;
}
static inline int blk_rq_map_integrity_sg(struct request_queue *q,
					  struct bio *b,
					  struct scatterlist *s)
{
	return 0;
}
static inline struct blk_integrity *bdev_get_integrity(struct block_device *b)
{
1961
	return NULL;
1962 1963 1964 1965 1966 1967 1968 1969 1970
}
static inline struct blk_integrity *blk_get_integrity(struct gendisk *disk)
{
	return NULL;
}
static inline int blk_integrity_compare(struct gendisk *a, struct gendisk *b)
{
	return 0;
}
1971
static inline void blk_integrity_register(struct gendisk *d,
1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985
					 struct blk_integrity *b)
{
}
static inline void blk_integrity_unregister(struct gendisk *d)
{
}
static inline void blk_queue_max_integrity_segments(struct request_queue *q,
						    unsigned int segs)
{
}
static inline unsigned short queue_max_integrity_segments(struct request_queue *q)
{
	return 0;
}
1986 1987 1988
static inline bool blk_integrity_merge_rq(struct request_queue *rq,
					  struct request *r1,
					  struct request *r2)
1989
{
1990
	return true;
1991
}
1992 1993 1994
static inline bool blk_integrity_merge_bio(struct request_queue *rq,
					   struct request *r,
					   struct bio *b)
1995
{
1996
	return true;
1997
}
1998

1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
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;
}
2009

2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021
static inline unsigned int bio_integrity_intervals(struct blk_integrity *bi,
						   unsigned int sectors)
{
	return 0;
}

static inline unsigned int bio_integrity_bytes(struct blk_integrity *bi,
					       unsigned int sectors)
{
	return 0;
}

2022 2023
#endif /* CONFIG_BLK_DEV_INTEGRITY */

2024
struct block_device_operations {
A
Al Viro 已提交
2025
	int (*open) (struct block_device *, fmode_t);
2026
	void (*release) (struct gendisk *, fmode_t);
2027
	int (*rw_page)(struct block_device *, sector_t, struct page *, unsigned int);
A
Al Viro 已提交
2028 2029
	int (*ioctl) (struct block_device *, fmode_t, unsigned, unsigned long);
	int (*compat_ioctl) (struct block_device *, fmode_t, unsigned, unsigned long);
2030 2031 2032
	unsigned int (*check_events) (struct gendisk *disk,
				      unsigned int clearing);
	/* ->media_changed() is DEPRECATED, use ->check_events() instead */
2033
	int (*media_changed) (struct gendisk *);
2034
	void (*unlock_native_capacity) (struct gendisk *);
2035 2036
	int (*revalidate_disk) (struct gendisk *);
	int (*getgeo)(struct block_device *, struct hd_geometry *);
2037 2038
	/* this callback is with swap_lock and sometimes page table lock held */
	void (*swap_slot_free_notify) (struct block_device *, unsigned long);
2039
	struct module *owner;
2040
	const struct pr_ops *pr_ops;
2041 2042
};

2043 2044
extern int __blkdev_driver_ioctl(struct block_device *, fmode_t, unsigned int,
				 unsigned long);
2045 2046 2047
extern int bdev_read_page(struct block_device *, sector_t, struct page *);
extern int bdev_write_page(struct block_device *, sector_t, struct page *,
						struct writeback_control *);
2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101

#ifdef CONFIG_BLK_DEV_ZONED
bool blk_req_needs_zone_write_lock(struct request *rq);
void __blk_req_zone_write_lock(struct request *rq);
void __blk_req_zone_write_unlock(struct request *rq);

static inline void blk_req_zone_write_lock(struct request *rq)
{
	if (blk_req_needs_zone_write_lock(rq))
		__blk_req_zone_write_lock(rq);
}

static inline void blk_req_zone_write_unlock(struct request *rq)
{
	if (rq->rq_flags & RQF_ZONE_WRITE_LOCKED)
		__blk_req_zone_write_unlock(rq);
}

static inline bool blk_req_zone_is_write_locked(struct request *rq)
{
	return rq->q->seq_zones_wlock &&
		test_bit(blk_rq_zone_no(rq), rq->q->seq_zones_wlock);
}

static inline bool blk_req_can_dispatch_to_zone(struct request *rq)
{
	if (!blk_req_needs_zone_write_lock(rq))
		return true;
	return !blk_req_zone_is_write_locked(rq);
}
#else
static inline bool blk_req_needs_zone_write_lock(struct request *rq)
{
	return false;
}

static inline void blk_req_zone_write_lock(struct request *rq)
{
}

static inline void blk_req_zone_write_unlock(struct request *rq)
{
}
static inline bool blk_req_zone_is_write_locked(struct request *rq)
{
	return false;
}

static inline bool blk_req_can_dispatch_to_zone(struct request *rq)
{
	return true;
}
#endif /* CONFIG_BLK_DEV_ZONED */

2102
#else /* CONFIG_BLOCK */
2103 2104 2105

struct block_device;

2106 2107 2108 2109 2110 2111 2112 2113 2114 2115
/*
 * stubs for when the block layer is configured out
 */
#define buffer_heads_over_limit 0

static inline long nr_blockdev_pages(void)
{
	return 0;
}

2116 2117 2118 2119
struct blk_plug {
};

static inline void blk_start_plug(struct blk_plug *plug)
2120 2121 2122
{
}

2123
static inline void blk_finish_plug(struct blk_plug *plug)
2124 2125 2126
{
}

2127
static inline void blk_flush_plug(struct task_struct *task)
2128 2129 2130
{
}

2131 2132 2133 2134 2135
static inline void blk_schedule_flush_plug(struct task_struct *task)
{
}


2136 2137 2138 2139 2140
static inline bool blk_needs_flush_plug(struct task_struct *tsk)
{
	return false;
}

2141 2142 2143 2144 2145 2146
static inline int blkdev_issue_flush(struct block_device *bdev, gfp_t gfp_mask,
				     sector_t *error_sector)
{
	return 0;
}

2147 2148
#endif /* CONFIG_BLOCK */

L
Linus Torvalds 已提交
2149
#endif