blkdev.h 57.6 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3
#ifndef _LINUX_BLKDEV_H
#define _LINUX_BLKDEV_H

4
#include <linux/sched.h>
5
#include <linux/sched/clock.h>
6

J
Jens Axboe 已提交
7 8
#ifdef CONFIG_BLOCK

L
Linus Torvalds 已提交
9 10 11
#include <linux/major.h>
#include <linux/genhd.h>
#include <linux/list.h>
12
#include <linux/llist.h>
L
Linus Torvalds 已提交
13 14 15
#include <linux/timer.h>
#include <linux/workqueue.h>
#include <linux/pagemap.h>
16
#include <linux/backing-dev-defs.h>
L
Linus Torvalds 已提交
17 18
#include <linux/wait.h>
#include <linux/mempool.h>
D
Dan Williams 已提交
19
#include <linux/pfn.h>
L
Linus Torvalds 已提交
20 21
#include <linux/bio.h>
#include <linux/stringify.h>
22
#include <linux/gfp.h>
23
#include <linux/bsg.h>
24
#include <linux/smp.h>
T
Tejun Heo 已提交
25
#include <linux/rcupdate.h>
26
#include <linux/percpu-refcount.h>
C
Christoph Hellwig 已提交
27
#include <linux/scatterlist.h>
28
#include <linux/blkzoned.h>
L
Linus Torvalds 已提交
29

30
struct module;
31 32
struct scsi_ioctl_command;

L
Linus Torvalds 已提交
33 34
struct request_queue;
struct elevator_queue;
35
struct blk_trace;
36 37
struct request;
struct sg_io_hdr;
M
Mike Christie 已提交
38
struct bsg_job;
T
Tejun Heo 已提交
39
struct blkcg_gq;
40
struct blk_flush_queue;
41
struct pr_ops;
J
Jens Axboe 已提交
42
struct rq_wb;
43 44
struct blk_queue_stats;
struct blk_stat_callback;
L
Linus Torvalds 已提交
45 46 47 48

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

T
Tejun Heo 已提交
49 50 51 52 53 54
/*
 * Maximum number of blkcg policies allowed to be registered concurrently.
 * Defined here to simplify include dependency.
 */
#define BLKCG_MAX_POLS		2

55
typedef void (rq_end_io_fn)(struct request *, int);
L
Linus Torvalds 已提交
56

57 58 59
#define BLK_RL_SYNCFULL		(1U << 0)
#define BLK_RL_ASYNCFULL	(1U << 1)

L
Linus Torvalds 已提交
60
struct request_list {
61
	struct request_queue	*q;	/* the queue this rl belongs to */
62 63 64
#ifdef CONFIG_BLK_CGROUP
	struct blkcg_gq		*blkg;	/* blkg this request pool belongs to */
#endif
65 66 67 68
	/*
	 * count[], starved[], and wait[] are indexed by
	 * BLK_RW_SYNC/BLK_RW_ASYNC
	 */
69 70 71 72
	int			count[2];
	int			starved[2];
	mempool_t		*rq_pool;
	wait_queue_head_t	wait[2];
73
	unsigned int		flags;
L
Linus Torvalds 已提交
74 75
};

76 77 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
/*
 * 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))
115 116
/* IO stats tracking on */
#define RQF_STATS		((__force req_flags_t)(1 << 17))
117 118 119
/* Look at ->special_vec for the actual data payload instead of the
   bio chain. */
#define RQF_SPECIAL_PAYLOAD	((__force req_flags_t)(1 << 18))
120 121 122

/* flags that prevent us from merging requests: */
#define RQF_NOMERGE_FLAGS \
123
	(RQF_STARTED | RQF_SOFTBARRIER | RQF_FLUSH_SEQ | RQF_SPECIAL_PAYLOAD)
124

L
Linus Torvalds 已提交
125
/*
126 127 128 129
 * 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
130 131
 */
struct request {
132
	struct list_head queuelist;
133 134
	union {
		struct call_single_data csd;
135
		u64 fifo_time;
136
	};
137

138
	struct request_queue *q;
139
	struct blk_mq_ctx *mq_ctx;
140

141
	int cpu;
142
	unsigned int cmd_flags;		/* op and common flags */
143
	req_flags_t rq_flags;
144 145 146

	int internal_tag;

J
Jens Axboe 已提交
147
	unsigned long atomic_flags;
L
Linus Torvalds 已提交
148

149 150
	/* the following two fields are internal, NEVER access directly */
	unsigned int __data_len;	/* total data len */
151
	int tag;
152
	sector_t __sector;		/* sector cursor */
L
Linus Torvalds 已提交
153 154 155 156

	struct bio *bio;
	struct bio *biotail;

157 158 159 160 161 162 163 164 165 166 167 168
	/*
	 * 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;
	};

169 170 171
	/*
	 * The rb_node is only used inside the io scheduler, requests
	 * are pruned when moved to the dispatch queue. So let the
172
	 * completion_data share space with the rb_node.
173 174 175
	 */
	union {
		struct rb_node rb_node;	/* sort/lookup */
176
		struct bio_vec special_vec;
177
		void *completion_data;
178
	};
179

180
	/*
181
	 * Three pointers are available for the IO schedulers, if they need
182 183
	 * more they have to dynamically allocate it.  Flush requests are
	 * never put on the IO scheduler. So let the flush fields share
184
	 * space with the elevator data.
185
	 */
186
	union {
187 188 189 190 191
		struct {
			struct io_cq		*icq;
			void			*priv[2];
		} elv;

192 193 194
		struct {
			unsigned int		seq;
			struct list_head	list;
195
			rq_end_io_fn		*saved_end_io;
196 197
		} flush;
	};
198

199
	struct gendisk *rq_disk;
200
	struct hd_struct *part;
L
Linus Torvalds 已提交
201
	unsigned long start_time;
202
	struct blk_issue_stat issue_stat;
203
#ifdef CONFIG_BLK_CGROUP
204
	struct request_list *rl;		/* rl this rq is alloced from */
205 206 207
	unsigned long long start_time_ns;
	unsigned long long io_start_time_ns;    /* when passed to hardware */
#endif
L
Linus Torvalds 已提交
208 209 210 211
	/* Number of scatter-gather DMA addr+len pairs after
	 * physical address coalescing is performed.
	 */
	unsigned short nr_phys_segments;
212 213 214
#if defined(CONFIG_BLK_DEV_INTEGRITY)
	unsigned short nr_integrity_segments;
#endif
L
Linus Torvalds 已提交
215

216 217
	unsigned short ioprio;

218 219
	unsigned int timeout;

T
Tejun Heo 已提交
220
	void *special;		/* opaque pointer available for LLD use */
L
Linus Torvalds 已提交
221

222 223
	int errors;

224
	unsigned int extra_len;	/* length of alignment and padding */
L
Linus Torvalds 已提交
225

J
Jens Axboe 已提交
226 227
	unsigned long deadline;
	struct list_head timeout_list;
L
Linus Torvalds 已提交
228 229

	/*
230
	 * completion callback.
L
Linus Torvalds 已提交
231 232 233
	 */
	rq_end_io_fn *end_io;
	void *end_io_data;
234 235 236

	/* for bidi */
	struct request *next_rq;
L
Linus Torvalds 已提交
237 238
};

239 240 241 242 243 244 245 246 247 248
static inline bool blk_rq_is_scsi(struct request *rq)
{
	return req_op(rq) == REQ_OP_SCSI_IN || req_op(rq) == REQ_OP_SCSI_OUT;
}

static inline bool blk_rq_is_private(struct request *rq)
{
	return req_op(rq) == REQ_OP_DRV_IN || req_op(rq) == REQ_OP_DRV_OUT;
}

249 250
static inline bool blk_rq_is_passthrough(struct request *rq)
{
251
	return blk_rq_is_scsi(rq) || blk_rq_is_private(rq);
252 253
}

254 255 256 257 258
static inline unsigned short req_get_ioprio(struct request *req)
{
	return req->ioprio;
}

L
Linus Torvalds 已提交
259 260
#include <linux/elevator.h>

261 262
struct blk_queue_ctx;

263
typedef void (request_fn_proc) (struct request_queue *q);
264
typedef blk_qc_t (make_request_fn) (struct request_queue *q, struct bio *bio);
265
typedef int (prep_rq_fn) (struct request_queue *, struct request *);
266
typedef void (unprep_rq_fn) (struct request_queue *, struct request *);
L
Linus Torvalds 已提交
267 268

struct bio_vec;
269
typedef void (softirq_done_fn)(struct request *);
270
typedef int (dma_drain_needed_fn)(struct request *);
271
typedef int (lld_busy_fn) (struct request_queue *q);
M
Mike Christie 已提交
272
typedef int (bsg_job_fn) (struct bsg_job *);
273 274
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
275

J
Jens Axboe 已提交
276 277 278 279 280 281 282 283
enum blk_eh_timer_return {
	BLK_EH_NOT_HANDLED,
	BLK_EH_HANDLED,
	BLK_EH_RESET_TIMER,
};

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

L
Linus Torvalds 已提交
284 285 286 287 288 289 290 291 292
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 */
293
	int real_max_depth;		/* what the array can hold */
L
Linus Torvalds 已提交
294
	atomic_t refcnt;		/* map can be shared */
295 296
	int alloc_policy;		/* tag allocation policy */
	int next_tag;			/* next tag */
L
Linus Torvalds 已提交
297
};
298 299
#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
300

301 302 303
#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
304 305 306 307 308 309 310 311 312
/*
 * 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 */
};

313 314 315
struct queue_limits {
	unsigned long		bounce_pfn;
	unsigned long		seg_boundary_mask;
316
	unsigned long		virt_boundary_mask;
317 318

	unsigned int		max_hw_sectors;
319
	unsigned int		max_dev_sectors;
320
	unsigned int		chunk_sectors;
321 322
	unsigned int		max_sectors;
	unsigned int		max_segment_size;
323 324 325 326
	unsigned int		physical_block_size;
	unsigned int		alignment_offset;
	unsigned int		io_min;
	unsigned int		io_opt;
327
	unsigned int		max_discard_sectors;
328
	unsigned int		max_hw_discard_sectors;
329
	unsigned int		max_write_same_sectors;
330
	unsigned int		max_write_zeroes_sectors;
331 332
	unsigned int		discard_granularity;
	unsigned int		discard_alignment;
333 334

	unsigned short		logical_block_size;
335
	unsigned short		max_segments;
336
	unsigned short		max_integrity_segments;
337
	unsigned short		max_discard_segments;
338

339
	unsigned char		misaligned;
340
	unsigned char		discard_misaligned;
341
	unsigned char		cluster;
342
	unsigned char		discard_zeroes_data;
343
	unsigned char		raid_partial_stripes_expensive;
D
Damien Le Moal 已提交
344
	enum blk_zoned_model	zoned;
345 346
};

347 348 349 350 351 352 353 354 355 356 357 358 359
#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
360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380
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;
}

381 382
#endif /* CONFIG_BLK_DEV_ZONED */

383
struct request_queue {
L
Linus Torvalds 已提交
384 385 386 387 388
	/*
	 * Together with queue_head for cacheline sharing
	 */
	struct list_head	queue_head;
	struct request		*last_merge;
J
Jens Axboe 已提交
389
	struct elevator_queue	*elevator;
390 391
	int			nr_rqs[2];	/* # allocated [a]sync rqs */
	int			nr_rqs_elvpriv;	/* # allocated rqs w/ elvpriv */
L
Linus Torvalds 已提交
392

393
	struct blk_queue_stats	*stats;
J
Jens Axboe 已提交
394 395
	struct rq_wb		*rq_wb;

L
Linus Torvalds 已提交
396
	/*
397 398 399 400
	 * 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
401
	 */
402
	struct request_list	root_rl;
L
Linus Torvalds 已提交
403 404 405 406

	request_fn_proc		*request_fn;
	make_request_fn		*make_request_fn;
	prep_rq_fn		*prep_rq_fn;
407
	unprep_rq_fn		*unprep_rq_fn;
408
	softirq_done_fn		*softirq_done_fn;
J
Jens Axboe 已提交
409
	rq_timed_out_fn		*rq_timed_out_fn;
410
	dma_drain_needed_fn	*dma_drain_needed;
411
	lld_busy_fn		*lld_busy_fn;
412 413
	init_rq_fn		*init_rq_fn;
	exit_rq_fn		*exit_rq_fn;
L
Linus Torvalds 已提交
414

J
Jens Axboe 已提交
415
	const struct blk_mq_ops	*mq_ops;
416 417 418 419

	unsigned int		*mq_map;

	/* sw queues */
420
	struct blk_mq_ctx __percpu	*queue_ctx;
421 422
	unsigned int		nr_queues;

423 424
	unsigned int		queue_depth;

425 426 427 428
	/* hw dispatch queues */
	struct blk_mq_hw_ctx	**queue_hw_ctx;
	unsigned int		nr_hw_queues;

429 430 431
	/*
	 * Dispatch queue sorting
	 */
J
Jens Axboe 已提交
432
	sector_t		end_sector;
433 434
	struct request		*boundary_rq;

L
Linus Torvalds 已提交
435
	/*
436
	 * Delayed queue handling
L
Linus Torvalds 已提交
437
	 */
438
	struct delayed_work	delay_work;
L
Linus Torvalds 已提交
439

440
	struct backing_dev_info	*backing_dev_info;
L
Linus Torvalds 已提交
441 442 443 444 445 446 447 448

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

	/*
449
	 * various queue flags, see QUEUE_* below
L
Linus Torvalds 已提交
450
	 */
451
	unsigned long		queue_flags;
L
Linus Torvalds 已提交
452

453 454 455 456 457 458
	/*
	 * ida allocated id for this queue.  Used to index queues from
	 * ioctx.
	 */
	int			id;

L
Linus Torvalds 已提交
459
	/*
460
	 * queue needs bounce pages for pages above this limit
L
Linus Torvalds 已提交
461
	 */
462
	gfp_t			bounce_gfp;
L
Linus Torvalds 已提交
463 464

	/*
已提交
465 466 467
	 * 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
468
	 */
已提交
469
	spinlock_t		__queue_lock;
L
Linus Torvalds 已提交
470 471 472 473 474 475 476
	spinlock_t		*queue_lock;

	/*
	 * queue kobject
	 */
	struct kobject kobj;

477 478 479 480 481
	/*
	 * mq queue kobject
	 */
	struct kobject mq_kobj;

482 483 484 485
#ifdef  CONFIG_BLK_DEV_INTEGRITY
	struct blk_integrity integrity;
#endif	/* CONFIG_BLK_DEV_INTEGRITY */

486
#ifdef CONFIG_PM
L
Lin Ming 已提交
487 488 489 490 491
	struct device		*dev;
	int			rpm_status;
	unsigned int		nr_pending;
#endif

L
Linus Torvalds 已提交
492 493 494 495 496 497 498 499
	/*
	 * 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
500
	unsigned int		dma_drain_size;
501
	void			*dma_drain_buffer;
502
	unsigned int		dma_pad_mask;
L
Linus Torvalds 已提交
503 504 505
	unsigned int		dma_alignment;

	struct blk_queue_tag	*queue_tags;
506
	struct list_head	tag_busy_list;
L
Linus Torvalds 已提交
507

508
	unsigned int		nr_sorted;
509
	unsigned int		in_flight[2];
510

511 512 513 514 515 516
	/*
	 * 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
517

J
Jens Axboe 已提交
518
	unsigned int		rq_timeout;
519
	int			poll_nsec;
520 521 522 523

	struct blk_stat_callback	*poll_cb;
	struct blk_rq_stat	poll_stat[2];

J
Jens Axboe 已提交
524
	struct timer_list	timeout;
525
	struct work_struct	timeout_work;
J
Jens Axboe 已提交
526 527
	struct list_head	timeout_list;

528
	struct list_head	icq_list;
529
#ifdef CONFIG_BLK_CGROUP
530
	DECLARE_BITMAP		(blkcg_pols, BLKCG_MAX_POLS);
T
Tejun Heo 已提交
531
	struct blkcg_gq		*root_blkg;
532
	struct list_head	blkg_list;
533
#endif
534

535 536
	struct queue_limits	limits;

L
Linus Torvalds 已提交
537 538 539 540 541
	/*
	 * sg stuff
	 */
	unsigned int		sg_timeout;
	unsigned int		sg_reserved_size;
542
	int			node;
543
#ifdef CONFIG_BLK_DEV_IO_TRACE
544
	struct blk_trace	*blk_trace;
545
#endif
L
Linus Torvalds 已提交
546
	/*
547
	 * for flush operations
L
Linus Torvalds 已提交
548
	 */
549
	struct blk_flush_queue	*fq;
550

551 552
	struct list_head	requeue_list;
	spinlock_t		requeue_lock;
553
	struct delayed_work	requeue_work;
554

555
	struct mutex		sysfs_lock;
556

557
	int			bypass_depth;
558
	atomic_t		mq_freeze_depth;
559

560
#if defined(CONFIG_BLK_DEV_BSG)
M
Mike Christie 已提交
561 562
	bsg_job_fn		*bsg_job_fn;
	int			bsg_job_size;
563 564
	struct bsg_class_device bsg_dev;
#endif
565 566 567 568 569

#ifdef CONFIG_BLK_DEV_THROTTLING
	/* Throttle data */
	struct throtl_data *td;
#endif
T
Tejun Heo 已提交
570
	struct rcu_head		rcu_head;
571
	wait_queue_head_t	mq_freeze_wq;
572
	struct percpu_ref	q_usage_counter;
573
	struct list_head	all_q_node;
574 575 576

	struct blk_mq_tag_set	*tag_set;
	struct list_head	tag_set_list;
577
	struct bio_set		*bio_split;
578

579
#ifdef CONFIG_BLK_DEBUG_FS
580 581 582 583
	struct dentry		*debugfs_dir;
	struct dentry		*mq_debugfs_dir;
#endif

584
	bool			mq_sysfs_init_done;
585 586 587

	size_t			cmd_size;
	void			*rq_alloc_data;
L
Linus Torvalds 已提交
588 589 590 591
};

#define QUEUE_FLAG_QUEUED	1	/* uses generic tag queueing */
#define QUEUE_FLAG_STOPPED	2	/* queue is stopped */
592 593
#define	QUEUE_FLAG_SYNCFULL	3	/* read queue has been filled */
#define QUEUE_FLAG_ASYNCFULL	4	/* write queue has been filled */
B
Bart Van Assche 已提交
594
#define QUEUE_FLAG_DYING	5	/* queue being torn down */
595
#define QUEUE_FLAG_BYPASS	6	/* act as dumb FIFO queue */
596 597
#define QUEUE_FLAG_BIDI		7	/* queue supports bidi requests */
#define QUEUE_FLAG_NOMERGES     8	/* disable merge attempts */
D
Dan Williams 已提交
598
#define QUEUE_FLAG_SAME_COMP	9	/* complete on same CPU-group */
599 600 601
#define QUEUE_FLAG_FAIL_IO     10	/* fake timeout */
#define QUEUE_FLAG_STACKABLE   11	/* supports request stacking */
#define QUEUE_FLAG_NONROT      12	/* non-rotational device (SSD) */
602
#define QUEUE_FLAG_VIRT        QUEUE_FLAG_NONROT /* paravirt device */
603 604 605 606
#define QUEUE_FLAG_IO_STAT     13	/* do IO stats */
#define QUEUE_FLAG_DISCARD     14	/* supports DISCARD */
#define QUEUE_FLAG_NOXMERGES   15	/* No extended merges */
#define QUEUE_FLAG_ADD_RANDOM  16	/* Contributes to random pool */
607
#define QUEUE_FLAG_SECERASE    17	/* supports secure erase */
D
Dan Williams 已提交
608
#define QUEUE_FLAG_SAME_FORCE  18	/* force complete on same CPU */
609
#define QUEUE_FLAG_DEAD        19	/* queue tear-down finished */
610
#define QUEUE_FLAG_INIT_DONE   20	/* queue is initialized */
611
#define QUEUE_FLAG_NO_SG_MERGE 21	/* don't attempt to merge SG segments*/
J
Jens Axboe 已提交
612
#define QUEUE_FLAG_POLL	       22	/* IO polling enabled if set */
613 614
#define QUEUE_FLAG_WC	       23	/* Write back caching */
#define QUEUE_FLAG_FUA	       24	/* device supports FUA writes */
J
Jens Axboe 已提交
615
#define QUEUE_FLAG_FLUSH_NQ    25	/* flush not queueuable */
616
#define QUEUE_FLAG_DAX         26	/* device supports DAX */
617
#define QUEUE_FLAG_STATS       27	/* track rq completion times */
618 619
#define QUEUE_FLAG_POLL_STATS  28	/* collecting stats for hybrid polling */
#define QUEUE_FLAG_REGISTERED  29	/* queue has been registered to a disk */
620 621

#define QUEUE_FLAG_DEFAULT	((1 << QUEUE_FLAG_IO_STAT) |		\
622
				 (1 << QUEUE_FLAG_STACKABLE)	|	\
623 624
				 (1 << QUEUE_FLAG_SAME_COMP)	|	\
				 (1 << QUEUE_FLAG_ADD_RANDOM))
625

626
#define QUEUE_FLAG_MQ_DEFAULT	((1 << QUEUE_FLAG_IO_STAT) |		\
627
				 (1 << QUEUE_FLAG_STACKABLE)	|	\
628 629
				 (1 << QUEUE_FLAG_SAME_COMP)	|	\
				 (1 << QUEUE_FLAG_POLL))
630

631
static inline void queue_lockdep_assert_held(struct request_queue *q)
632
{
633 634
	if (q->queue_lock)
		lockdep_assert_held(q->queue_lock);
635 636
}

N
Nick Piggin 已提交
637 638 639 640 641 642
static inline void queue_flag_set_unlocked(unsigned int flag,
					   struct request_queue *q)
{
	__set_bit(flag, &q->queue_flags);
}

J
Jens Axboe 已提交
643 644 645
static inline int queue_flag_test_and_clear(unsigned int flag,
					    struct request_queue *q)
{
646
	queue_lockdep_assert_held(q);
J
Jens Axboe 已提交
647 648 649 650 651 652 653 654 655 656 657 658

	if (test_bit(flag, &q->queue_flags)) {
		__clear_bit(flag, &q->queue_flags);
		return 1;
	}

	return 0;
}

static inline int queue_flag_test_and_set(unsigned int flag,
					  struct request_queue *q)
{
659
	queue_lockdep_assert_held(q);
J
Jens Axboe 已提交
660 661 662 663 664 665 666 667 668

	if (!test_bit(flag, &q->queue_flags)) {
		__set_bit(flag, &q->queue_flags);
		return 0;
	}

	return 1;
}

N
Nick Piggin 已提交
669 670
static inline void queue_flag_set(unsigned int flag, struct request_queue *q)
{
671
	queue_lockdep_assert_held(q);
N
Nick Piggin 已提交
672 673 674 675 676 677 678 679 680
	__set_bit(flag, &q->queue_flags);
}

static inline void queue_flag_clear_unlocked(unsigned int flag,
					     struct request_queue *q)
{
	__clear_bit(flag, &q->queue_flags);
}

681 682 683 684 685
static inline int queue_in_flight(struct request_queue *q)
{
	return q->in_flight[0] + q->in_flight[1];
}

N
Nick Piggin 已提交
686 687
static inline void queue_flag_clear(unsigned int flag, struct request_queue *q)
{
688
	queue_lockdep_assert_held(q);
N
Nick Piggin 已提交
689 690 691
	__clear_bit(flag, &q->queue_flags);
}

L
Linus Torvalds 已提交
692 693
#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
694
#define blk_queue_dying(q)	test_bit(QUEUE_FLAG_DYING, &(q)->queue_flags)
695
#define blk_queue_dead(q)	test_bit(QUEUE_FLAG_DEAD, &(q)->queue_flags)
696
#define blk_queue_bypass(q)	test_bit(QUEUE_FLAG_BYPASS, &(q)->queue_flags)
697
#define blk_queue_init_done(q)	test_bit(QUEUE_FLAG_INIT_DONE, &(q)->queue_flags)
698
#define blk_queue_nomerges(q)	test_bit(QUEUE_FLAG_NOMERGES, &(q)->queue_flags)
699 700
#define blk_queue_noxmerges(q)	\
	test_bit(QUEUE_FLAG_NOXMERGES, &(q)->queue_flags)
701
#define blk_queue_nonrot(q)	test_bit(QUEUE_FLAG_NONROT, &(q)->queue_flags)
702
#define blk_queue_io_stat(q)	test_bit(QUEUE_FLAG_IO_STAT, &(q)->queue_flags)
703
#define blk_queue_add_random(q)	test_bit(QUEUE_FLAG_ADD_RANDOM, &(q)->queue_flags)
704 705
#define blk_queue_stackable(q)	\
	test_bit(QUEUE_FLAG_STACKABLE, &(q)->queue_flags)
706
#define blk_queue_discard(q)	test_bit(QUEUE_FLAG_DISCARD, &(q)->queue_flags)
707 708
#define blk_queue_secure_erase(q) \
	(test_bit(QUEUE_FLAG_SECERASE, &(q)->queue_flags))
709
#define blk_queue_dax(q)	test_bit(QUEUE_FLAG_DAX, &(q)->queue_flags)
L
Linus Torvalds 已提交
710

711 712 713 714
#define blk_noretry_request(rq) \
	((rq)->cmd_flags & (REQ_FAILFAST_DEV|REQ_FAILFAST_TRANSPORT| \
			     REQ_FAILFAST_DRIVER))

715 716 717 718
static inline bool blk_account_rq(struct request *rq)
{
	return (rq->rq_flags & RQF_STARTED) && !blk_rq_is_passthrough(rq);
}
719

720
#define blk_rq_cpu_valid(rq)	((rq)->cpu != -1)
721
#define blk_bidi_rq(rq)		((rq)->next_rq != NULL)
722 723
/* rq->queuelist of dequeued request must be list_empty() */
#define blk_queued_rq(rq)	(!list_empty(&(rq)->queuelist))
L
Linus Torvalds 已提交
724 725 726

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

727
#define rq_data_dir(rq)		(op_is_write(req_op(rq)) ? WRITE : READ)
L
Linus Torvalds 已提交
728

729 730 731 732 733 734 735 736 737
/*
 * 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;
}

738 739 740 741 742
static inline unsigned int blk_queue_cluster(struct request_queue *q)
{
	return q->limits.cluster;
}

D
Damien Le Moal 已提交
743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759
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;
	}
}

760
static inline unsigned int blk_queue_zone_sectors(struct request_queue *q)
761 762 763 764
{
	return blk_queue_is_zoned(q) ? q->limits.chunk_sectors : 0;
}

765 766
static inline bool rq_is_sync(struct request *rq)
{
767
	return op_is_sync(rq->cmd_flags);
768 769
}

770
static inline bool blk_rl_full(struct request_list *rl, bool sync)
L
Linus Torvalds 已提交
771
{
772 773 774
	unsigned int flag = sync ? BLK_RL_SYNCFULL : BLK_RL_ASYNCFULL;

	return rl->flags & flag;
L
Linus Torvalds 已提交
775 776
}

777
static inline void blk_set_rl_full(struct request_list *rl, bool sync)
L
Linus Torvalds 已提交
778
{
779 780 781
	unsigned int flag = sync ? BLK_RL_SYNCFULL : BLK_RL_ASYNCFULL;

	rl->flags |= flag;
L
Linus Torvalds 已提交
782 783
}

784
static inline void blk_clear_rl_full(struct request_list *rl, bool sync)
L
Linus Torvalds 已提交
785
{
786 787 788
	unsigned int flag = sync ? BLK_RL_SYNCFULL : BLK_RL_ASYNCFULL;

	rl->flags &= ~flag;
L
Linus Torvalds 已提交
789 790
}

791 792
static inline bool rq_mergeable(struct request *rq)
{
793
	if (blk_rq_is_passthrough(rq))
794
		return false;
L
Linus Torvalds 已提交
795

796 797 798
	if (req_op(rq) == REQ_OP_FLUSH)
		return false;

799 800 801
	if (req_op(rq) == REQ_OP_WRITE_ZEROES)
		return false;

802
	if (rq->cmd_flags & REQ_NOMERGE_FLAGS)
803 804
		return false;
	if (rq->rq_flags & RQF_NOMERGE_FLAGS)
805 806 807 808
		return false;

	return true;
}
L
Linus Torvalds 已提交
809

810 811 812 813 814 815 816 817
static inline bool blk_write_same_mergeable(struct bio *a, struct bio *b)
{
	if (bio_data(a) == bio_data(b))
		return true;

	return false;
}

818 819 820 821 822 823 824 825
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
826 827 828
/*
 * q->prep_rq_fn return values
 */
829 830 831 832 833 834
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
835 836 837 838 839 840 841 842 843 844

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
 */
845 846

#if BITS_PER_LONG == 32
L
Linus Torvalds 已提交
847
#define BLK_BOUNCE_HIGH		((u64)blk_max_low_pfn << PAGE_SHIFT)
848 849 850 851
#else
#define BLK_BOUNCE_HIGH		-1ULL
#endif
#define BLK_BOUNCE_ANY		(-1ULL)
852
#define BLK_BOUNCE_ISA		(DMA_BIT_MASK(24))
L
Linus Torvalds 已提交
853

854 855 856 857
/*
 * default timeout for SG_IO if none specified
 */
#define BLK_DEFAULT_SG_TIMEOUT	(60 * HZ)
L
Linus Torvalds 已提交
858
#define BLK_MIN_SG_TIMEOUT	(7 * HZ)
859

860
#ifdef CONFIG_BOUNCE
L
Linus Torvalds 已提交
861
extern int init_emergency_isa_pool(void);
862
extern void blk_queue_bounce(struct request_queue *q, struct bio **bio);
L
Linus Torvalds 已提交
863 864 865 866 867
#else
static inline int init_emergency_isa_pool(void)
{
	return 0;
}
868
static inline void blk_queue_bounce(struct request_queue *q, struct bio **bio)
L
Linus Torvalds 已提交
869 870 871 872
{
}
#endif /* CONFIG_MMU */

873 874 875 876
struct rq_map_data {
	struct page **pages;
	int page_order;
	int nr_entries;
877
	unsigned long offset;
878
	int null_mapped;
879
	int from_user;
880 881
};

882
struct req_iterator {
883
	struct bvec_iter iter;
884 885 886 887
	struct bio *bio;
};

/* This should not be used directly - use rq_for_each_segment */
888 889
#define for_each_bio(_bio)		\
	for (; _bio; _bio = _bio->bi_next)
890
#define __rq_for_each_bio(_bio, rq)	\
L
Linus Torvalds 已提交
891 892 893
	if ((rq->bio))			\
		for (_bio = (rq)->bio; _bio; _bio = _bio->bi_next)

894 895
#define rq_for_each_segment(bvl, _rq, _iter)			\
	__rq_for_each_bio(_iter.bio, _rq)			\
896
		bio_for_each_segment(bvl, _iter.bio, _iter.iter)
897

K
Kent Overstreet 已提交
898
#define rq_iter_last(bvec, _iter)				\
899
		(_iter.bio->bi_next == NULL &&			\
K
Kent Overstreet 已提交
900
		 bio_iter_last(bvec, _iter.iter))
901

902 903 904 905 906 907 908 909 910 911 912
#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

T
Toshi Kani 已提交
913 914 915 916 917 918 919 920 921 922 923
#ifdef CONFIG_PRINTK
#define vfs_msg(sb, level, fmt, ...)				\
	__vfs_msg(sb, level, fmt, ##__VA_ARGS__)
#else
#define vfs_msg(sb, level, fmt, ...)				\
do {								\
	no_printk(fmt, ##__VA_ARGS__);				\
	__vfs_msg(sb, "", " ");					\
} while (0)
#endif

L
Linus Torvalds 已提交
924 925
extern int blk_register_queue(struct gendisk *disk);
extern void blk_unregister_queue(struct gendisk *disk);
926
extern blk_qc_t generic_make_request(struct bio *bio);
927
extern void blk_rq_init(struct request_queue *q, struct request *rq);
L
Linus Torvalds 已提交
928
extern void blk_put_request(struct request *);
929 930 931
extern void __blk_put_request(struct request_queue *, struct request *);
extern struct request *blk_get_request(struct request_queue *, int, gfp_t);
extern void blk_requeue_request(struct request_queue *, struct request *);
932
extern int blk_lld_busy(struct request_queue *q);
933 934 935 936 937
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);
938 939
extern int blk_insert_cloned_request(struct request_queue *q,
				     struct request *rq);
940
extern int blk_rq_append_bio(struct request *rq, struct bio *bio);
941
extern void blk_delay_queue(struct request_queue *, unsigned long);
942 943
extern void blk_queue_split(struct request_queue *, struct bio **,
			    struct bio_set *);
944
extern void blk_recount_segments(struct request_queue *, struct bio *);
945
extern int scsi_verify_blk_ioctl(struct block_device *, unsigned int);
946 947
extern int scsi_cmd_blk_ioctl(struct block_device *, fmode_t,
			      unsigned int, void __user *);
948 949
extern int scsi_cmd_ioctl(struct request_queue *, struct gendisk *, fmode_t,
			  unsigned int, void __user *);
950 951
extern int sg_scsi_ioctl(struct request_queue *, struct gendisk *, fmode_t,
			 struct scsi_ioctl_command __user *);
952

953
extern int blk_queue_enter(struct request_queue *q, bool nowait);
954
extern void blk_queue_exit(struct request_queue *q);
955
extern void blk_start_queue(struct request_queue *q);
J
Jens Axboe 已提交
956
extern void blk_start_queue_async(struct request_queue *q);
957
extern void blk_stop_queue(struct request_queue *q);
L
Linus Torvalds 已提交
958
extern void blk_sync_queue(struct request_queue *q);
959
extern void __blk_stop_queue(struct request_queue *q);
960
extern void __blk_run_queue(struct request_queue *q);
961
extern void __blk_run_queue_uncond(struct request_queue *q);
962
extern void blk_run_queue(struct request_queue *);
963
extern void blk_run_queue_async(struct request_queue *q);
964
extern void blk_mq_quiesce_queue(struct request_queue *q);
965
extern int blk_rq_map_user(struct request_queue *, struct request *,
966 967
			   struct rq_map_data *, void __user *, unsigned long,
			   gfp_t);
J
Jens Axboe 已提交
968
extern int blk_rq_unmap_user(struct bio *);
969 970
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 *,
971 972
			       struct rq_map_data *, const struct iov_iter *,
			       gfp_t);
973
extern int blk_execute_rq(struct request_queue *, struct gendisk *,
974
			  struct request *, int);
975
extern void blk_execute_rq_nowait(struct request_queue *, struct gendisk *,
976
				  struct request *, int, rq_end_io_fn *);
977

J
Jens Axboe 已提交
978
bool blk_mq_poll(struct request_queue *q, blk_qc_t cookie);
J
Jens Axboe 已提交
979

980
static inline struct request_queue *bdev_get_queue(struct block_device *bdev)
L
Linus Torvalds 已提交
981
{
982
	return bdev->bd_disk->queue;	/* this is never NULL */
L
Linus Torvalds 已提交
983 984
}

985
/*
986 987 988 989 990 991
 * 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
992
 */
993 994
static inline sector_t blk_rq_pos(const struct request *rq)
{
995
	return rq->__sector;
996 997 998 999
}

static inline unsigned int blk_rq_bytes(const struct request *rq)
{
1000
	return rq->__data_len;
1001 1002
}

1003 1004 1005 1006
static inline int blk_rq_cur_bytes(const struct request *rq)
{
	return rq->bio ? bio_cur_bytes(rq->bio) : 0;
}
1007

1008 1009
extern unsigned int blk_rq_err_bytes(const struct request *rq);

1010 1011
static inline unsigned int blk_rq_sectors(const struct request *rq)
{
1012
	return blk_rq_bytes(rq) >> 9;
1013 1014 1015 1016
}

static inline unsigned int blk_rq_cur_sectors(const struct request *rq)
{
1017
	return blk_rq_cur_bytes(rq) >> 9;
1018 1019
}

1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032
/*
 * 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);
}

1033
static inline unsigned int blk_queue_get_max_sectors(struct request_queue *q,
1034
						     int op)
1035
{
A
Adrian Hunter 已提交
1036
	if (unlikely(op == REQ_OP_DISCARD || op == REQ_OP_SECURE_ERASE))
1037
		return min(q->limits.max_discard_sectors, UINT_MAX >> 9);
1038

1039
	if (unlikely(op == REQ_OP_WRITE_SAME))
1040 1041
		return q->limits.max_write_same_sectors;

1042 1043 1044
	if (unlikely(op == REQ_OP_WRITE_ZEROES))
		return q->limits.max_write_zeroes_sectors;

1045 1046 1047
	return q->limits.max_sectors;
}

1048 1049 1050 1051 1052 1053 1054 1055
/*
 * 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)
1056
		return q->limits.max_sectors;
1057 1058 1059 1060 1061

	return q->limits.chunk_sectors -
			(offset & (q->limits.chunk_sectors - 1));
}

D
Damien Le Moal 已提交
1062 1063
static inline unsigned int blk_rq_get_max_sectors(struct request *rq,
						  sector_t offset)
1064 1065 1066
{
	struct request_queue *q = rq->q;

1067
	if (blk_rq_is_passthrough(rq))
1068 1069
		return q->limits.max_hw_sectors;

A
Adrian Hunter 已提交
1070 1071 1072
	if (!q->limits.chunk_sectors ||
	    req_op(rq) == REQ_OP_DISCARD ||
	    req_op(rq) == REQ_OP_SECURE_ERASE)
1073
		return blk_queue_get_max_sectors(q, req_op(rq));
1074

D
Damien Le Moal 已提交
1075
	return min(blk_max_size_offset(q, offset),
1076
			blk_queue_get_max_sectors(q, req_op(rq)));
1077 1078
}

1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089
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;
}

1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103
/*
 * blk_rq_set_prio - associate a request with prio from ioc
 * @rq: request of interest
 * @ioc: target iocontext
 *
 * Assocate request prio with ioc prio so request based drivers
 * can leverage priority information.
 */
static inline void blk_rq_set_prio(struct request *rq, struct io_context *ioc)
{
	if (ioc)
		rq->ioprio = ioc->ioprio;
}

1104 1105 1106 1107 1108 1109 1110
/*
 * 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);

L
Linus Torvalds 已提交
1111
/*
1112 1113 1114 1115 1116
 * Request completion related functions.
 *
 * blk_update_request() completes given number of bytes and updates
 * the request without completing it.
 *
1117 1118
 * blk_end_request() and friends.  __blk_end_request() must be called
 * with the request queue spinlock acquired.
L
Linus Torvalds 已提交
1119 1120
 *
 * Several drivers define their own end_request and call
1121 1122
 * blk_end_request() for parts of the original function.
 * This prevents code duplication in drivers.
L
Linus Torvalds 已提交
1123
 */
1124 1125
extern bool blk_update_request(struct request *rq, int error,
			       unsigned int nr_bytes);
1126
extern void blk_finish_request(struct request *rq, int error);
1127 1128 1129 1130
extern bool blk_end_request(struct request *rq, int error,
			    unsigned int nr_bytes);
extern void blk_end_request_all(struct request *rq, int error);
extern bool blk_end_request_cur(struct request *rq, int error);
1131
extern bool blk_end_request_err(struct request *rq, int error);
1132 1133 1134 1135
extern bool __blk_end_request(struct request *rq, int error,
			      unsigned int nr_bytes);
extern void __blk_end_request_all(struct request *rq, int error);
extern bool __blk_end_request_cur(struct request *rq, int error);
1136
extern bool __blk_end_request_err(struct request *rq, int error);
1137

1138
extern void blk_complete_request(struct request *);
J
Jens Axboe 已提交
1139 1140
extern void __blk_complete_request(struct request *);
extern void blk_abort_request(struct request *);
1141
extern void blk_unprep_request(struct request *);
1142

L
Linus Torvalds 已提交
1143 1144 1145
/*
 * Access functions for manipulating queue properties
 */
1146
extern struct request_queue *blk_init_queue_node(request_fn_proc *rfn,
1147
					spinlock_t *lock, int node_id);
1148
extern struct request_queue *blk_init_queue(request_fn_proc *, spinlock_t *);
1149
extern int blk_init_allocated_queue(struct request_queue *);
1150 1151 1152
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);
1153
extern void blk_queue_max_hw_sectors(struct request_queue *, unsigned int);
1154
extern void blk_queue_chunk_sectors(struct request_queue *, unsigned int);
1155
extern void blk_queue_max_segments(struct request_queue *, unsigned short);
1156 1157
extern void blk_queue_max_discard_segments(struct request_queue *,
		unsigned short);
1158
extern void blk_queue_max_segment_size(struct request_queue *, unsigned int);
1159 1160
extern void blk_queue_max_discard_sectors(struct request_queue *q,
		unsigned int max_discard_sectors);
1161 1162
extern void blk_queue_max_write_same_sectors(struct request_queue *q,
		unsigned int max_write_same_sectors);
1163 1164
extern void blk_queue_max_write_zeroes_sectors(struct request_queue *q,
		unsigned int max_write_same_sectors);
1165
extern void blk_queue_logical_block_size(struct request_queue *, unsigned short);
1166
extern void blk_queue_physical_block_size(struct request_queue *, unsigned int);
1167 1168
extern void blk_queue_alignment_offset(struct request_queue *q,
				       unsigned int alignment);
1169
extern void blk_limits_io_min(struct queue_limits *limits, unsigned int min);
1170
extern void blk_queue_io_min(struct request_queue *q, unsigned int min);
1171
extern void blk_limits_io_opt(struct queue_limits *limits, unsigned int opt);
1172
extern void blk_queue_io_opt(struct request_queue *q, unsigned int opt);
1173
extern void blk_set_queue_depth(struct request_queue *q, unsigned int depth);
1174
extern void blk_set_default_limits(struct queue_limits *lim);
1175
extern void blk_set_stacking_limits(struct queue_limits *lim);
1176 1177
extern int blk_stack_limits(struct queue_limits *t, struct queue_limits *b,
			    sector_t offset);
1178 1179
extern int bdev_stack_limits(struct queue_limits *t, struct block_device *bdev,
			    sector_t offset);
1180 1181
extern void disk_stack_limits(struct gendisk *disk, struct block_device *bdev,
			      sector_t offset);
1182
extern void blk_queue_stack_limits(struct request_queue *t, struct request_queue *b);
1183
extern void blk_queue_dma_pad(struct request_queue *, unsigned int);
1184
extern void blk_queue_update_dma_pad(struct request_queue *, unsigned int);
1185 1186 1187
extern int blk_queue_dma_drain(struct request_queue *q,
			       dma_drain_needed_fn *dma_drain_needed,
			       void *buf, unsigned int size);
1188
extern void blk_queue_lld_busy(struct request_queue *q, lld_busy_fn *fn);
1189
extern void blk_queue_segment_boundary(struct request_queue *, unsigned long);
1190
extern void blk_queue_virt_boundary(struct request_queue *, unsigned long);
1191
extern void blk_queue_prep_rq(struct request_queue *, prep_rq_fn *pfn);
1192
extern void blk_queue_unprep_rq(struct request_queue *, unprep_rq_fn *ufn);
1193
extern void blk_queue_dma_alignment(struct request_queue *, int);
1194
extern void blk_queue_update_dma_alignment(struct request_queue *, int);
1195
extern void blk_queue_softirq_done(struct request_queue *, softirq_done_fn *);
J
Jens Axboe 已提交
1196 1197
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);
1198
extern void blk_queue_flush_queueable(struct request_queue *q, bool queueable);
1199
extern void blk_queue_write_cache(struct request_queue *q, bool enabled, bool fua);
L
Linus Torvalds 已提交
1200

1201 1202 1203 1204 1205 1206 1207 1208 1209
/*
 * 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.
 */
1210 1211 1212 1213 1214 1215 1216
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;
}

1217 1218 1219 1220 1221 1222 1223 1224 1225
/*
 * 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);
}

1226
extern int blk_rq_map_sg(struct request_queue *, struct request *, struct scatterlist *);
L
Linus Torvalds 已提交
1227 1228 1229
extern void blk_dump_rq_flags(struct request *, char *);
extern long nr_blockdev_pages(void);

T
Tejun Heo 已提交
1230
bool __must_check blk_get_queue(struct request_queue *);
1231 1232 1233
struct request_queue *blk_alloc_queue(gfp_t);
struct request_queue *blk_alloc_queue_node(gfp_t, int);
extern void blk_put_queue(struct request_queue *);
1234
extern void blk_set_queue_dying(struct request_queue *);
L
Linus Torvalds 已提交
1235

L
Lin Ming 已提交
1236 1237 1238
/*
 * block layer runtime pm functions
 */
1239
#ifdef CONFIG_PM
L
Lin Ming 已提交
1240 1241 1242 1243 1244
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);
1245
extern void blk_set_runtime_active(struct request_queue *q);
L
Lin Ming 已提交
1246 1247 1248 1249 1250 1251 1252 1253 1254 1255
#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) {}
1256
static inline void blk_set_runtime_active(struct request_queue *q) {}
L
Lin Ming 已提交
1257 1258
#endif

1259
/*
S
Suresh Jayaraman 已提交
1260 1261 1262 1263 1264 1265 1266 1267 1268 1269
 * 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.
1270
 */
1271
struct blk_plug {
S
Suresh Jayaraman 已提交
1272
	struct list_head list; /* requests */
1273
	struct list_head mq_list; /* blk-mq requests */
S
Suresh Jayaraman 已提交
1274
	struct list_head cb_list; /* md requires an unplug callback */
1275
};
1276
#define BLK_MAX_REQUEST_COUNT 16
1277
#define BLK_PLUG_FLUSH_SIZE (128 * 1024)
1278

1279
struct blk_plug_cb;
1280
typedef void (*blk_plug_cb_fn)(struct blk_plug_cb *, bool);
1281 1282
struct blk_plug_cb {
	struct list_head list;
1283 1284
	blk_plug_cb_fn callback;
	void *data;
1285
};
1286 1287
extern struct blk_plug_cb *blk_check_plugged(blk_plug_cb_fn unplug,
					     void *data, int size);
1288 1289
extern void blk_start_plug(struct blk_plug *);
extern void blk_finish_plug(struct blk_plug *);
1290
extern void blk_flush_plug_list(struct blk_plug *, bool);
1291 1292 1293 1294 1295

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

1296 1297 1298 1299 1300 1301 1302 1303
	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;

1304
	if (plug)
1305
		blk_flush_plug_list(plug, true);
1306 1307 1308 1309 1310 1311
}

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

1312 1313 1314 1315
	return plug &&
		(!list_empty(&plug->list) ||
		 !list_empty(&plug->mq_list) ||
		 !list_empty(&plug->cb_list));
1316 1317
}

L
Linus Torvalds 已提交
1318 1319 1320
/*
 * tag stuff
 */
1321 1322 1323
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 *);
1324
extern int blk_queue_init_tags(struct request_queue *, int, struct blk_queue_tag *, int);
1325 1326 1327
extern void blk_queue_free_tags(struct request_queue *);
extern int blk_queue_resize_tags(struct request_queue *, int);
extern void blk_queue_invalidate_tags(struct request_queue *);
1328
extern struct blk_queue_tag *blk_init_tags(int, int);
1329
extern void blk_free_tags(struct blk_queue_tag *);
L
Linus Torvalds 已提交
1330

1331 1332 1333 1334 1335 1336 1337
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];
}
1338

1339 1340 1341
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);
1342 1343 1344

#define BLKDEV_DISCARD_SECURE	(1 << 0)	/* issue a secure erase */
#define BLKDEV_DISCARD_ZERO	(1 << 1)	/* must reliably zero data */
1345

1346 1347
extern int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
		sector_t nr_sects, gfp_t gfp_mask, unsigned long flags);
1348
extern int __blkdev_issue_discard(struct block_device *bdev, sector_t sector,
1349
		sector_t nr_sects, gfp_t gfp_mask, int flags,
1350
		struct bio **biop);
1351 1352

#define BLKDEV_ZERO_NOUNMAP	(1 << 0)  /* do not free blocks */
1353
#define BLKDEV_ZERO_NOFALLBACK	(1 << 1)  /* don't write explicit zeroes */
1354

1355 1356
extern int __blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
		sector_t nr_sects, gfp_t gfp_mask, struct bio **biop,
1357
		unsigned flags);
1358
extern int blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
1359 1360
		sector_t nr_sects, gfp_t gfp_mask, unsigned flags);

1361 1362
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
1363
{
1364 1365 1366
	return blkdev_issue_discard(sb->s_bdev, block << (sb->s_blocksize_bits - 9),
				    nr_blocks << (sb->s_blocksize_bits - 9),
				    gfp_mask, flags);
D
David Woodhouse 已提交
1367
}
1368
static inline int sb_issue_zeroout(struct super_block *sb, sector_t block,
1369
		sector_t nr_blocks, gfp_t gfp_mask)
1370 1371 1372 1373
{
	return blkdev_issue_zeroout(sb->s_bdev,
				    block << (sb->s_blocksize_bits - 9),
				    nr_blocks << (sb->s_blocksize_bits - 9),
1374
				    gfp_mask, 0);
1375
}
L
Linus Torvalds 已提交
1376

1377
extern int blk_verify_command(unsigned char *cmd, fmode_t has_write_perm);
1378

1379 1380 1381
enum blk_default_limits {
	BLK_MAX_SEGMENTS	= 128,
	BLK_SAFE_MAX_SECTORS	= 255,
1382
	BLK_DEF_MAX_SECTORS	= 2560,
1383 1384 1385
	BLK_MAX_SEGMENT_SIZE	= 65536,
	BLK_SEG_BOUNDARY_MASK	= 0xFFFFFFFFUL,
};
1386

L
Linus Torvalds 已提交
1387 1388
#define blkdev_entry_to_request(entry) list_entry((entry), struct request, queuelist)

1389 1390
static inline unsigned long queue_bounce_pfn(struct request_queue *q)
{
1391
	return q->limits.bounce_pfn;
1392 1393 1394 1395
}

static inline unsigned long queue_segment_boundary(struct request_queue *q)
{
1396
	return q->limits.seg_boundary_mask;
1397 1398
}

1399 1400 1401 1402 1403
static inline unsigned long queue_virt_boundary(struct request_queue *q)
{
	return q->limits.virt_boundary_mask;
}

1404 1405
static inline unsigned int queue_max_sectors(struct request_queue *q)
{
1406
	return q->limits.max_sectors;
1407 1408 1409 1410
}

static inline unsigned int queue_max_hw_sectors(struct request_queue *q)
{
1411
	return q->limits.max_hw_sectors;
1412 1413
}

1414
static inline unsigned short queue_max_segments(struct request_queue *q)
1415
{
1416
	return q->limits.max_segments;
1417 1418
}

1419 1420 1421 1422 1423
static inline unsigned short queue_max_discard_segments(struct request_queue *q)
{
	return q->limits.max_discard_segments;
}

1424 1425
static inline unsigned int queue_max_segment_size(struct request_queue *q)
{
1426
	return q->limits.max_segment_size;
1427 1428
}

1429
static inline unsigned short queue_logical_block_size(struct request_queue *q)
L
Linus Torvalds 已提交
1430 1431 1432
{
	int retval = 512;

1433 1434
	if (q && q->limits.logical_block_size)
		retval = q->limits.logical_block_size;
L
Linus Torvalds 已提交
1435 1436 1437 1438

	return retval;
}

1439
static inline unsigned short bdev_logical_block_size(struct block_device *bdev)
L
Linus Torvalds 已提交
1440
{
1441
	return queue_logical_block_size(bdev_get_queue(bdev));
L
Linus Torvalds 已提交
1442 1443
}

1444 1445 1446 1447 1448
static inline unsigned int queue_physical_block_size(struct request_queue *q)
{
	return q->limits.physical_block_size;
}

1449
static inline unsigned int bdev_physical_block_size(struct block_device *bdev)
M
Martin K. Petersen 已提交
1450 1451 1452 1453
{
	return queue_physical_block_size(bdev_get_queue(bdev));
}

1454 1455 1456 1457 1458
static inline unsigned int queue_io_min(struct request_queue *q)
{
	return q->limits.io_min;
}

M
Martin K. Petersen 已提交
1459 1460 1461 1462 1463
static inline int bdev_io_min(struct block_device *bdev)
{
	return queue_io_min(bdev_get_queue(bdev));
}

1464 1465 1466 1467 1468
static inline unsigned int queue_io_opt(struct request_queue *q)
{
	return q->limits.io_opt;
}

M
Martin K. Petersen 已提交
1469 1470 1471 1472 1473
static inline int bdev_io_opt(struct block_device *bdev)
{
	return queue_io_opt(bdev_get_queue(bdev));
}

1474 1475
static inline int queue_alignment_offset(struct request_queue *q)
{
M
Martin K. Petersen 已提交
1476
	if (q->limits.misaligned)
1477 1478
		return -1;

M
Martin K. Petersen 已提交
1479
	return q->limits.alignment_offset;
1480 1481
}

1482
static inline int queue_limit_alignment_offset(struct queue_limits *lim, sector_t sector)
1483 1484
{
	unsigned int granularity = max(lim->physical_block_size, lim->io_min);
1485
	unsigned int alignment = sector_div(sector, granularity >> 9) << 9;
1486

1487
	return (granularity + lim->alignment_offset - alignment) % granularity;
1488 1489
}

M
Martin K. Petersen 已提交
1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502
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;
}

1503 1504 1505 1506 1507 1508 1509 1510
static inline int queue_discard_alignment(struct request_queue *q)
{
	if (q->limits.discard_misaligned)
		return -1;

	return q->limits.discard_alignment;
}

1511
static inline int queue_limit_discard_alignment(struct queue_limits *lim, sector_t sector)
1512
{
1513
	unsigned int alignment, granularity, offset;
1514

1515 1516 1517
	if (!lim->max_discard_sectors)
		return 0;

1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531
	/* Why are these in bytes, not sectors? */
	alignment = lim->discard_alignment >> 9;
	granularity = lim->discard_granularity >> 9;
	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 */
	return offset << 9;
1532 1533
}

1534 1535 1536 1537 1538 1539 1540 1541 1542 1543
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;
}

1544 1545
static inline unsigned int queue_discard_zeroes_data(struct request_queue *q)
{
1546
	if (q->limits.max_discard_sectors && q->limits.discard_zeroes_data == 1)
1547 1548 1549 1550 1551 1552 1553 1554 1555 1556
		return 1;

	return 0;
}

static inline unsigned int bdev_discard_zeroes_data(struct block_device *bdev)
{
	return queue_discard_zeroes_data(bdev_get_queue(bdev));
}

1557 1558 1559 1560 1561 1562 1563 1564 1565 1566
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;
}

1567 1568 1569 1570 1571 1572 1573 1574 1575 1576
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
1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596
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;
}

1597
static inline unsigned int bdev_zone_sectors(struct block_device *bdev)
1598 1599 1600 1601
{
	struct request_queue *q = bdev_get_queue(bdev);

	if (q)
1602
		return blk_queue_zone_sectors(q);
1603 1604 1605 1606

	return 0;
}

1607
static inline int queue_dma_alignment(struct request_queue *q)
L
Linus Torvalds 已提交
1608
{
1609
	return q ? q->dma_alignment : 511;
L
Linus Torvalds 已提交
1610 1611
}

1612
static inline int blk_rq_aligned(struct request_queue *q, unsigned long addr,
1613 1614 1615
				 unsigned int len)
{
	unsigned int alignment = queue_dma_alignment(q) | q->dma_pad_mask;
1616
	return !(addr & alignment) && !(len & alignment);
1617 1618
}

L
Linus Torvalds 已提交
1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629
/* 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;
}

1630
static inline unsigned int block_size(struct block_device *bdev)
L
Linus Torvalds 已提交
1631 1632 1633 1634
{
	return bdev->bd_block_size;
}

1635 1636
static inline bool queue_flush_queueable(struct request_queue *q)
{
J
Jens Axboe 已提交
1637
	return !test_bit(QUEUE_FLAG_FLUSH_NQ, &q->queue_flags);
1638 1639
}

L
Linus Torvalds 已提交
1640 1641 1642 1643 1644 1645
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)
{
1646
	put_page(p.v);
L
Linus Torvalds 已提交
1647 1648
}

1649 1650 1651 1652 1653 1654 1655
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));
}

1656 1657 1658 1659 1660 1661 1662 1663 1664
/*
 * 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;
1665
	return __bvec_gap_to_prev(q, bprv, offset);
1666 1667
}

M
Ming Lei 已提交
1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686
/*
 * 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;
}

1687 1688 1689
static inline bool bio_will_gap(struct request_queue *q, struct bio *prev,
			 struct bio *next)
{
1690 1691 1692 1693 1694
	if (bio_has_data(prev) && queue_virt_boundary(q)) {
		struct bio_vec pb, nb;

		bio_get_last_bvec(prev, &pb);
		bio_get_first_bvec(next, &nb);
1695

M
Ming Lei 已提交
1696 1697
		if (!bios_segs_mergeable(q, prev, &pb, &nb))
			return __bvec_gap_to_prev(q, &pb, nb.bv_offset);
1698 1699 1700
	}

	return false;
1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712
}

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

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

1713
int kblockd_schedule_work(struct work_struct *work);
1714
int kblockd_schedule_work_on(int cpu, struct work_struct *work);
1715
int kblockd_schedule_delayed_work(struct delayed_work *dwork, unsigned long delay);
1716
int kblockd_schedule_delayed_work_on(int cpu, struct delayed_work *dwork, unsigned long delay);
L
Linus Torvalds 已提交
1717

1718
#ifdef CONFIG_BLK_CGROUP
1719 1720 1721 1722 1723
/*
 * This should not be using sched_clock(). A real patch is in progress
 * to fix this up, until that is in place we need to disable preemption
 * around sched_clock() in this function and set_io_start_time_ns().
 */
1724 1725
static inline void set_start_time_ns(struct request *req)
{
1726
	preempt_disable();
1727
	req->start_time_ns = sched_clock();
1728
	preempt_enable();
1729 1730 1731 1732
}

static inline void set_io_start_time_ns(struct request *req)
{
1733
	preempt_disable();
1734
	req->io_start_time_ns = sched_clock();
1735
	preempt_enable();
1736
}
1737 1738 1739 1740 1741 1742 1743 1744 1745 1746

static inline uint64_t rq_start_time_ns(struct request *req)
{
        return req->start_time_ns;
}

static inline uint64_t rq_io_start_time_ns(struct request *req)
{
        return req->io_start_time_ns;
}
1747 1748 1749
#else
static inline void set_start_time_ns(struct request *req) {}
static inline void set_io_start_time_ns(struct request *req) {}
1750 1751 1752 1753 1754 1755 1756 1757
static inline uint64_t rq_start_time_ns(struct request *req)
{
	return 0;
}
static inline uint64_t rq_io_start_time_ns(struct request *req)
{
	return 0;
}
1758 1759
#endif

L
Linus Torvalds 已提交
1760 1761 1762 1763 1764
#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) "-*")

1765 1766
#if defined(CONFIG_BLK_DEV_INTEGRITY)

1767 1768 1769
enum blk_integrity_flags {
	BLK_INTEGRITY_VERIFY		= 1 << 0,
	BLK_INTEGRITY_GENERATE		= 1 << 1,
1770
	BLK_INTEGRITY_DEVICE_CAPABLE	= 1 << 2,
1771
	BLK_INTEGRITY_IP_CHECKSUM	= 1 << 3,
1772
};
1773

1774
struct blk_integrity_iter {
1775 1776
	void			*prot_buf;
	void			*data_buf;
1777
	sector_t		seed;
1778
	unsigned int		data_size;
1779
	unsigned short		interval;
1780 1781 1782
	const char		*disk_name;
};

1783
typedef int (integrity_processing_fn) (struct blk_integrity_iter *);
1784

1785 1786 1787 1788 1789
struct blk_integrity_profile {
	integrity_processing_fn		*generate_fn;
	integrity_processing_fn		*verify_fn;
	const char			*name;
};
1790

1791
extern void blk_integrity_register(struct gendisk *, struct blk_integrity *);
1792
extern void blk_integrity_unregister(struct gendisk *);
1793
extern int blk_integrity_compare(struct gendisk *, struct gendisk *);
1794 1795 1796
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 *);
1797 1798 1799 1800
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 *);
1801

1802
static inline struct blk_integrity *blk_get_integrity(struct gendisk *disk)
1803
{
1804
	struct blk_integrity *bi = &disk->queue->integrity;
1805 1806 1807 1808 1809

	if (!bi->profile)
		return NULL;

	return bi;
1810 1811
}

1812 1813
static inline
struct blk_integrity *bdev_get_integrity(struct block_device *bdev)
1814
{
1815
	return blk_get_integrity(bdev->bd_disk);
1816 1817
}

1818
static inline bool blk_integrity_rq(struct request *rq)
1819
{
1820
	return rq->cmd_flags & REQ_INTEGRITY;
1821 1822
}

1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834
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;
}

1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854
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);
}

1855 1856
#else /* CONFIG_BLK_DEV_INTEGRITY */

1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878
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)
{
1879
	return NULL;
1880 1881 1882 1883 1884 1885 1886 1887 1888
}
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;
}
1889
static inline void blk_integrity_register(struct gendisk *d,
1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903
					 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;
}
1904 1905 1906
static inline bool blk_integrity_merge_rq(struct request_queue *rq,
					  struct request *r1,
					  struct request *r2)
1907
{
1908
	return true;
1909
}
1910 1911 1912
static inline bool blk_integrity_merge_bio(struct request_queue *rq,
					   struct request *r,
					   struct bio *b)
1913
{
1914
	return true;
1915
}
1916

1917 1918 1919 1920 1921 1922 1923 1924 1925 1926
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;
}
1927 1928 1929

#endif /* CONFIG_BLK_DEV_INTEGRITY */

1930 1931 1932 1933 1934 1935 1936 1937 1938
/**
 * struct blk_dax_ctl - control and output parameters for ->direct_access
 * @sector: (input) offset relative to a block_device
 * @addr: (output) kernel virtual address for @sector populated by driver
 * @pfn: (output) page frame number for @addr populated by driver
 * @size: (input) number of bytes requested
 */
struct blk_dax_ctl {
	sector_t sector;
D
Dan Williams 已提交
1939
	void *addr;
1940
	long size;
D
Dan Williams 已提交
1941
	pfn_t pfn;
1942 1943
};

1944
struct block_device_operations {
A
Al Viro 已提交
1945
	int (*open) (struct block_device *, fmode_t);
1946
	void (*release) (struct gendisk *, fmode_t);
1947
	int (*rw_page)(struct block_device *, sector_t, struct page *, bool);
A
Al Viro 已提交
1948 1949
	int (*ioctl) (struct block_device *, fmode_t, unsigned, unsigned long);
	int (*compat_ioctl) (struct block_device *, fmode_t, unsigned, unsigned long);
D
Dan Williams 已提交
1950 1951
	long (*direct_access)(struct block_device *, sector_t, void **, pfn_t *,
			long);
1952 1953 1954
	unsigned int (*check_events) (struct gendisk *disk,
				      unsigned int clearing);
	/* ->media_changed() is DEPRECATED, use ->check_events() instead */
1955
	int (*media_changed) (struct gendisk *);
1956
	void (*unlock_native_capacity) (struct gendisk *);
1957 1958
	int (*revalidate_disk) (struct gendisk *);
	int (*getgeo)(struct block_device *, struct hd_geometry *);
1959 1960
	/* this callback is with swap_lock and sometimes page table lock held */
	void (*swap_slot_free_notify) (struct block_device *, unsigned long);
1961
	struct module *owner;
1962
	const struct pr_ops *pr_ops;
1963 1964
};

1965 1966
extern int __blkdev_driver_ioctl(struct block_device *, fmode_t, unsigned int,
				 unsigned long);
1967 1968 1969
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 *);
1970
extern long bdev_direct_access(struct block_device *, struct blk_dax_ctl *);
1971
extern int bdev_dax_supported(struct super_block *, int);
1972
extern bool bdev_dax_capable(struct block_device *);
1973
#else /* CONFIG_BLOCK */
1974 1975 1976

struct block_device;

1977 1978 1979 1980 1981 1982 1983 1984 1985 1986
/*
 * stubs for when the block layer is configured out
 */
#define buffer_heads_over_limit 0

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

1987 1988 1989 1990
struct blk_plug {
};

static inline void blk_start_plug(struct blk_plug *plug)
1991 1992 1993
{
}

1994
static inline void blk_finish_plug(struct blk_plug *plug)
1995 1996 1997
{
}

1998
static inline void blk_flush_plug(struct task_struct *task)
1999 2000 2001
{
}

2002 2003 2004 2005 2006
static inline void blk_schedule_flush_plug(struct task_struct *task)
{
}


2007 2008 2009 2010 2011
static inline bool blk_needs_flush_plug(struct task_struct *tsk)
{
	return false;
}

2012 2013 2014 2015 2016 2017
static inline int blkdev_issue_flush(struct block_device *bdev, gfp_t gfp_mask,
				     sector_t *error_sector)
{
	return 0;
}

2018 2019
#endif /* CONFIG_BLOCK */

L
Linus Torvalds 已提交
2020
#endif