blkdev.h 53.9 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3
#ifndef _LINUX_BLKDEV_H
#define _LINUX_BLKDEV_H

4 5
#include <linux/sched.h>

J
Jens Axboe 已提交
6 7
#ifdef CONFIG_BLOCK

L
Linus Torvalds 已提交
8 9 10
#include <linux/major.h>
#include <linux/genhd.h>
#include <linux/list.h>
11
#include <linux/llist.h>
L
Linus Torvalds 已提交
12 13 14
#include <linux/timer.h>
#include <linux/workqueue.h>
#include <linux/pagemap.h>
15
#include <linux/backing-dev-defs.h>
L
Linus Torvalds 已提交
16 17
#include <linux/wait.h>
#include <linux/mempool.h>
D
Dan Williams 已提交
18
#include <linux/pfn.h>
L
Linus Torvalds 已提交
19 20
#include <linux/bio.h>
#include <linux/stringify.h>
21
#include <linux/gfp.h>
22
#include <linux/bsg.h>
23
#include <linux/smp.h>
T
Tejun Heo 已提交
24
#include <linux/rcupdate.h>
25
#include <linux/percpu-refcount.h>
C
Christoph Hellwig 已提交
26
#include <linux/scatterlist.h>
27
#include <linux/blkzoned.h>
L
Linus Torvalds 已提交
28

29
struct module;
30 31
struct scsi_ioctl_command;

L
Linus Torvalds 已提交
32 33
struct request_queue;
struct elevator_queue;
34
struct blk_trace;
35 36
struct request;
struct sg_io_hdr;
M
Mike Christie 已提交
37
struct bsg_job;
T
Tejun Heo 已提交
38
struct blkcg_gq;
39
struct blk_flush_queue;
40
struct pr_ops;
J
Jens Axboe 已提交
41
struct rq_wb;
L
Linus Torvalds 已提交
42 43 44 45

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

T
Tejun Heo 已提交
46 47 48 49 50 51
/*
 * Maximum number of blkcg policies allowed to be registered concurrently.
 * Defined here to simplify include dependency.
 */
#define BLKCG_MAX_POLS		2

52
typedef void (rq_end_io_fn)(struct request *, int);
L
Linus Torvalds 已提交
53

54 55 56
#define BLK_RL_SYNCFULL		(1U << 0)
#define BLK_RL_ASYNCFULL	(1U << 1)

L
Linus Torvalds 已提交
57
struct request_list {
58
	struct request_queue	*q;	/* the queue this rl belongs to */
59 60 61
#ifdef CONFIG_BLK_CGROUP
	struct blkcg_gq		*blkg;	/* blkg this request pool belongs to */
#endif
62 63 64 65
	/*
	 * count[], starved[], and wait[] are indexed by
	 * BLK_RW_SYNC/BLK_RW_ASYNC
	 */
66 67 68 69
	int			count[2];
	int			starved[2];
	mempool_t		*rq_pool;
	wait_queue_head_t	wait[2];
70
	unsigned int		flags;
L
Linus Torvalds 已提交
71 72
};

73 74 75 76 77 78
/*
 * request command types
 */
enum rq_cmd_type_bits {
	REQ_TYPE_FS		= 1,	/* fs request */
	REQ_TYPE_BLOCK_PC,		/* scsi command */
79
	REQ_TYPE_DRV_PRIV,		/* driver defined types from here */
80 81
};

82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120
/*
 * 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))
121 122
/* IO stats tracking on */
#define RQF_STATS		((__force req_flags_t)(1 << 17))
123 124 125 126 127

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

L
Linus Torvalds 已提交
128 129 130
#define BLK_MAX_CDB	16

/*
131 132 133 134
 * 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
135 136
 */
struct request {
137
	struct list_head queuelist;
138 139
	union {
		struct call_single_data csd;
140
		u64 fifo_time;
141
	};
142

143
	struct request_queue *q;
144
	struct blk_mq_ctx *mq_ctx;
145

146
	int cpu;
147
	unsigned cmd_type;
148
	unsigned int cmd_flags;		/* op and common flags */
149
	req_flags_t rq_flags;
J
Jens Axboe 已提交
150
	unsigned long atomic_flags;
L
Linus Torvalds 已提交
151

152 153
	/* the following two fields are internal, NEVER access directly */
	unsigned int __data_len;	/* total data len */
154
	sector_t __sector;		/* sector cursor */
L
Linus Torvalds 已提交
155 156 157 158

	struct bio *bio;
	struct bio *biotail;

159 160 161 162 163 164 165 166 167 168 169 170
	/*
	 * 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;
	};

171 172 173
	/*
	 * The rb_node is only used inside the io scheduler, requests
	 * are pruned when moved to the dispatch queue. So let the
174
	 * completion_data share space with the rb_node.
175 176 177
	 */
	union {
		struct rb_node rb_node;	/* sort/lookup */
178
		void *completion_data;
179
	};
180

181
	/*
182
	 * Three pointers are available for the IO schedulers, if they need
183 184
	 * more they have to dynamically allocate it.  Flush requests are
	 * never put on the IO scheduler. So let the flush fields share
185
	 * space with the elevator data.
186
	 */
187
	union {
188 189 190 191 192
		struct {
			struct io_cq		*icq;
			void			*priv[2];
		} elv;

193 194 195
		struct {
			unsigned int		seq;
			struct list_head	list;
196
			rq_end_io_fn		*saved_end_io;
197 198
		} flush;
	};
199

200
	struct gendisk *rq_disk;
201
	struct hd_struct *part;
L
Linus Torvalds 已提交
202
	unsigned long start_time;
203
	struct blk_issue_stat issue_stat;
204
#ifdef CONFIG_BLK_CGROUP
205
	struct request_list *rl;		/* rl this rq is alloced from */
206 207 208
	unsigned long long start_time_ns;
	unsigned long long io_start_time_ns;    /* when passed to hardware */
#endif
L
Linus Torvalds 已提交
209 210 211 212
	/* Number of scatter-gather DMA addr+len pairs after
	 * physical address coalescing is performed.
	 */
	unsigned short nr_phys_segments;
213 214 215
#if defined(CONFIG_BLK_DEV_INTEGRITY)
	unsigned short nr_integrity_segments;
#endif
L
Linus Torvalds 已提交
216

217 218
	unsigned short ioprio;

T
Tejun Heo 已提交
219
	void *special;		/* opaque pointer available for LLD use */
L
Linus Torvalds 已提交
220

221 222 223
	int tag;
	int errors;

L
Linus Torvalds 已提交
224 225 226
	/*
	 * when request is used as a packet command carrier
	 */
227 228
	unsigned char __cmd[BLK_MAX_CDB];
	unsigned char *cmd;
229
	unsigned short cmd_len;
L
Linus Torvalds 已提交
230

231
	unsigned int extra_len;	/* length of alignment and padding */
L
Linus Torvalds 已提交
232
	unsigned int sense_len;
T
Tejun Heo 已提交
233
	unsigned int resid_len;	/* residual count */
L
Linus Torvalds 已提交
234 235
	void *sense;

J
Jens Axboe 已提交
236 237
	unsigned long deadline;
	struct list_head timeout_list;
L
Linus Torvalds 已提交
238
	unsigned int timeout;
239
	int retries;
L
Linus Torvalds 已提交
240 241

	/*
242
	 * completion callback.
L
Linus Torvalds 已提交
243 244 245
	 */
	rq_end_io_fn *end_io;
	void *end_io_data;
246 247 248

	/* for bidi */
	struct request *next_rq;
L
Linus Torvalds 已提交
249 250
};

251 252 253 254 255
static inline unsigned short req_get_ioprio(struct request *req)
{
	return req->ioprio;
}

L
Linus Torvalds 已提交
256 257
#include <linux/elevator.h>

258 259
struct blk_queue_ctx;

260
typedef void (request_fn_proc) (struct request_queue *q);
261
typedef blk_qc_t (make_request_fn) (struct request_queue *q, struct bio *bio);
262
typedef int (prep_rq_fn) (struct request_queue *, struct request *);
263
typedef void (unprep_rq_fn) (struct request_queue *, struct request *);
L
Linus Torvalds 已提交
264 265

struct bio_vec;
266
typedef void (softirq_done_fn)(struct request *);
267
typedef int (dma_drain_needed_fn)(struct request *);
268
typedef int (lld_busy_fn) (struct request_queue *q);
M
Mike Christie 已提交
269
typedef int (bsg_job_fn) (struct bsg_job *);
L
Linus Torvalds 已提交
270

J
Jens Axboe 已提交
271 272 273 274 275 276 277 278
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
279 280 281 282 283 284 285 286 287 288
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 busy;			/* current depth */
	int max_depth;			/* what we will send to device */
289
	int real_max_depth;		/* what the array can hold */
L
Linus Torvalds 已提交
290
	atomic_t refcnt;		/* map can be shared */
291 292
	int alloc_policy;		/* tag allocation policy */
	int next_tag;			/* next tag */
L
Linus Torvalds 已提交
293
};
294 295
#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
296

297 298 299
#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
300 301 302 303 304 305 306 307 308
/*
 * 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 */
};

309 310 311
struct queue_limits {
	unsigned long		bounce_pfn;
	unsigned long		seg_boundary_mask;
312
	unsigned long		virt_boundary_mask;
313 314

	unsigned int		max_hw_sectors;
315
	unsigned int		max_dev_sectors;
316
	unsigned int		chunk_sectors;
317 318
	unsigned int		max_sectors;
	unsigned int		max_segment_size;
319 320 321 322
	unsigned int		physical_block_size;
	unsigned int		alignment_offset;
	unsigned int		io_min;
	unsigned int		io_opt;
323
	unsigned int		max_discard_sectors;
324
	unsigned int		max_hw_discard_sectors;
325
	unsigned int		max_write_same_sectors;
326 327
	unsigned int		discard_granularity;
	unsigned int		discard_alignment;
328 329

	unsigned short		logical_block_size;
330
	unsigned short		max_segments;
331
	unsigned short		max_integrity_segments;
332

333
	unsigned char		misaligned;
334
	unsigned char		discard_misaligned;
335
	unsigned char		cluster;
336
	unsigned char		discard_zeroes_data;
337
	unsigned char		raid_partial_stripes_expensive;
D
Damien Le Moal 已提交
338
	enum blk_zoned_model	zoned;
339 340
};

341 342 343 344 345 346 347 348 349 350 351 352 353
#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
354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374
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;
}

375 376
#endif /* CONFIG_BLK_DEV_ZONED */

377
struct request_queue {
L
Linus Torvalds 已提交
378 379 380 381 382
	/*
	 * Together with queue_head for cacheline sharing
	 */
	struct list_head	queue_head;
	struct request		*last_merge;
J
Jens Axboe 已提交
383
	struct elevator_queue	*elevator;
384 385
	int			nr_rqs[2];	/* # allocated [a]sync rqs */
	int			nr_rqs_elvpriv;	/* # allocated rqs w/ elvpriv */
L
Linus Torvalds 已提交
386

J
Jens Axboe 已提交
387 388
	struct rq_wb		*rq_wb;

L
Linus Torvalds 已提交
389
	/*
390 391 392 393
	 * 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
394
	 */
395
	struct request_list	root_rl;
L
Linus Torvalds 已提交
396 397 398 399

	request_fn_proc		*request_fn;
	make_request_fn		*make_request_fn;
	prep_rq_fn		*prep_rq_fn;
400
	unprep_rq_fn		*unprep_rq_fn;
401
	softirq_done_fn		*softirq_done_fn;
J
Jens Axboe 已提交
402
	rq_timed_out_fn		*rq_timed_out_fn;
403
	dma_drain_needed_fn	*dma_drain_needed;
404
	lld_busy_fn		*lld_busy_fn;
L
Linus Torvalds 已提交
405

406 407 408 409 410
	struct blk_mq_ops	*mq_ops;

	unsigned int		*mq_map;

	/* sw queues */
411
	struct blk_mq_ctx __percpu	*queue_ctx;
412 413
	unsigned int		nr_queues;

414 415
	unsigned int		queue_depth;

416 417 418 419
	/* hw dispatch queues */
	struct blk_mq_hw_ctx	**queue_hw_ctx;
	unsigned int		nr_hw_queues;

420 421 422
	/*
	 * Dispatch queue sorting
	 */
J
Jens Axboe 已提交
423
	sector_t		end_sector;
424 425
	struct request		*boundary_rq;

L
Linus Torvalds 已提交
426
	/*
427
	 * Delayed queue handling
L
Linus Torvalds 已提交
428
	 */
429
	struct delayed_work	delay_work;
L
Linus Torvalds 已提交
430 431 432 433 434 435 436 437 438 439

	struct backing_dev_info	backing_dev_info;

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

	/*
440
	 * various queue flags, see QUEUE_* below
L
Linus Torvalds 已提交
441
	 */
442
	unsigned long		queue_flags;
L
Linus Torvalds 已提交
443

444 445 446 447 448 449
	/*
	 * ida allocated id for this queue.  Used to index queues from
	 * ioctx.
	 */
	int			id;

L
Linus Torvalds 已提交
450
	/*
451
	 * queue needs bounce pages for pages above this limit
L
Linus Torvalds 已提交
452
	 */
453
	gfp_t			bounce_gfp;
L
Linus Torvalds 已提交
454 455

	/*
已提交
456 457 458
	 * 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
459
	 */
已提交
460
	spinlock_t		__queue_lock;
L
Linus Torvalds 已提交
461 462 463 464 465 466 467
	spinlock_t		*queue_lock;

	/*
	 * queue kobject
	 */
	struct kobject kobj;

468 469 470 471 472
	/*
	 * mq queue kobject
	 */
	struct kobject mq_kobj;

473 474 475 476
#ifdef  CONFIG_BLK_DEV_INTEGRITY
	struct blk_integrity integrity;
#endif	/* CONFIG_BLK_DEV_INTEGRITY */

477
#ifdef CONFIG_PM
L
Lin Ming 已提交
478 479 480 481 482
	struct device		*dev;
	int			rpm_status;
	unsigned int		nr_pending;
#endif

L
Linus Torvalds 已提交
483 484 485 486 487 488 489 490
	/*
	 * 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
491
	unsigned int		dma_drain_size;
492
	void			*dma_drain_buffer;
493
	unsigned int		dma_pad_mask;
L
Linus Torvalds 已提交
494 495 496
	unsigned int		dma_alignment;

	struct blk_queue_tag	*queue_tags;
497
	struct list_head	tag_busy_list;
L
Linus Torvalds 已提交
498

499
	unsigned int		nr_sorted;
500
	unsigned int		in_flight[2];
501 502 503

	struct blk_rq_stat	rq_stats[2];

504 505 506 507 508 509
	/*
	 * 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
510

J
Jens Axboe 已提交
511 512
	unsigned int		rq_timeout;
	struct timer_list	timeout;
513
	struct work_struct	timeout_work;
J
Jens Axboe 已提交
514 515
	struct list_head	timeout_list;

516
	struct list_head	icq_list;
517
#ifdef CONFIG_BLK_CGROUP
518
	DECLARE_BITMAP		(blkcg_pols, BLKCG_MAX_POLS);
T
Tejun Heo 已提交
519
	struct blkcg_gq		*root_blkg;
520
	struct list_head	blkg_list;
521
#endif
522

523 524
	struct queue_limits	limits;

L
Linus Torvalds 已提交
525 526 527 528 529
	/*
	 * sg stuff
	 */
	unsigned int		sg_timeout;
	unsigned int		sg_reserved_size;
530
	int			node;
531
#ifdef CONFIG_BLK_DEV_IO_TRACE
532
	struct blk_trace	*blk_trace;
533
#endif
L
Linus Torvalds 已提交
534
	/*
535
	 * for flush operations
L
Linus Torvalds 已提交
536
	 */
537
	struct blk_flush_queue	*fq;
538

539 540
	struct list_head	requeue_list;
	spinlock_t		requeue_lock;
541
	struct delayed_work	requeue_work;
542

543
	struct mutex		sysfs_lock;
544

545
	int			bypass_depth;
546
	atomic_t		mq_freeze_depth;
547

548
#if defined(CONFIG_BLK_DEV_BSG)
M
Mike Christie 已提交
549 550
	bsg_job_fn		*bsg_job_fn;
	int			bsg_job_size;
551 552
	struct bsg_class_device bsg_dev;
#endif
553 554 555 556 557

#ifdef CONFIG_BLK_DEV_THROTTLING
	/* Throttle data */
	struct throtl_data *td;
#endif
T
Tejun Heo 已提交
558
	struct rcu_head		rcu_head;
559
	wait_queue_head_t	mq_freeze_wq;
560
	struct percpu_ref	q_usage_counter;
561
	struct list_head	all_q_node;
562 563 564

	struct blk_mq_tag_set	*tag_set;
	struct list_head	tag_set_list;
565
	struct bio_set		*bio_split;
566 567

	bool			mq_sysfs_init_done;
L
Linus Torvalds 已提交
568 569 570 571
};

#define QUEUE_FLAG_QUEUED	1	/* uses generic tag queueing */
#define QUEUE_FLAG_STOPPED	2	/* queue is stopped */
572 573
#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
574
#define QUEUE_FLAG_DYING	5	/* queue being torn down */
575
#define QUEUE_FLAG_BYPASS	6	/* act as dumb FIFO queue */
576 577
#define QUEUE_FLAG_BIDI		7	/* queue supports bidi requests */
#define QUEUE_FLAG_NOMERGES     8	/* disable merge attempts */
D
Dan Williams 已提交
578
#define QUEUE_FLAG_SAME_COMP	9	/* complete on same CPU-group */
579 580 581
#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) */
582
#define QUEUE_FLAG_VIRT        QUEUE_FLAG_NONROT /* paravirt device */
583 584 585 586
#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 */
587
#define QUEUE_FLAG_SECERASE    17	/* supports secure erase */
D
Dan Williams 已提交
588
#define QUEUE_FLAG_SAME_FORCE  18	/* force complete on same CPU */
589
#define QUEUE_FLAG_DEAD        19	/* queue tear-down finished */
590
#define QUEUE_FLAG_INIT_DONE   20	/* queue is initialized */
591
#define QUEUE_FLAG_NO_SG_MERGE 21	/* don't attempt to merge SG segments*/
J
Jens Axboe 已提交
592
#define QUEUE_FLAG_POLL	       22	/* IO polling enabled if set */
593 594
#define QUEUE_FLAG_WC	       23	/* Write back caching */
#define QUEUE_FLAG_FUA	       24	/* device supports FUA writes */
J
Jens Axboe 已提交
595
#define QUEUE_FLAG_FLUSH_NQ    25	/* flush not queueuable */
596
#define QUEUE_FLAG_DAX         26	/* device supports DAX */
597
#define QUEUE_FLAG_STATS       27	/* track rq completion times */
598 599

#define QUEUE_FLAG_DEFAULT	((1 << QUEUE_FLAG_IO_STAT) |		\
600
				 (1 << QUEUE_FLAG_STACKABLE)	|	\
601 602
				 (1 << QUEUE_FLAG_SAME_COMP)	|	\
				 (1 << QUEUE_FLAG_ADD_RANDOM))
603

604
#define QUEUE_FLAG_MQ_DEFAULT	((1 << QUEUE_FLAG_IO_STAT) |		\
605
				 (1 << QUEUE_FLAG_STACKABLE)	|	\
606 607
				 (1 << QUEUE_FLAG_SAME_COMP)	|	\
				 (1 << QUEUE_FLAG_POLL))
608

609
static inline void queue_lockdep_assert_held(struct request_queue *q)
610
{
611 612
	if (q->queue_lock)
		lockdep_assert_held(q->queue_lock);
613 614
}

N
Nick Piggin 已提交
615 616 617 618 619 620
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
621 622 623
static inline int queue_flag_test_and_clear(unsigned int flag,
					    struct request_queue *q)
{
624
	queue_lockdep_assert_held(q);
J
Jens Axboe 已提交
625 626 627 628 629 630 631 632 633 634 635 636

	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)
{
637
	queue_lockdep_assert_held(q);
J
Jens Axboe 已提交
638 639 640 641 642 643 644 645 646

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

	return 1;
}

N
Nick Piggin 已提交
647 648
static inline void queue_flag_set(unsigned int flag, struct request_queue *q)
{
649
	queue_lockdep_assert_held(q);
N
Nick Piggin 已提交
650 651 652 653 654 655 656 657 658
	__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);
}

659 660 661 662 663
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
664 665
static inline void queue_flag_clear(unsigned int flag, struct request_queue *q)
{
666
	queue_lockdep_assert_held(q);
N
Nick Piggin 已提交
667 668 669
	__clear_bit(flag, &q->queue_flags);
}

L
Linus Torvalds 已提交
670 671
#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
672
#define blk_queue_dying(q)	test_bit(QUEUE_FLAG_DYING, &(q)->queue_flags)
673
#define blk_queue_dead(q)	test_bit(QUEUE_FLAG_DEAD, &(q)->queue_flags)
674
#define blk_queue_bypass(q)	test_bit(QUEUE_FLAG_BYPASS, &(q)->queue_flags)
675
#define blk_queue_init_done(q)	test_bit(QUEUE_FLAG_INIT_DONE, &(q)->queue_flags)
676
#define blk_queue_nomerges(q)	test_bit(QUEUE_FLAG_NOMERGES, &(q)->queue_flags)
677 678
#define blk_queue_noxmerges(q)	\
	test_bit(QUEUE_FLAG_NOXMERGES, &(q)->queue_flags)
679
#define blk_queue_nonrot(q)	test_bit(QUEUE_FLAG_NONROT, &(q)->queue_flags)
680
#define blk_queue_io_stat(q)	test_bit(QUEUE_FLAG_IO_STAT, &(q)->queue_flags)
681
#define blk_queue_add_random(q)	test_bit(QUEUE_FLAG_ADD_RANDOM, &(q)->queue_flags)
682 683
#define blk_queue_stackable(q)	\
	test_bit(QUEUE_FLAG_STACKABLE, &(q)->queue_flags)
684
#define blk_queue_discard(q)	test_bit(QUEUE_FLAG_DISCARD, &(q)->queue_flags)
685 686
#define blk_queue_secure_erase(q) \
	(test_bit(QUEUE_FLAG_SECERASE, &(q)->queue_flags))
687
#define blk_queue_dax(q)	test_bit(QUEUE_FLAG_DAX, &(q)->queue_flags)
L
Linus Torvalds 已提交
688

689 690 691 692 693
#define blk_noretry_request(rq) \
	((rq)->cmd_flags & (REQ_FAILFAST_DEV|REQ_FAILFAST_TRANSPORT| \
			     REQ_FAILFAST_DRIVER))

#define blk_account_rq(rq) \
694
	(((rq)->rq_flags & RQF_STARTED) && \
695
	 ((rq)->cmd_type == REQ_TYPE_FS))
696

697
#define blk_rq_cpu_valid(rq)	((rq)->cpu != -1)
698
#define blk_bidi_rq(rq)		((rq)->next_rq != NULL)
699 700
/* rq->queuelist of dequeued request must be list_empty() */
#define blk_queued_rq(rq)	(!list_empty(&(rq)->queuelist))
L
Linus Torvalds 已提交
701 702 703

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

704
#define rq_data_dir(rq)		(op_is_write(req_op(rq)) ? WRITE : READ)
L
Linus Torvalds 已提交
705

706 707 708 709 710 711 712 713 714
/*
 * 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;
}

715 716 717 718 719
static inline unsigned int blk_queue_cluster(struct request_queue *q)
{
	return q->limits.cluster;
}

D
Damien Le Moal 已提交
720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736
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;
	}
}

737 738 739 740 741
static inline unsigned int blk_queue_zone_size(struct request_queue *q)
{
	return blk_queue_is_zoned(q) ? q->limits.chunk_sectors : 0;
}

742 743
static inline bool rq_is_sync(struct request *rq)
{
744
	return op_is_sync(rq->cmd_flags);
745 746
}

747
static inline bool blk_rl_full(struct request_list *rl, bool sync)
L
Linus Torvalds 已提交
748
{
749 750 751
	unsigned int flag = sync ? BLK_RL_SYNCFULL : BLK_RL_ASYNCFULL;

	return rl->flags & flag;
L
Linus Torvalds 已提交
752 753
}

754
static inline void blk_set_rl_full(struct request_list *rl, bool sync)
L
Linus Torvalds 已提交
755
{
756 757 758
	unsigned int flag = sync ? BLK_RL_SYNCFULL : BLK_RL_ASYNCFULL;

	rl->flags |= flag;
L
Linus Torvalds 已提交
759 760
}

761
static inline void blk_clear_rl_full(struct request_list *rl, bool sync)
L
Linus Torvalds 已提交
762
{
763 764 765
	unsigned int flag = sync ? BLK_RL_SYNCFULL : BLK_RL_ASYNCFULL;

	rl->flags &= ~flag;
L
Linus Torvalds 已提交
766 767
}

768 769 770 771
static inline bool rq_mergeable(struct request *rq)
{
	if (rq->cmd_type != REQ_TYPE_FS)
		return false;
L
Linus Torvalds 已提交
772

773 774 775
	if (req_op(rq) == REQ_OP_FLUSH)
		return false;

776
	if (rq->cmd_flags & REQ_NOMERGE_FLAGS)
777 778
		return false;
	if (rq->rq_flags & RQF_NOMERGE_FLAGS)
779 780 781 782
		return false;

	return true;
}
L
Linus Torvalds 已提交
783

784 785 786 787 788 789 790 791
static inline bool blk_write_same_mergeable(struct bio *a, struct bio *b)
{
	if (bio_data(a) == bio_data(b))
		return true;

	return false;
}

792 793 794 795 796 797 798 799
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
800 801 802
/*
 * q->prep_rq_fn return values
 */
803 804 805 806 807 808
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
809 810 811 812 813 814 815 816 817 818

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
 */
819 820

#if BITS_PER_LONG == 32
L
Linus Torvalds 已提交
821
#define BLK_BOUNCE_HIGH		((u64)blk_max_low_pfn << PAGE_SHIFT)
822 823 824 825
#else
#define BLK_BOUNCE_HIGH		-1ULL
#endif
#define BLK_BOUNCE_ANY		(-1ULL)
826
#define BLK_BOUNCE_ISA		(DMA_BIT_MASK(24))
L
Linus Torvalds 已提交
827

828 829 830 831
/*
 * default timeout for SG_IO if none specified
 */
#define BLK_DEFAULT_SG_TIMEOUT	(60 * HZ)
L
Linus Torvalds 已提交
832
#define BLK_MIN_SG_TIMEOUT	(7 * HZ)
833

834
#ifdef CONFIG_BOUNCE
L
Linus Torvalds 已提交
835
extern int init_emergency_isa_pool(void);
836
extern void blk_queue_bounce(struct request_queue *q, struct bio **bio);
L
Linus Torvalds 已提交
837 838 839 840 841
#else
static inline int init_emergency_isa_pool(void)
{
	return 0;
}
842
static inline void blk_queue_bounce(struct request_queue *q, struct bio **bio)
L
Linus Torvalds 已提交
843 844 845 846
{
}
#endif /* CONFIG_MMU */

847 848 849 850
struct rq_map_data {
	struct page **pages;
	int page_order;
	int nr_entries;
851
	unsigned long offset;
852
	int null_mapped;
853
	int from_user;
854 855
};

856
struct req_iterator {
857
	struct bvec_iter iter;
858 859 860 861
	struct bio *bio;
};

/* This should not be used directly - use rq_for_each_segment */
862 863
#define for_each_bio(_bio)		\
	for (; _bio; _bio = _bio->bi_next)
864
#define __rq_for_each_bio(_bio, rq)	\
L
Linus Torvalds 已提交
865 866 867
	if ((rq->bio))			\
		for (_bio = (rq)->bio; _bio; _bio = _bio->bi_next)

868 869
#define rq_for_each_segment(bvl, _rq, _iter)			\
	__rq_for_each_bio(_iter.bio, _rq)			\
870
		bio_for_each_segment(bvl, _iter.bio, _iter.iter)
871

K
Kent Overstreet 已提交
872
#define rq_iter_last(bvec, _iter)				\
873
		(_iter.bio->bi_next == NULL &&			\
K
Kent Overstreet 已提交
874
		 bio_iter_last(bvec, _iter.iter))
875

876 877 878 879 880 881 882 883 884 885 886
#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
887 888 889 890 891 892 893 894 895 896 897
#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
898 899
extern int blk_register_queue(struct gendisk *disk);
extern void blk_unregister_queue(struct gendisk *disk);
900
extern blk_qc_t generic_make_request(struct bio *bio);
901
extern void blk_rq_init(struct request_queue *q, struct request *rq);
L
Linus Torvalds 已提交
902
extern void blk_put_request(struct request *);
903 904
extern void __blk_put_request(struct request_queue *, struct request *);
extern struct request *blk_get_request(struct request_queue *, int, gfp_t);
J
Jens Axboe 已提交
905
extern void blk_rq_set_block_pc(struct request *);
906
extern void blk_requeue_request(struct request_queue *, struct request *);
907
extern void blk_add_request_payload(struct request *rq, struct page *page,
908
		int offset, unsigned int len);
909
extern int blk_lld_busy(struct request_queue *q);
910 911 912 913 914
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);
915 916
extern int blk_insert_cloned_request(struct request_queue *q,
				     struct request *rq);
917
extern int blk_rq_append_bio(struct request *rq, struct bio *bio);
918
extern void blk_delay_queue(struct request_queue *, unsigned long);
919 920
extern void blk_queue_split(struct request_queue *, struct bio **,
			    struct bio_set *);
921
extern void blk_recount_segments(struct request_queue *, struct bio *);
922
extern int scsi_verify_blk_ioctl(struct block_device *, unsigned int);
923 924
extern int scsi_cmd_blk_ioctl(struct block_device *, fmode_t,
			      unsigned int, void __user *);
925 926
extern int scsi_cmd_ioctl(struct request_queue *, struct gendisk *, fmode_t,
			  unsigned int, void __user *);
927 928
extern int sg_scsi_ioctl(struct request_queue *, struct gendisk *, fmode_t,
			 struct scsi_ioctl_command __user *);
929

930
extern int blk_queue_enter(struct request_queue *q, bool nowait);
931
extern void blk_queue_exit(struct request_queue *q);
932
extern void blk_start_queue(struct request_queue *q);
J
Jens Axboe 已提交
933
extern void blk_start_queue_async(struct request_queue *q);
934
extern void blk_stop_queue(struct request_queue *q);
L
Linus Torvalds 已提交
935
extern void blk_sync_queue(struct request_queue *q);
936
extern void __blk_stop_queue(struct request_queue *q);
937
extern void __blk_run_queue(struct request_queue *q);
938
extern void __blk_run_queue_uncond(struct request_queue *q);
939
extern void blk_run_queue(struct request_queue *);
940
extern void blk_run_queue_async(struct request_queue *q);
941
extern void blk_mq_quiesce_queue(struct request_queue *q);
942
extern int blk_rq_map_user(struct request_queue *, struct request *,
943 944
			   struct rq_map_data *, void __user *, unsigned long,
			   gfp_t);
J
Jens Axboe 已提交
945
extern int blk_rq_unmap_user(struct bio *);
946 947
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 *,
948 949
			       struct rq_map_data *, const struct iov_iter *,
			       gfp_t);
950
extern int blk_execute_rq(struct request_queue *, struct gendisk *,
951
			  struct request *, int);
952
extern void blk_execute_rq_nowait(struct request_queue *, struct gendisk *,
953
				  struct request *, int, rq_end_io_fn *);
954

J
Jens Axboe 已提交
955 956
bool blk_poll(struct request_queue *q, blk_qc_t cookie);

957
static inline struct request_queue *bdev_get_queue(struct block_device *bdev)
L
Linus Torvalds 已提交
958
{
959
	return bdev->bd_disk->queue;	/* this is never NULL */
L
Linus Torvalds 已提交
960 961
}

962
/*
963 964 965 966 967 968
 * 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
969
 */
970 971
static inline sector_t blk_rq_pos(const struct request *rq)
{
972
	return rq->__sector;
973 974 975 976
}

static inline unsigned int blk_rq_bytes(const struct request *rq)
{
977
	return rq->__data_len;
978 979
}

980 981 982 983
static inline int blk_rq_cur_bytes(const struct request *rq)
{
	return rq->bio ? bio_cur_bytes(rq->bio) : 0;
}
984

985 986
extern unsigned int blk_rq_err_bytes(const struct request *rq);

987 988
static inline unsigned int blk_rq_sectors(const struct request *rq)
{
989
	return blk_rq_bytes(rq) >> 9;
990 991 992 993
}

static inline unsigned int blk_rq_cur_sectors(const struct request *rq)
{
994
	return blk_rq_cur_bytes(rq) >> 9;
995 996
}

997
static inline unsigned int blk_queue_get_max_sectors(struct request_queue *q,
998
						     int op)
999
{
A
Adrian Hunter 已提交
1000
	if (unlikely(op == REQ_OP_DISCARD || op == REQ_OP_SECURE_ERASE))
1001
		return min(q->limits.max_discard_sectors, UINT_MAX >> 9);
1002

1003
	if (unlikely(op == REQ_OP_WRITE_SAME))
1004 1005
		return q->limits.max_write_same_sectors;

1006 1007 1008
	return q->limits.max_sectors;
}

1009 1010 1011 1012 1013 1014 1015 1016
/*
 * 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)
1017
		return q->limits.max_sectors;
1018 1019 1020 1021 1022

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

D
Damien Le Moal 已提交
1023 1024
static inline unsigned int blk_rq_get_max_sectors(struct request *rq,
						  sector_t offset)
1025 1026 1027
{
	struct request_queue *q = rq->q;

1028
	if (unlikely(rq->cmd_type != REQ_TYPE_FS))
1029 1030
		return q->limits.max_hw_sectors;

A
Adrian Hunter 已提交
1031 1032 1033
	if (!q->limits.chunk_sectors ||
	    req_op(rq) == REQ_OP_DISCARD ||
	    req_op(rq) == REQ_OP_SECURE_ERASE)
1034
		return blk_queue_get_max_sectors(q, req_op(rq));
1035

D
Damien Le Moal 已提交
1036
	return min(blk_max_size_offset(q, offset),
1037
			blk_queue_get_max_sectors(q, req_op(rq)));
1038 1039
}

1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050
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;
}

1051 1052 1053 1054 1055 1056 1057
/*
 * 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
1058
/*
1059 1060 1061 1062 1063
 * Request completion related functions.
 *
 * blk_update_request() completes given number of bytes and updates
 * the request without completing it.
 *
1064 1065
 * blk_end_request() and friends.  __blk_end_request() must be called
 * with the request queue spinlock acquired.
L
Linus Torvalds 已提交
1066 1067
 *
 * Several drivers define their own end_request and call
1068 1069
 * blk_end_request() for parts of the original function.
 * This prevents code duplication in drivers.
L
Linus Torvalds 已提交
1070
 */
1071 1072
extern bool blk_update_request(struct request *rq, int error,
			       unsigned int nr_bytes);
1073
extern void blk_finish_request(struct request *rq, int error);
1074 1075 1076 1077
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);
1078
extern bool blk_end_request_err(struct request *rq, int error);
1079 1080 1081 1082
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);
1083
extern bool __blk_end_request_err(struct request *rq, int error);
1084

1085
extern void blk_complete_request(struct request *);
J
Jens Axboe 已提交
1086 1087
extern void __blk_complete_request(struct request *);
extern void blk_abort_request(struct request *);
1088
extern void blk_unprep_request(struct request *);
1089

L
Linus Torvalds 已提交
1090 1091 1092
/*
 * Access functions for manipulating queue properties
 */
1093
extern struct request_queue *blk_init_queue_node(request_fn_proc *rfn,
1094
					spinlock_t *lock, int node_id);
1095
extern struct request_queue *blk_init_queue(request_fn_proc *, spinlock_t *);
1096 1097
extern struct request_queue *blk_init_allocated_queue(struct request_queue *,
						      request_fn_proc *, spinlock_t *);
1098 1099 1100
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);
1101
extern void blk_queue_max_hw_sectors(struct request_queue *, unsigned int);
1102
extern void blk_queue_chunk_sectors(struct request_queue *, unsigned int);
1103
extern void blk_queue_max_segments(struct request_queue *, unsigned short);
1104
extern void blk_queue_max_segment_size(struct request_queue *, unsigned int);
1105 1106
extern void blk_queue_max_discard_sectors(struct request_queue *q,
		unsigned int max_discard_sectors);
1107 1108
extern void blk_queue_max_write_same_sectors(struct request_queue *q,
		unsigned int max_write_same_sectors);
1109
extern void blk_queue_logical_block_size(struct request_queue *, unsigned short);
1110
extern void blk_queue_physical_block_size(struct request_queue *, unsigned int);
1111 1112
extern void blk_queue_alignment_offset(struct request_queue *q,
				       unsigned int alignment);
1113
extern void blk_limits_io_min(struct queue_limits *limits, unsigned int min);
1114
extern void blk_queue_io_min(struct request_queue *q, unsigned int min);
1115
extern void blk_limits_io_opt(struct queue_limits *limits, unsigned int opt);
1116
extern void blk_queue_io_opt(struct request_queue *q, unsigned int opt);
1117
extern void blk_set_queue_depth(struct request_queue *q, unsigned int depth);
1118
extern void blk_set_default_limits(struct queue_limits *lim);
1119
extern void blk_set_stacking_limits(struct queue_limits *lim);
1120 1121
extern int blk_stack_limits(struct queue_limits *t, struct queue_limits *b,
			    sector_t offset);
1122 1123
extern int bdev_stack_limits(struct queue_limits *t, struct block_device *bdev,
			    sector_t offset);
1124 1125
extern void disk_stack_limits(struct gendisk *disk, struct block_device *bdev,
			      sector_t offset);
1126
extern void blk_queue_stack_limits(struct request_queue *t, struct request_queue *b);
1127
extern void blk_queue_dma_pad(struct request_queue *, unsigned int);
1128
extern void blk_queue_update_dma_pad(struct request_queue *, unsigned int);
1129 1130 1131
extern int blk_queue_dma_drain(struct request_queue *q,
			       dma_drain_needed_fn *dma_drain_needed,
			       void *buf, unsigned int size);
1132
extern void blk_queue_lld_busy(struct request_queue *q, lld_busy_fn *fn);
1133
extern void blk_queue_segment_boundary(struct request_queue *, unsigned long);
1134
extern void blk_queue_virt_boundary(struct request_queue *, unsigned long);
1135
extern void blk_queue_prep_rq(struct request_queue *, prep_rq_fn *pfn);
1136
extern void blk_queue_unprep_rq(struct request_queue *, unprep_rq_fn *ufn);
1137
extern void blk_queue_dma_alignment(struct request_queue *, int);
1138
extern void blk_queue_update_dma_alignment(struct request_queue *, int);
1139
extern void blk_queue_softirq_done(struct request_queue *, softirq_done_fn *);
J
Jens Axboe 已提交
1140 1141
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);
1142
extern void blk_queue_flush_queueable(struct request_queue *q, bool queueable);
1143
extern void blk_queue_write_cache(struct request_queue *q, bool enabled, bool fua);
L
Linus Torvalds 已提交
1144 1145
extern struct backing_dev_info *blk_get_backing_dev_info(struct block_device *bdev);

1146
extern int blk_rq_map_sg(struct request_queue *, struct request *, struct scatterlist *);
L
Linus Torvalds 已提交
1147 1148 1149
extern void blk_dump_rq_flags(struct request *, char *);
extern long nr_blockdev_pages(void);

T
Tejun Heo 已提交
1150
bool __must_check blk_get_queue(struct request_queue *);
1151 1152 1153
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 *);
1154
extern void blk_set_queue_dying(struct request_queue *);
L
Linus Torvalds 已提交
1155

L
Lin Ming 已提交
1156 1157 1158
/*
 * block layer runtime pm functions
 */
1159
#ifdef CONFIG_PM
L
Lin Ming 已提交
1160 1161 1162 1163 1164
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);
1165
extern void blk_set_runtime_active(struct request_queue *q);
L
Lin Ming 已提交
1166 1167 1168 1169 1170 1171 1172 1173 1174 1175
#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) {}
1176
extern inline void blk_set_runtime_active(struct request_queue *q) {}
L
Lin Ming 已提交
1177 1178
#endif

1179
/*
S
Suresh Jayaraman 已提交
1180 1181 1182 1183 1184 1185 1186 1187 1188 1189
 * 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.
1190
 */
1191
struct blk_plug {
S
Suresh Jayaraman 已提交
1192
	struct list_head list; /* requests */
1193
	struct list_head mq_list; /* blk-mq requests */
S
Suresh Jayaraman 已提交
1194
	struct list_head cb_list; /* md requires an unplug callback */
1195
};
1196
#define BLK_MAX_REQUEST_COUNT 16
1197
#define BLK_PLUG_FLUSH_SIZE (128 * 1024)
1198

1199
struct blk_plug_cb;
1200
typedef void (*blk_plug_cb_fn)(struct blk_plug_cb *, bool);
1201 1202
struct blk_plug_cb {
	struct list_head list;
1203 1204
	blk_plug_cb_fn callback;
	void *data;
1205
};
1206 1207
extern struct blk_plug_cb *blk_check_plugged(blk_plug_cb_fn unplug,
					     void *data, int size);
1208 1209
extern void blk_start_plug(struct blk_plug *);
extern void blk_finish_plug(struct blk_plug *);
1210
extern void blk_flush_plug_list(struct blk_plug *, bool);
1211 1212 1213 1214 1215

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

1216 1217 1218 1219 1220 1221 1222 1223
	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;

1224
	if (plug)
1225
		blk_flush_plug_list(plug, true);
1226 1227 1228 1229 1230 1231
}

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

1232 1233 1234 1235
	return plug &&
		(!list_empty(&plug->list) ||
		 !list_empty(&plug->mq_list) ||
		 !list_empty(&plug->cb_list));
1236 1237
}

L
Linus Torvalds 已提交
1238 1239 1240
/*
 * tag stuff
 */
1241 1242 1243
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 *);
1244
extern int blk_queue_init_tags(struct request_queue *, int, struct blk_queue_tag *, int);
1245 1246 1247
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 *);
1248
extern struct blk_queue_tag *blk_init_tags(int, int);
1249
extern void blk_free_tags(struct blk_queue_tag *);
L
Linus Torvalds 已提交
1250

1251 1252 1253 1254 1255 1256 1257
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];
}
1258

1259 1260 1261

#define BLKDEV_DISCARD_SECURE	(1 << 0)	/* issue a secure erase */
#define BLKDEV_DISCARD_ZERO	(1 << 1)	/* must reliably zero data */
1262 1263

extern int blkdev_issue_flush(struct block_device *, gfp_t, sector_t *);
1264 1265
extern int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
		sector_t nr_sects, gfp_t gfp_mask, unsigned long flags);
1266
extern int __blkdev_issue_discard(struct block_device *bdev, sector_t sector,
1267
		sector_t nr_sects, gfp_t gfp_mask, int flags,
1268
		struct bio **biop);
1269 1270
extern int blkdev_issue_write_same(struct block_device *bdev, sector_t sector,
		sector_t nr_sects, gfp_t gfp_mask, struct page *page);
1271
extern int blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
1272
		sector_t nr_sects, gfp_t gfp_mask, bool discard);
1273 1274
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
1275
{
1276 1277 1278
	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
1279
}
1280
static inline int sb_issue_zeroout(struct super_block *sb, sector_t block,
1281
		sector_t nr_blocks, gfp_t gfp_mask)
1282 1283 1284 1285
{
	return blkdev_issue_zeroout(sb->s_bdev,
				    block << (sb->s_blocksize_bits - 9),
				    nr_blocks << (sb->s_blocksize_bits - 9),
1286
				    gfp_mask, true);
1287
}
L
Linus Torvalds 已提交
1288

1289
extern int blk_verify_command(unsigned char *cmd, fmode_t has_write_perm);
1290

1291 1292 1293
enum blk_default_limits {
	BLK_MAX_SEGMENTS	= 128,
	BLK_SAFE_MAX_SECTORS	= 255,
1294
	BLK_DEF_MAX_SECTORS	= 2560,
1295 1296 1297
	BLK_MAX_SEGMENT_SIZE	= 65536,
	BLK_SEG_BOUNDARY_MASK	= 0xFFFFFFFFUL,
};
1298

L
Linus Torvalds 已提交
1299 1300
#define blkdev_entry_to_request(entry) list_entry((entry), struct request, queuelist)

1301 1302
static inline unsigned long queue_bounce_pfn(struct request_queue *q)
{
1303
	return q->limits.bounce_pfn;
1304 1305 1306 1307
}

static inline unsigned long queue_segment_boundary(struct request_queue *q)
{
1308
	return q->limits.seg_boundary_mask;
1309 1310
}

1311 1312 1313 1314 1315
static inline unsigned long queue_virt_boundary(struct request_queue *q)
{
	return q->limits.virt_boundary_mask;
}

1316 1317
static inline unsigned int queue_max_sectors(struct request_queue *q)
{
1318
	return q->limits.max_sectors;
1319 1320 1321 1322
}

static inline unsigned int queue_max_hw_sectors(struct request_queue *q)
{
1323
	return q->limits.max_hw_sectors;
1324 1325
}

1326
static inline unsigned short queue_max_segments(struct request_queue *q)
1327
{
1328
	return q->limits.max_segments;
1329 1330 1331 1332
}

static inline unsigned int queue_max_segment_size(struct request_queue *q)
{
1333
	return q->limits.max_segment_size;
1334 1335
}

1336
static inline unsigned short queue_logical_block_size(struct request_queue *q)
L
Linus Torvalds 已提交
1337 1338 1339
{
	int retval = 512;

1340 1341
	if (q && q->limits.logical_block_size)
		retval = q->limits.logical_block_size;
L
Linus Torvalds 已提交
1342 1343 1344 1345

	return retval;
}

1346
static inline unsigned short bdev_logical_block_size(struct block_device *bdev)
L
Linus Torvalds 已提交
1347
{
1348
	return queue_logical_block_size(bdev_get_queue(bdev));
L
Linus Torvalds 已提交
1349 1350
}

1351 1352 1353 1354 1355
static inline unsigned int queue_physical_block_size(struct request_queue *q)
{
	return q->limits.physical_block_size;
}

1356
static inline unsigned int bdev_physical_block_size(struct block_device *bdev)
M
Martin K. Petersen 已提交
1357 1358 1359 1360
{
	return queue_physical_block_size(bdev_get_queue(bdev));
}

1361 1362 1363 1364 1365
static inline unsigned int queue_io_min(struct request_queue *q)
{
	return q->limits.io_min;
}

M
Martin K. Petersen 已提交
1366 1367 1368 1369 1370
static inline int bdev_io_min(struct block_device *bdev)
{
	return queue_io_min(bdev_get_queue(bdev));
}

1371 1372 1373 1374 1375
static inline unsigned int queue_io_opt(struct request_queue *q)
{
	return q->limits.io_opt;
}

M
Martin K. Petersen 已提交
1376 1377 1378 1379 1380
static inline int bdev_io_opt(struct block_device *bdev)
{
	return queue_io_opt(bdev_get_queue(bdev));
}

1381 1382
static inline int queue_alignment_offset(struct request_queue *q)
{
M
Martin K. Petersen 已提交
1383
	if (q->limits.misaligned)
1384 1385
		return -1;

M
Martin K. Petersen 已提交
1386
	return q->limits.alignment_offset;
1387 1388
}

1389
static inline int queue_limit_alignment_offset(struct queue_limits *lim, sector_t sector)
1390 1391
{
	unsigned int granularity = max(lim->physical_block_size, lim->io_min);
1392
	unsigned int alignment = sector_div(sector, granularity >> 9) << 9;
1393

1394
	return (granularity + lim->alignment_offset - alignment) % granularity;
1395 1396
}

M
Martin K. Petersen 已提交
1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409
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;
}

1410 1411 1412 1413 1414 1415 1416 1417
static inline int queue_discard_alignment(struct request_queue *q)
{
	if (q->limits.discard_misaligned)
		return -1;

	return q->limits.discard_alignment;
}

1418
static inline int queue_limit_discard_alignment(struct queue_limits *lim, sector_t sector)
1419
{
1420
	unsigned int alignment, granularity, offset;
1421

1422 1423 1424
	if (!lim->max_discard_sectors)
		return 0;

1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438
	/* 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;
1439 1440
}

1441 1442 1443 1444 1445 1446 1447 1448 1449 1450
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;
}

1451 1452
static inline unsigned int queue_discard_zeroes_data(struct request_queue *q)
{
1453
	if (q->limits.max_discard_sectors && q->limits.discard_zeroes_data == 1)
1454 1455 1456 1457 1458 1459 1460 1461 1462 1463
		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));
}

1464 1465 1466 1467 1468 1469 1470 1471 1472 1473
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;
}

D
Damien Le Moal 已提交
1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493
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;
}

1494 1495 1496 1497 1498 1499 1500 1501 1502 1503
static inline unsigned int bdev_zone_size(struct block_device *bdev)
{
	struct request_queue *q = bdev_get_queue(bdev);

	if (q)
		return blk_queue_zone_size(q);

	return 0;
}

1504
static inline int queue_dma_alignment(struct request_queue *q)
L
Linus Torvalds 已提交
1505
{
1506
	return q ? q->dma_alignment : 511;
L
Linus Torvalds 已提交
1507 1508
}

1509
static inline int blk_rq_aligned(struct request_queue *q, unsigned long addr,
1510 1511 1512
				 unsigned int len)
{
	unsigned int alignment = queue_dma_alignment(q) | q->dma_pad_mask;
1513
	return !(addr & alignment) && !(len & alignment);
1514 1515
}

L
Linus Torvalds 已提交
1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526
/* 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;
}

1527
static inline unsigned int block_size(struct block_device *bdev)
L
Linus Torvalds 已提交
1528 1529 1530 1531
{
	return bdev->bd_block_size;
}

1532 1533
static inline bool queue_flush_queueable(struct request_queue *q)
{
J
Jens Axboe 已提交
1534
	return !test_bit(QUEUE_FLAG_FLUSH_NQ, &q->queue_flags);
1535 1536
}

L
Linus Torvalds 已提交
1537 1538 1539 1540 1541 1542
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)
{
1543
	put_page(p.v);
L
Linus Torvalds 已提交
1544 1545
}

1546 1547 1548 1549 1550 1551 1552
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));
}

1553 1554 1555 1556 1557 1558 1559 1560 1561
/*
 * 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;
1562
	return __bvec_gap_to_prev(q, bprv, offset);
1563 1564
}

1565 1566 1567
static inline bool bio_will_gap(struct request_queue *q, struct bio *prev,
			 struct bio *next)
{
1568 1569 1570 1571 1572
	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);
1573

1574 1575 1576 1577
		return __bvec_gap_to_prev(q, &pb, nb.bv_offset);
	}

	return false;
1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589
}

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);
}

1590
int kblockd_schedule_work(struct work_struct *work);
1591
int kblockd_schedule_work_on(int cpu, struct work_struct *work);
1592
int kblockd_schedule_delayed_work(struct delayed_work *dwork, unsigned long delay);
1593
int kblockd_schedule_delayed_work_on(int cpu, struct delayed_work *dwork, unsigned long delay);
L
Linus Torvalds 已提交
1594

1595
#ifdef CONFIG_BLK_CGROUP
1596 1597 1598 1599 1600
/*
 * 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().
 */
1601 1602
static inline void set_start_time_ns(struct request *req)
{
1603
	preempt_disable();
1604
	req->start_time_ns = sched_clock();
1605
	preempt_enable();
1606 1607 1608 1609
}

static inline void set_io_start_time_ns(struct request *req)
{
1610
	preempt_disable();
1611
	req->io_start_time_ns = sched_clock();
1612
	preempt_enable();
1613
}
1614 1615 1616 1617 1618 1619 1620 1621 1622 1623

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;
}
1624 1625 1626
#else
static inline void set_start_time_ns(struct request *req) {}
static inline void set_io_start_time_ns(struct request *req) {}
1627 1628 1629 1630 1631 1632 1633 1634
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;
}
1635 1636
#endif

L
Linus Torvalds 已提交
1637 1638 1639 1640 1641
#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) "-*")

1642 1643
#if defined(CONFIG_BLK_DEV_INTEGRITY)

1644 1645 1646
enum blk_integrity_flags {
	BLK_INTEGRITY_VERIFY		= 1 << 0,
	BLK_INTEGRITY_GENERATE		= 1 << 1,
1647
	BLK_INTEGRITY_DEVICE_CAPABLE	= 1 << 2,
1648
	BLK_INTEGRITY_IP_CHECKSUM	= 1 << 3,
1649
};
1650

1651
struct blk_integrity_iter {
1652 1653
	void			*prot_buf;
	void			*data_buf;
1654
	sector_t		seed;
1655
	unsigned int		data_size;
1656
	unsigned short		interval;
1657 1658 1659
	const char		*disk_name;
};

1660
typedef int (integrity_processing_fn) (struct blk_integrity_iter *);
1661

1662 1663 1664 1665 1666
struct blk_integrity_profile {
	integrity_processing_fn		*generate_fn;
	integrity_processing_fn		*verify_fn;
	const char			*name;
};
1667

1668
extern void blk_integrity_register(struct gendisk *, struct blk_integrity *);
1669
extern void blk_integrity_unregister(struct gendisk *);
1670
extern int blk_integrity_compare(struct gendisk *, struct gendisk *);
1671 1672 1673
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 *);
1674 1675 1676 1677
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 *);
1678

1679
static inline struct blk_integrity *blk_get_integrity(struct gendisk *disk)
1680
{
1681
	struct blk_integrity *bi = &disk->queue->integrity;
1682 1683 1684 1685 1686

	if (!bi->profile)
		return NULL;

	return bi;
1687 1688
}

1689 1690
static inline
struct blk_integrity *bdev_get_integrity(struct block_device *bdev)
1691
{
1692
	return blk_get_integrity(bdev->bd_disk);
1693 1694
}

1695
static inline bool blk_integrity_rq(struct request *rq)
1696
{
1697
	return rq->cmd_flags & REQ_INTEGRITY;
1698 1699
}

1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711
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;
}

1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731
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);
}

1732 1733
#else /* CONFIG_BLK_DEV_INTEGRITY */

1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755
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)
{
1756
	return NULL;
1757 1758 1759 1760 1761 1762 1763 1764 1765
}
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;
}
1766
static inline void blk_integrity_register(struct gendisk *d,
1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780
					 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;
}
1781 1782 1783
static inline bool blk_integrity_merge_rq(struct request_queue *rq,
					  struct request *r1,
					  struct request *r2)
1784
{
1785
	return true;
1786
}
1787 1788 1789
static inline bool blk_integrity_merge_bio(struct request_queue *rq,
					   struct request *r,
					   struct bio *b)
1790
{
1791
	return true;
1792
}
1793

1794 1795 1796 1797 1798 1799 1800 1801 1802 1803
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;
}
1804 1805 1806

#endif /* CONFIG_BLK_DEV_INTEGRITY */

1807 1808 1809 1810 1811 1812 1813 1814 1815
/**
 * 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
1816
	void *addr;
1817
	long size;
D
Dan Williams 已提交
1818
	pfn_t pfn;
1819 1820
};

1821
struct block_device_operations {
A
Al Viro 已提交
1822
	int (*open) (struct block_device *, fmode_t);
1823
	void (*release) (struct gendisk *, fmode_t);
1824
	int (*rw_page)(struct block_device *, sector_t, struct page *, bool);
A
Al Viro 已提交
1825 1826
	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
1827 1828
	long (*direct_access)(struct block_device *, sector_t, void **, pfn_t *,
			long);
1829 1830 1831
	unsigned int (*check_events) (struct gendisk *disk,
				      unsigned int clearing);
	/* ->media_changed() is DEPRECATED, use ->check_events() instead */
1832
	int (*media_changed) (struct gendisk *);
1833
	void (*unlock_native_capacity) (struct gendisk *);
1834 1835
	int (*revalidate_disk) (struct gendisk *);
	int (*getgeo)(struct block_device *, struct hd_geometry *);
1836 1837
	/* this callback is with swap_lock and sometimes page table lock held */
	void (*swap_slot_free_notify) (struct block_device *, unsigned long);
1838
	struct module *owner;
1839
	const struct pr_ops *pr_ops;
1840 1841
};

1842 1843
extern int __blkdev_driver_ioctl(struct block_device *, fmode_t, unsigned int,
				 unsigned long);
1844 1845 1846
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 *);
1847
extern long bdev_direct_access(struct block_device *, struct blk_dax_ctl *);
1848
extern int bdev_dax_supported(struct super_block *, int);
1849
extern bool bdev_dax_capable(struct block_device *);
1850
#else /* CONFIG_BLOCK */
1851 1852 1853

struct block_device;

1854 1855 1856 1857 1858 1859 1860 1861 1862 1863
/*
 * stubs for when the block layer is configured out
 */
#define buffer_heads_over_limit 0

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

1864 1865 1866 1867
struct blk_plug {
};

static inline void blk_start_plug(struct blk_plug *plug)
1868 1869 1870
{
}

1871
static inline void blk_finish_plug(struct blk_plug *plug)
1872 1873 1874
{
}

1875
static inline void blk_flush_plug(struct task_struct *task)
1876 1877 1878
{
}

1879 1880 1881 1882 1883
static inline void blk_schedule_flush_plug(struct task_struct *task)
{
}


1884 1885 1886 1887 1888
static inline bool blk_needs_flush_plug(struct task_struct *tsk)
{
	return false;
}

1889 1890 1891 1892 1893 1894
static inline int blkdev_issue_flush(struct block_device *bdev, gfp_t gfp_mask,
				     sector_t *error_sector)
{
	return 0;
}

1895 1896
#endif /* CONFIG_BLOCK */

L
Linus Torvalds 已提交
1897
#endif