blkdev.h 47.5 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 15 16 17 18 19
#include <linux/timer.h>
#include <linux/workqueue.h>
#include <linux/pagemap.h>
#include <linux/backing-dev.h>
#include <linux/wait.h>
#include <linux/mempool.h>
#include <linux/bio.h>
#include <linux/stringify.h>
20
#include <linux/gfp.h>
21
#include <linux/bsg.h>
22
#include <linux/smp.h>
T
Tejun Heo 已提交
23
#include <linux/rcupdate.h>
24
#include <linux/percpu-refcount.h>
L
Linus Torvalds 已提交
25 26 27

#include <asm/scatterlist.h>

28
struct module;
29 30
struct scsi_ioctl_command;

L
Linus Torvalds 已提交
31 32 33
struct request_queue;
struct elevator_queue;
struct request_pm_state;
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;
L
Linus Torvalds 已提交
40 41 42 43

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

T
Tejun Heo 已提交
44 45 46 47 48 49
/*
 * Maximum number of blkcg policies allowed to be registered concurrently.
 * Defined here to simplify include dependency.
 */
#define BLKCG_MAX_POLS		2

L
Linus Torvalds 已提交
50
struct request;
51
typedef void (rq_end_io_fn)(struct request *, int);
L
Linus Torvalds 已提交
52

53 54 55
#define BLK_RL_SYNCFULL		(1U << 0)
#define BLK_RL_ASYNCFULL	(1U << 1)

L
Linus Torvalds 已提交
56
struct request_list {
57
	struct request_queue	*q;	/* the queue this rl belongs to */
58 59 60
#ifdef CONFIG_BLK_CGROUP
	struct blkcg_gq		*blkg;	/* blkg this request pool belongs to */
#endif
61 62 63 64
	/*
	 * count[], starved[], and wait[] are indexed by
	 * BLK_RW_SYNC/BLK_RW_ASYNC
	 */
65 66 67 68
	int			count[2];
	int			starved[2];
	mempool_t		*rq_pool;
	wait_queue_head_t	wait[2];
69
	unsigned int		flags;
L
Linus Torvalds 已提交
70 71
};

72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88
/*
 * request command types
 */
enum rq_cmd_type_bits {
	REQ_TYPE_FS		= 1,	/* fs request */
	REQ_TYPE_BLOCK_PC,		/* scsi command */
	REQ_TYPE_SENSE,			/* sense request */
	REQ_TYPE_PM_SUSPEND,		/* suspend request */
	REQ_TYPE_PM_RESUME,		/* resume request */
	REQ_TYPE_PM_SHUTDOWN,		/* shutdown request */
	REQ_TYPE_SPECIAL,		/* driver defined type */
	/*
	 * for ATA/ATAPI devices. this really doesn't belong here, ide should
	 * use REQ_TYPE_SPECIAL and use rq->cmd[0] with the range of driver
	 * private REQ_LB opcodes to differentiate what type of request this is
	 */
	REQ_TYPE_ATA_TASKFILE,
89
	REQ_TYPE_ATA_PC,
90 91
};

L
Linus Torvalds 已提交
92 93 94
#define BLK_MAX_CDB	16

/*
95 96 97 98
 * 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
99 100
 */
struct request {
101
	struct list_head queuelist;
102 103
	union {
		struct call_single_data csd;
104
		unsigned long fifo_time;
105
	};
106

107
	struct request_queue *q;
108
	struct blk_mq_ctx *mq_ctx;
109

J
Jens Axboe 已提交
110
	u64 cmd_flags;
111
	enum rq_cmd_type_bits cmd_type;
J
Jens Axboe 已提交
112
	unsigned long atomic_flags;
L
Linus Torvalds 已提交
113

114 115
	int cpu;

116 117
	/* the following two fields are internal, NEVER access directly */
	unsigned int __data_len;	/* total data len */
118
	sector_t __sector;		/* sector cursor */
L
Linus Torvalds 已提交
119 120 121 122

	struct bio *bio;
	struct bio *biotail;

123 124 125 126 127 128 129 130 131 132 133 134
	/*
	 * 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;
	};

135 136 137
	/*
	 * The rb_node is only used inside the io scheduler, requests
	 * are pruned when moved to the dispatch queue. So let the
138
	 * completion_data share space with the rb_node.
139 140 141
	 */
	union {
		struct rb_node rb_node;	/* sort/lookup */
142
		void *completion_data;
143
	};
144

145
	/*
146
	 * Three pointers are available for the IO schedulers, if they need
147 148
	 * more they have to dynamically allocate it.  Flush requests are
	 * never put on the IO scheduler. So let the flush fields share
149
	 * space with the elevator data.
150
	 */
151
	union {
152 153 154 155 156
		struct {
			struct io_cq		*icq;
			void			*priv[2];
		} elv;

157 158 159
		struct {
			unsigned int		seq;
			struct list_head	list;
160
			rq_end_io_fn		*saved_end_io;
161 162
		} flush;
	};
163

164
	struct gendisk *rq_disk;
165
	struct hd_struct *part;
L
Linus Torvalds 已提交
166
	unsigned long start_time;
167
#ifdef CONFIG_BLK_CGROUP
168
	struct request_list *rl;		/* rl this rq is alloced from */
169 170 171
	unsigned long long start_time_ns;
	unsigned long long io_start_time_ns;    /* when passed to hardware */
#endif
L
Linus Torvalds 已提交
172 173 174 175
	/* Number of scatter-gather DMA addr+len pairs after
	 * physical address coalescing is performed.
	 */
	unsigned short nr_phys_segments;
176 177 178
#if defined(CONFIG_BLK_DEV_INTEGRITY)
	unsigned short nr_integrity_segments;
#endif
L
Linus Torvalds 已提交
179

180 181
	unsigned short ioprio;

T
Tejun Heo 已提交
182
	void *special;		/* opaque pointer available for LLD use */
L
Linus Torvalds 已提交
183

184 185 186
	int tag;
	int errors;

L
Linus Torvalds 已提交
187 188 189
	/*
	 * when request is used as a packet command carrier
	 */
190 191
	unsigned char __cmd[BLK_MAX_CDB];
	unsigned char *cmd;
192
	unsigned short cmd_len;
L
Linus Torvalds 已提交
193

194
	unsigned int extra_len;	/* length of alignment and padding */
L
Linus Torvalds 已提交
195
	unsigned int sense_len;
T
Tejun Heo 已提交
196
	unsigned int resid_len;	/* residual count */
L
Linus Torvalds 已提交
197 198
	void *sense;

J
Jens Axboe 已提交
199 200
	unsigned long deadline;
	struct list_head timeout_list;
L
Linus Torvalds 已提交
201
	unsigned int timeout;
202
	int retries;
L
Linus Torvalds 已提交
203 204

	/*
205
	 * completion callback.
L
Linus Torvalds 已提交
206 207 208
	 */
	rq_end_io_fn *end_io;
	void *end_io_data;
209 210 211

	/* for bidi */
	struct request *next_rq;
L
Linus Torvalds 已提交
212 213
};

214 215 216 217 218
static inline unsigned short req_get_ioprio(struct request *req)
{
	return req->ioprio;
}

L
Linus Torvalds 已提交
219
/*
220
 * State information carried for REQ_TYPE_PM_SUSPEND and REQ_TYPE_PM_RESUME
L
Linus Torvalds 已提交
221 222 223 224 225 226 227 228 229 230 231 232 233
 * requests. Some step values could eventually be made generic.
 */
struct request_pm_state
{
	/* PM state machine step value, currently driver specific */
	int	pm_step;
	/* requested PM state value (S1, S2, S3, S4, ...) */
	u32	pm_state;
	void*	data;		/* for driver use */
};

#include <linux/elevator.h>

234 235
struct blk_queue_ctx;

236
typedef void (request_fn_proc) (struct request_queue *q);
237
typedef void (make_request_fn) (struct request_queue *q, struct bio *bio);
238
typedef int (prep_rq_fn) (struct request_queue *, struct request *);
239
typedef void (unprep_rq_fn) (struct request_queue *, struct request *);
L
Linus Torvalds 已提交
240 241

struct bio_vec;
242 243 244 245 246 247 248 249
struct bvec_merge_data {
	struct block_device *bi_bdev;
	sector_t bi_sector;
	unsigned bi_size;
	unsigned long bi_rw;
};
typedef int (merge_bvec_fn) (struct request_queue *, struct bvec_merge_data *,
			     struct bio_vec *);
250
typedef void (softirq_done_fn)(struct request *);
251
typedef int (dma_drain_needed_fn)(struct request *);
252
typedef int (lld_busy_fn) (struct request_queue *q);
M
Mike Christie 已提交
253
typedef int (bsg_job_fn) (struct bsg_job *);
L
Linus Torvalds 已提交
254

J
Jens Axboe 已提交
255 256 257 258 259 260 261 262
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
263 264 265 266 267 268 269 270 271 272
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 */
273
	int real_max_depth;		/* what the array can hold */
L
Linus Torvalds 已提交
274
	atomic_t refcnt;		/* map can be shared */
275 276
	int alloc_policy;		/* tag allocation policy */
	int next_tag;			/* next tag */
L
Linus Torvalds 已提交
277
};
278 279
#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
280

281 282 283
#define BLK_SCSI_MAX_CMDS	(256)
#define BLK_SCSI_CMD_PER_LONG	(BLK_SCSI_MAX_CMDS / (sizeof(long) * 8))

284 285 286 287 288
struct queue_limits {
	unsigned long		bounce_pfn;
	unsigned long		seg_boundary_mask;

	unsigned int		max_hw_sectors;
289
	unsigned int		chunk_sectors;
290 291
	unsigned int		max_sectors;
	unsigned int		max_segment_size;
292 293 294 295
	unsigned int		physical_block_size;
	unsigned int		alignment_offset;
	unsigned int		io_min;
	unsigned int		io_opt;
296
	unsigned int		max_discard_sectors;
297
	unsigned int		max_write_same_sectors;
298 299
	unsigned int		discard_granularity;
	unsigned int		discard_alignment;
300 301

	unsigned short		logical_block_size;
302
	unsigned short		max_segments;
303
	unsigned short		max_integrity_segments;
304

305
	unsigned char		misaligned;
306
	unsigned char		discard_misaligned;
307
	unsigned char		cluster;
308
	unsigned char		discard_zeroes_data;
309
	unsigned char		raid_partial_stripes_expensive;
310 311
};

312
struct request_queue {
L
Linus Torvalds 已提交
313 314 315 316 317
	/*
	 * Together with queue_head for cacheline sharing
	 */
	struct list_head	queue_head;
	struct request		*last_merge;
J
Jens Axboe 已提交
318
	struct elevator_queue	*elevator;
319 320
	int			nr_rqs[2];	/* # allocated [a]sync rqs */
	int			nr_rqs_elvpriv;	/* # allocated rqs w/ elvpriv */
L
Linus Torvalds 已提交
321 322

	/*
323 324 325 326
	 * 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
327
	 */
328
	struct request_list	root_rl;
L
Linus Torvalds 已提交
329 330 331 332

	request_fn_proc		*request_fn;
	make_request_fn		*make_request_fn;
	prep_rq_fn		*prep_rq_fn;
333
	unprep_rq_fn		*unprep_rq_fn;
L
Linus Torvalds 已提交
334
	merge_bvec_fn		*merge_bvec_fn;
335
	softirq_done_fn		*softirq_done_fn;
J
Jens Axboe 已提交
336
	rq_timed_out_fn		*rq_timed_out_fn;
337
	dma_drain_needed_fn	*dma_drain_needed;
338
	lld_busy_fn		*lld_busy_fn;
L
Linus Torvalds 已提交
339

340 341 342 343 344
	struct blk_mq_ops	*mq_ops;

	unsigned int		*mq_map;

	/* sw queues */
345
	struct blk_mq_ctx __percpu	*queue_ctx;
346 347 348 349 350 351
	unsigned int		nr_queues;

	/* hw dispatch queues */
	struct blk_mq_hw_ctx	**queue_hw_ctx;
	unsigned int		nr_hw_queues;

352 353 354
	/*
	 * Dispatch queue sorting
	 */
J
Jens Axboe 已提交
355
	sector_t		end_sector;
356 357
	struct request		*boundary_rq;

L
Linus Torvalds 已提交
358
	/*
359
	 * Delayed queue handling
L
Linus Torvalds 已提交
360
	 */
361
	struct delayed_work	delay_work;
L
Linus Torvalds 已提交
362 363 364 365 366 367 368 369 370 371

	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;

	/*
372
	 * various queue flags, see QUEUE_* below
L
Linus Torvalds 已提交
373
	 */
374
	unsigned long		queue_flags;
L
Linus Torvalds 已提交
375

376 377 378 379 380 381
	/*
	 * ida allocated id for this queue.  Used to index queues from
	 * ioctx.
	 */
	int			id;

L
Linus Torvalds 已提交
382
	/*
383
	 * queue needs bounce pages for pages above this limit
L
Linus Torvalds 已提交
384
	 */
385
	gfp_t			bounce_gfp;
L
Linus Torvalds 已提交
386 387

	/*
已提交
388 389 390
	 * 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
391
	 */
已提交
392
	spinlock_t		__queue_lock;
L
Linus Torvalds 已提交
393 394 395 396 397 398 399
	spinlock_t		*queue_lock;

	/*
	 * queue kobject
	 */
	struct kobject kobj;

400 401 402 403 404
	/*
	 * mq queue kobject
	 */
	struct kobject mq_kobj;

405
#ifdef CONFIG_PM
L
Lin Ming 已提交
406 407 408 409 410
	struct device		*dev;
	int			rpm_status;
	unsigned int		nr_pending;
#endif

L
Linus Torvalds 已提交
411 412 413 414 415 416 417 418
	/*
	 * 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
419
	unsigned int		dma_drain_size;
420
	void			*dma_drain_buffer;
421
	unsigned int		dma_pad_mask;
L
Linus Torvalds 已提交
422 423 424
	unsigned int		dma_alignment;

	struct blk_queue_tag	*queue_tags;
425
	struct list_head	tag_busy_list;
L
Linus Torvalds 已提交
426

427
	unsigned int		nr_sorted;
428
	unsigned int		in_flight[2];
429 430 431 432 433 434
	/*
	 * 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
435

J
Jens Axboe 已提交
436 437 438 439
	unsigned int		rq_timeout;
	struct timer_list	timeout;
	struct list_head	timeout_list;

440
	struct list_head	icq_list;
441
#ifdef CONFIG_BLK_CGROUP
442
	DECLARE_BITMAP		(blkcg_pols, BLKCG_MAX_POLS);
T
Tejun Heo 已提交
443
	struct blkcg_gq		*root_blkg;
444
	struct list_head	blkg_list;
445
#endif
446

447 448
	struct queue_limits	limits;

L
Linus Torvalds 已提交
449 450 451 452 453
	/*
	 * sg stuff
	 */
	unsigned int		sg_timeout;
	unsigned int		sg_reserved_size;
454
	int			node;
455
#ifdef CONFIG_BLK_DEV_IO_TRACE
456
	struct blk_trace	*blk_trace;
457
#endif
L
Linus Torvalds 已提交
458
	/*
459
	 * for flush operations
L
Linus Torvalds 已提交
460
	 */
461
	unsigned int		flush_flags;
462
	unsigned int		flush_not_queueable:1;
463
	struct blk_flush_queue	*fq;
464

465 466 467 468
	struct list_head	requeue_list;
	spinlock_t		requeue_lock;
	struct work_struct	requeue_work;

469
	struct mutex		sysfs_lock;
470

471
	int			bypass_depth;
472
	int			mq_freeze_depth;
473

474
#if defined(CONFIG_BLK_DEV_BSG)
M
Mike Christie 已提交
475 476
	bsg_job_fn		*bsg_job_fn;
	int			bsg_job_size;
477 478
	struct bsg_class_device bsg_dev;
#endif
479 480 481 482 483

#ifdef CONFIG_BLK_DEV_THROTTLING
	/* Throttle data */
	struct throtl_data *td;
#endif
T
Tejun Heo 已提交
484
	struct rcu_head		rcu_head;
485
	wait_queue_head_t	mq_freeze_wq;
486
	struct percpu_ref	mq_usage_counter;
487
	struct list_head	all_q_node;
488 489 490

	struct blk_mq_tag_set	*tag_set;
	struct list_head	tag_set_list;
L
Linus Torvalds 已提交
491 492 493 494
};

#define QUEUE_FLAG_QUEUED	1	/* uses generic tag queueing */
#define QUEUE_FLAG_STOPPED	2	/* queue is stopped */
495 496
#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
497
#define QUEUE_FLAG_DYING	5	/* queue being torn down */
498
#define QUEUE_FLAG_BYPASS	6	/* act as dumb FIFO queue */
499 500
#define QUEUE_FLAG_BIDI		7	/* queue supports bidi requests */
#define QUEUE_FLAG_NOMERGES     8	/* disable merge attempts */
D
Dan Williams 已提交
501
#define QUEUE_FLAG_SAME_COMP	9	/* complete on same CPU-group */
502 503 504
#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) */
505
#define QUEUE_FLAG_VIRT        QUEUE_FLAG_NONROT /* paravirt device */
506 507 508 509 510
#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 */
#define QUEUE_FLAG_SECDISCARD  17	/* supports SECDISCARD */
D
Dan Williams 已提交
511
#define QUEUE_FLAG_SAME_FORCE  18	/* force complete on same CPU */
512
#define QUEUE_FLAG_DEAD        19	/* queue tear-down finished */
513
#define QUEUE_FLAG_INIT_DONE   20	/* queue is initialized */
514
#define QUEUE_FLAG_NO_SG_MERGE 21	/* don't attempt to merge SG segments*/
515
#define QUEUE_FLAG_SG_GAPS     22	/* queue doesn't support SG gaps */
516 517

#define QUEUE_FLAG_DEFAULT	((1 << QUEUE_FLAG_IO_STAT) |		\
518
				 (1 << QUEUE_FLAG_STACKABLE)	|	\
519 520
				 (1 << QUEUE_FLAG_SAME_COMP)	|	\
				 (1 << QUEUE_FLAG_ADD_RANDOM))
521

522
#define QUEUE_FLAG_MQ_DEFAULT	((1 << QUEUE_FLAG_IO_STAT) |		\
523
				 (1 << QUEUE_FLAG_STACKABLE)	|	\
524 525
				 (1 << QUEUE_FLAG_SAME_COMP))

526
static inline void queue_lockdep_assert_held(struct request_queue *q)
527
{
528 529
	if (q->queue_lock)
		lockdep_assert_held(q->queue_lock);
530 531
}

N
Nick Piggin 已提交
532 533 534 535 536 537
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
538 539 540
static inline int queue_flag_test_and_clear(unsigned int flag,
					    struct request_queue *q)
{
541
	queue_lockdep_assert_held(q);
J
Jens Axboe 已提交
542 543 544 545 546 547 548 549 550 551 552 553

	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)
{
554
	queue_lockdep_assert_held(q);
J
Jens Axboe 已提交
555 556 557 558 559 560 561 562 563

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

	return 1;
}

N
Nick Piggin 已提交
564 565
static inline void queue_flag_set(unsigned int flag, struct request_queue *q)
{
566
	queue_lockdep_assert_held(q);
N
Nick Piggin 已提交
567 568 569 570 571 572 573 574 575
	__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);
}

576 577 578 579 580
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
581 582
static inline void queue_flag_clear(unsigned int flag, struct request_queue *q)
{
583
	queue_lockdep_assert_held(q);
N
Nick Piggin 已提交
584 585 586
	__clear_bit(flag, &q->queue_flags);
}

L
Linus Torvalds 已提交
587 588
#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
589
#define blk_queue_dying(q)	test_bit(QUEUE_FLAG_DYING, &(q)->queue_flags)
590
#define blk_queue_dead(q)	test_bit(QUEUE_FLAG_DEAD, &(q)->queue_flags)
591
#define blk_queue_bypass(q)	test_bit(QUEUE_FLAG_BYPASS, &(q)->queue_flags)
592
#define blk_queue_init_done(q)	test_bit(QUEUE_FLAG_INIT_DONE, &(q)->queue_flags)
593
#define blk_queue_nomerges(q)	test_bit(QUEUE_FLAG_NOMERGES, &(q)->queue_flags)
594 595
#define blk_queue_noxmerges(q)	\
	test_bit(QUEUE_FLAG_NOXMERGES, &(q)->queue_flags)
596
#define blk_queue_nonrot(q)	test_bit(QUEUE_FLAG_NONROT, &(q)->queue_flags)
597
#define blk_queue_io_stat(q)	test_bit(QUEUE_FLAG_IO_STAT, &(q)->queue_flags)
598
#define blk_queue_add_random(q)	test_bit(QUEUE_FLAG_ADD_RANDOM, &(q)->queue_flags)
599 600
#define blk_queue_stackable(q)	\
	test_bit(QUEUE_FLAG_STACKABLE, &(q)->queue_flags)
601
#define blk_queue_discard(q)	test_bit(QUEUE_FLAG_DISCARD, &(q)->queue_flags)
A
Adrian Hunter 已提交
602 603
#define blk_queue_secdiscard(q)	(blk_queue_discard(q) && \
	test_bit(QUEUE_FLAG_SECDISCARD, &(q)->queue_flags))
L
Linus Torvalds 已提交
604

605 606 607 608 609 610
#define blk_noretry_request(rq) \
	((rq)->cmd_flags & (REQ_FAILFAST_DEV|REQ_FAILFAST_TRANSPORT| \
			     REQ_FAILFAST_DRIVER))

#define blk_account_rq(rq) \
	(((rq)->cmd_flags & REQ_STARTED) && \
611
	 ((rq)->cmd_type == REQ_TYPE_FS))
612

L
Linus Torvalds 已提交
613
#define blk_pm_request(rq)	\
614 615
	((rq)->cmd_type == REQ_TYPE_PM_SUSPEND || \
	 (rq)->cmd_type == REQ_TYPE_PM_RESUME)
L
Linus Torvalds 已提交
616

617
#define blk_rq_cpu_valid(rq)	((rq)->cpu != -1)
618
#define blk_bidi_rq(rq)		((rq)->next_rq != NULL)
619 620
/* rq->queuelist of dequeued request must be list_empty() */
#define blk_queued_rq(rq)	(!list_empty(&(rq)->queuelist))
L
Linus Torvalds 已提交
621 622 623

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

J
Jens Axboe 已提交
624
#define rq_data_dir(rq)		(((rq)->cmd_flags & 1) != 0)
L
Linus Torvalds 已提交
625

626 627 628 629 630 631 632 633 634
/*
 * 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;
}

635 636 637 638 639
static inline unsigned int blk_queue_cluster(struct request_queue *q)
{
	return q->limits.cluster;
}

640
/*
641
 * We regard a request as sync, if either a read or a sync write
642
 */
643 644
static inline bool rw_is_sync(unsigned int rw_flags)
{
645
	return !(rw_flags & REQ_WRITE) || (rw_flags & REQ_SYNC);
646 647 648 649 650 651 652
}

static inline bool rq_is_sync(struct request *rq)
{
	return rw_is_sync(rq->cmd_flags);
}

653
static inline bool blk_rl_full(struct request_list *rl, bool sync)
L
Linus Torvalds 已提交
654
{
655 656 657
	unsigned int flag = sync ? BLK_RL_SYNCFULL : BLK_RL_ASYNCFULL;

	return rl->flags & flag;
L
Linus Torvalds 已提交
658 659
}

660
static inline void blk_set_rl_full(struct request_list *rl, bool sync)
L
Linus Torvalds 已提交
661
{
662 663 664
	unsigned int flag = sync ? BLK_RL_SYNCFULL : BLK_RL_ASYNCFULL;

	rl->flags |= flag;
L
Linus Torvalds 已提交
665 666
}

667
static inline void blk_clear_rl_full(struct request_list *rl, bool sync)
L
Linus Torvalds 已提交
668
{
669 670 671
	unsigned int flag = sync ? BLK_RL_SYNCFULL : BLK_RL_ASYNCFULL;

	rl->flags &= ~flag;
L
Linus Torvalds 已提交
672 673
}

674 675 676 677
static inline bool rq_mergeable(struct request *rq)
{
	if (rq->cmd_type != REQ_TYPE_FS)
		return false;
L
Linus Torvalds 已提交
678

679 680 681 682 683
	if (rq->cmd_flags & REQ_NOMERGE_FLAGS)
		return false;

	return true;
}
L
Linus Torvalds 已提交
684

685 686 687 688 689 690 691 692 693
static inline bool blk_check_merge_flags(unsigned int flags1,
					 unsigned int flags2)
{
	if ((flags1 & REQ_DISCARD) != (flags2 & REQ_DISCARD))
		return false;

	if ((flags1 & REQ_SECURE) != (flags2 & REQ_SECURE))
		return false;

694 695 696
	if ((flags1 & REQ_WRITE_SAME) != (flags2 & REQ_WRITE_SAME))
		return false;

697 698 699
	return true;
}

700 701 702 703 704 705 706 707
static inline bool blk_write_same_mergeable(struct bio *a, struct bio *b)
{
	if (bio_data(a) == bio_data(b))
		return true;

	return false;
}

L
Linus Torvalds 已提交
708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723
/*
 * q->prep_rq_fn return values
 */
#define BLKPREP_OK		0	/* serve it */
#define BLKPREP_KILL		1	/* fatal error, kill */
#define BLKPREP_DEFER		2	/* leave on queue */

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
 */
724 725

#if BITS_PER_LONG == 32
L
Linus Torvalds 已提交
726
#define BLK_BOUNCE_HIGH		((u64)blk_max_low_pfn << PAGE_SHIFT)
727 728 729 730
#else
#define BLK_BOUNCE_HIGH		-1ULL
#endif
#define BLK_BOUNCE_ANY		(-1ULL)
731
#define BLK_BOUNCE_ISA		(DMA_BIT_MASK(24))
L
Linus Torvalds 已提交
732

733 734 735 736
/*
 * default timeout for SG_IO if none specified
 */
#define BLK_DEFAULT_SG_TIMEOUT	(60 * HZ)
L
Linus Torvalds 已提交
737
#define BLK_MIN_SG_TIMEOUT	(7 * HZ)
738

739
#ifdef CONFIG_BOUNCE
L
Linus Torvalds 已提交
740
extern int init_emergency_isa_pool(void);
741
extern void blk_queue_bounce(struct request_queue *q, struct bio **bio);
L
Linus Torvalds 已提交
742 743 744 745 746
#else
static inline int init_emergency_isa_pool(void)
{
	return 0;
}
747
static inline void blk_queue_bounce(struct request_queue *q, struct bio **bio)
L
Linus Torvalds 已提交
748 749 750 751
{
}
#endif /* CONFIG_MMU */

752 753 754 755
struct rq_map_data {
	struct page **pages;
	int page_order;
	int nr_entries;
756
	unsigned long offset;
757
	int null_mapped;
758
	int from_user;
759 760
};

761
struct req_iterator {
762
	struct bvec_iter iter;
763 764 765 766
	struct bio *bio;
};

/* This should not be used directly - use rq_for_each_segment */
767 768
#define for_each_bio(_bio)		\
	for (; _bio; _bio = _bio->bi_next)
769
#define __rq_for_each_bio(_bio, rq)	\
L
Linus Torvalds 已提交
770 771 772
	if ((rq->bio))			\
		for (_bio = (rq)->bio; _bio; _bio = _bio->bi_next)

773 774
#define rq_for_each_segment(bvl, _rq, _iter)			\
	__rq_for_each_bio(_iter.bio, _rq)			\
775
		bio_for_each_segment(bvl, _iter.bio, _iter.iter)
776

K
Kent Overstreet 已提交
777
#define rq_iter_last(bvec, _iter)				\
778
		(_iter.bio->bi_next == NULL &&			\
K
Kent Overstreet 已提交
779
		 bio_iter_last(bvec, _iter.iter))
780

781 782 783 784 785 786 787 788 789 790 791
#ifndef ARCH_IMPLEMENTS_FLUSH_DCACHE_PAGE
# error	"You should define ARCH_IMPLEMENTS_FLUSH_DCACHE_PAGE for your platform"
#endif
#if ARCH_IMPLEMENTS_FLUSH_DCACHE_PAGE
extern void rq_flush_dcache_pages(struct request *rq);
#else
static inline void rq_flush_dcache_pages(struct request *rq)
{
}
#endif

L
Linus Torvalds 已提交
792 793 794
extern int blk_register_queue(struct gendisk *disk);
extern void blk_unregister_queue(struct gendisk *disk);
extern void generic_make_request(struct bio *bio);
795
extern void blk_rq_init(struct request_queue *q, struct request *rq);
L
Linus Torvalds 已提交
796
extern void blk_put_request(struct request *);
797 798
extern void __blk_put_request(struct request_queue *, struct request *);
extern struct request *blk_get_request(struct request_queue *, int, gfp_t);
799 800
extern struct request *blk_make_request(struct request_queue *, struct bio *,
					gfp_t);
J
Jens Axboe 已提交
801
extern void blk_rq_set_block_pc(struct request *);
802
extern void blk_requeue_request(struct request_queue *, struct request *);
803 804
extern void blk_add_request_payload(struct request *rq, struct page *page,
		unsigned int len);
805
extern int blk_rq_check_limits(struct request_queue *q, struct request *rq);
806
extern int blk_lld_busy(struct request_queue *q);
807 808 809 810 811
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);
812 813
extern int blk_insert_cloned_request(struct request_queue *q,
				     struct request *rq);
814
extern void blk_delay_queue(struct request_queue *, unsigned long);
815
extern void blk_recount_segments(struct request_queue *, struct bio *);
816
extern int scsi_verify_blk_ioctl(struct block_device *, unsigned int);
817 818
extern int scsi_cmd_blk_ioctl(struct block_device *, fmode_t,
			      unsigned int, void __user *);
819 820
extern int scsi_cmd_ioctl(struct request_queue *, struct gendisk *, fmode_t,
			  unsigned int, void __user *);
821 822
extern int sg_scsi_ioctl(struct request_queue *, struct gendisk *, fmode_t,
			 struct scsi_ioctl_command __user *);
823

824
extern void blk_queue_bio(struct request_queue *q, struct bio *bio);
C
Christoph Hellwig 已提交
825

826 827 828 829 830
/*
 * A queue has just exitted congestion.  Note this in the global counter of
 * congested queues, and wake up anyone who was waiting for requests to be
 * put back.
 */
831
static inline void blk_clear_queue_congested(struct request_queue *q, int sync)
832
{
833
	clear_bdi_congested(&q->backing_dev_info, sync);
834 835 836 837 838 839
}

/*
 * A queue has just entered congestion.  Flag that in the queue's VM-visible
 * state flags and increment the global gounter of congested queues.
 */
840
static inline void blk_set_queue_congested(struct request_queue *q, int sync)
841
{
842
	set_bdi_congested(&q->backing_dev_info, sync);
843 844
}

845 846
extern void blk_start_queue(struct request_queue *q);
extern void blk_stop_queue(struct request_queue *q);
L
Linus Torvalds 已提交
847
extern void blk_sync_queue(struct request_queue *q);
848
extern void __blk_stop_queue(struct request_queue *q);
849
extern void __blk_run_queue(struct request_queue *q);
850
extern void blk_run_queue(struct request_queue *);
851
extern void blk_run_queue_async(struct request_queue *q);
852
extern int blk_rq_map_user(struct request_queue *, struct request *,
853 854
			   struct rq_map_data *, void __user *, unsigned long,
			   gfp_t);
J
Jens Axboe 已提交
855
extern int blk_rq_unmap_user(struct bio *);
856 857
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 *,
858 859
			       struct rq_map_data *, const struct sg_iovec *,
			       int, unsigned int, gfp_t);
860
extern int blk_execute_rq(struct request_queue *, struct gendisk *,
861
			  struct request *, int);
862
extern void blk_execute_rq_nowait(struct request_queue *, struct gendisk *,
863
				  struct request *, int, rq_end_io_fn *);
864

865
static inline struct request_queue *bdev_get_queue(struct block_device *bdev)
L
Linus Torvalds 已提交
866
{
867
	return bdev->bd_disk->queue;	/* this is never NULL */
L
Linus Torvalds 已提交
868 869
}

870
/*
871 872 873 874 875 876
 * 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
877
 */
878 879
static inline sector_t blk_rq_pos(const struct request *rq)
{
880
	return rq->__sector;
881 882 883 884
}

static inline unsigned int blk_rq_bytes(const struct request *rq)
{
885
	return rq->__data_len;
886 887
}

888 889 890 891
static inline int blk_rq_cur_bytes(const struct request *rq)
{
	return rq->bio ? bio_cur_bytes(rq->bio) : 0;
}
892

893 894
extern unsigned int blk_rq_err_bytes(const struct request *rq);

895 896
static inline unsigned int blk_rq_sectors(const struct request *rq)
{
897
	return blk_rq_bytes(rq) >> 9;
898 899 900 901
}

static inline unsigned int blk_rq_cur_sectors(const struct request *rq)
{
902
	return blk_rq_cur_bytes(rq) >> 9;
903 904
}

905 906 907 908
static inline unsigned int blk_queue_get_max_sectors(struct request_queue *q,
						     unsigned int cmd_flags)
{
	if (unlikely(cmd_flags & REQ_DISCARD))
909
		return min(q->limits.max_discard_sectors, UINT_MAX >> 9);
910

911 912 913
	if (unlikely(cmd_flags & REQ_WRITE_SAME))
		return q->limits.max_write_same_sectors;

914 915 916
	return q->limits.max_sectors;
}

917 918 919 920 921 922 923 924
/*
 * 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)
925
		return q->limits.max_sectors;
926 927 928 929 930

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

931 932 933 934 935 936 937
static inline unsigned int blk_rq_get_max_sectors(struct request *rq)
{
	struct request_queue *q = rq->q;

	if (unlikely(rq->cmd_type == REQ_TYPE_BLOCK_PC))
		return q->limits.max_hw_sectors;

938 939 940 941 942
	if (!q->limits.chunk_sectors)
		return blk_queue_get_max_sectors(q, rq->cmd_flags);

	return min(blk_max_size_offset(q, blk_rq_pos(rq)),
			blk_queue_get_max_sectors(q, rq->cmd_flags));
943 944
}

945 946 947 948 949 950 951 952 953 954 955
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;
}

956 957 958 959 960 961 962
/*
 * 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
963
/*
964 965 966 967 968
 * Request completion related functions.
 *
 * blk_update_request() completes given number of bytes and updates
 * the request without completing it.
 *
969 970
 * blk_end_request() and friends.  __blk_end_request() must be called
 * with the request queue spinlock acquired.
L
Linus Torvalds 已提交
971 972
 *
 * Several drivers define their own end_request and call
973 974
 * blk_end_request() for parts of the original function.
 * This prevents code duplication in drivers.
L
Linus Torvalds 已提交
975
 */
976 977
extern bool blk_update_request(struct request *rq, int error,
			       unsigned int nr_bytes);
978
extern void blk_finish_request(struct request *rq, int error);
979 980 981 982
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);
983
extern bool blk_end_request_err(struct request *rq, int error);
984 985 986 987
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);
988
extern bool __blk_end_request_err(struct request *rq, int error);
989

990
extern void blk_complete_request(struct request *);
J
Jens Axboe 已提交
991 992
extern void __blk_complete_request(struct request *);
extern void blk_abort_request(struct request *);
993
extern void blk_unprep_request(struct request *);
994

L
Linus Torvalds 已提交
995 996 997
/*
 * Access functions for manipulating queue properties
 */
998
extern struct request_queue *blk_init_queue_node(request_fn_proc *rfn,
999
					spinlock_t *lock, int node_id);
1000
extern struct request_queue *blk_init_queue(request_fn_proc *, spinlock_t *);
1001 1002
extern struct request_queue *blk_init_allocated_queue(struct request_queue *,
						      request_fn_proc *, spinlock_t *);
1003 1004 1005
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);
1006
extern void blk_limits_max_hw_sectors(struct queue_limits *, unsigned int);
1007
extern void blk_queue_max_hw_sectors(struct request_queue *, unsigned int);
1008
extern void blk_queue_chunk_sectors(struct request_queue *, unsigned int);
1009
extern void blk_queue_max_segments(struct request_queue *, unsigned short);
1010
extern void blk_queue_max_segment_size(struct request_queue *, unsigned int);
1011 1012
extern void blk_queue_max_discard_sectors(struct request_queue *q,
		unsigned int max_discard_sectors);
1013 1014
extern void blk_queue_max_write_same_sectors(struct request_queue *q,
		unsigned int max_write_same_sectors);
1015
extern void blk_queue_logical_block_size(struct request_queue *, unsigned short);
1016
extern void blk_queue_physical_block_size(struct request_queue *, unsigned int);
1017 1018
extern void blk_queue_alignment_offset(struct request_queue *q,
				       unsigned int alignment);
1019
extern void blk_limits_io_min(struct queue_limits *limits, unsigned int min);
1020
extern void blk_queue_io_min(struct request_queue *q, unsigned int min);
1021
extern void blk_limits_io_opt(struct queue_limits *limits, unsigned int opt);
1022
extern void blk_queue_io_opt(struct request_queue *q, unsigned int opt);
1023
extern void blk_set_default_limits(struct queue_limits *lim);
1024
extern void blk_set_stacking_limits(struct queue_limits *lim);
1025 1026
extern int blk_stack_limits(struct queue_limits *t, struct queue_limits *b,
			    sector_t offset);
1027 1028
extern int bdev_stack_limits(struct queue_limits *t, struct block_device *bdev,
			    sector_t offset);
1029 1030
extern void disk_stack_limits(struct gendisk *disk, struct block_device *bdev,
			      sector_t offset);
1031
extern void blk_queue_stack_limits(struct request_queue *t, struct request_queue *b);
1032
extern void blk_queue_dma_pad(struct request_queue *, unsigned int);
1033
extern void blk_queue_update_dma_pad(struct request_queue *, unsigned int);
1034 1035 1036
extern int blk_queue_dma_drain(struct request_queue *q,
			       dma_drain_needed_fn *dma_drain_needed,
			       void *buf, unsigned int size);
1037
extern void blk_queue_lld_busy(struct request_queue *q, lld_busy_fn *fn);
1038 1039
extern void blk_queue_segment_boundary(struct request_queue *, unsigned long);
extern void blk_queue_prep_rq(struct request_queue *, prep_rq_fn *pfn);
1040
extern void blk_queue_unprep_rq(struct request_queue *, unprep_rq_fn *ufn);
1041 1042
extern void blk_queue_merge_bvec(struct request_queue *, merge_bvec_fn *);
extern void blk_queue_dma_alignment(struct request_queue *, int);
1043
extern void blk_queue_update_dma_alignment(struct request_queue *, int);
1044
extern void blk_queue_softirq_done(struct request_queue *, softirq_done_fn *);
J
Jens Axboe 已提交
1045 1046
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);
1047
extern void blk_queue_flush(struct request_queue *q, unsigned int flush);
1048
extern void blk_queue_flush_queueable(struct request_queue *q, bool queueable);
L
Linus Torvalds 已提交
1049 1050
extern struct backing_dev_info *blk_get_backing_dev_info(struct block_device *bdev);

1051
extern int blk_rq_map_sg(struct request_queue *, struct request *, struct scatterlist *);
A
Asias He 已提交
1052 1053
extern int blk_bio_map_sg(struct request_queue *q, struct bio *bio,
			  struct scatterlist *sglist);
L
Linus Torvalds 已提交
1054 1055 1056
extern void blk_dump_rq_flags(struct request *, char *);
extern long nr_blockdev_pages(void);

T
Tejun Heo 已提交
1057
bool __must_check blk_get_queue(struct request_queue *);
1058 1059 1060
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 *);
L
Linus Torvalds 已提交
1061

L
Lin Ming 已提交
1062 1063 1064
/*
 * block layer runtime pm functions
 */
1065
#ifdef CONFIG_PM
L
Lin Ming 已提交
1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082
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);
#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) {}
#endif

1083
/*
S
Suresh Jayaraman 已提交
1084 1085 1086 1087 1088 1089 1090 1091 1092 1093
 * 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.
1094
 */
1095
struct blk_plug {
S
Suresh Jayaraman 已提交
1096
	struct list_head list; /* requests */
1097
	struct list_head mq_list; /* blk-mq requests */
S
Suresh Jayaraman 已提交
1098
	struct list_head cb_list; /* md requires an unplug callback */
1099
};
1100 1101
#define BLK_MAX_REQUEST_COUNT 16

1102
struct blk_plug_cb;
1103
typedef void (*blk_plug_cb_fn)(struct blk_plug_cb *, bool);
1104 1105
struct blk_plug_cb {
	struct list_head list;
1106 1107
	blk_plug_cb_fn callback;
	void *data;
1108
};
1109 1110
extern struct blk_plug_cb *blk_check_plugged(blk_plug_cb_fn unplug,
					     void *data, int size);
1111 1112
extern void blk_start_plug(struct blk_plug *);
extern void blk_finish_plug(struct blk_plug *);
1113
extern void blk_flush_plug_list(struct blk_plug *, bool);
1114 1115 1116 1117 1118

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

1119 1120 1121 1122 1123 1124 1125 1126
	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;

1127
	if (plug)
1128
		blk_flush_plug_list(plug, true);
1129 1130 1131 1132 1133 1134
}

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

1135 1136 1137 1138
	return plug &&
		(!list_empty(&plug->list) ||
		 !list_empty(&plug->mq_list) ||
		 !list_empty(&plug->cb_list));
1139 1140
}

L
Linus Torvalds 已提交
1141 1142 1143
/*
 * tag stuff
 */
1144 1145 1146
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 *);
1147
extern int blk_queue_init_tags(struct request_queue *, int, struct blk_queue_tag *, int);
1148 1149 1150
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 *);
1151
extern struct blk_queue_tag *blk_init_tags(int, int);
1152
extern void blk_free_tags(struct blk_queue_tag *);
L
Linus Torvalds 已提交
1153

1154 1155 1156 1157 1158 1159 1160
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];
}
1161 1162 1163 1164

#define BLKDEV_DISCARD_SECURE  0x01    /* secure discard */

extern int blkdev_issue_flush(struct block_device *, gfp_t, sector_t *);
1165 1166
extern int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
		sector_t nr_sects, gfp_t gfp_mask, unsigned long flags);
1167 1168
extern int blkdev_issue_write_same(struct block_device *bdev, sector_t sector,
		sector_t nr_sects, gfp_t gfp_mask, struct page *page);
1169
extern int blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
1170
		sector_t nr_sects, gfp_t gfp_mask, bool discard);
1171 1172
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
1173
{
1174 1175 1176
	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
1177
}
1178
static inline int sb_issue_zeroout(struct super_block *sb, sector_t block,
1179
		sector_t nr_blocks, gfp_t gfp_mask)
1180 1181 1182 1183
{
	return blkdev_issue_zeroout(sb->s_bdev,
				    block << (sb->s_blocksize_bits - 9),
				    nr_blocks << (sb->s_blocksize_bits - 9),
1184
				    gfp_mask, true);
1185
}
L
Linus Torvalds 已提交
1186

1187
extern int blk_verify_command(unsigned char *cmd, fmode_t has_write_perm);
1188

1189 1190 1191 1192 1193 1194
enum blk_default_limits {
	BLK_MAX_SEGMENTS	= 128,
	BLK_SAFE_MAX_SECTORS	= 255,
	BLK_MAX_SEGMENT_SIZE	= 65536,
	BLK_SEG_BOUNDARY_MASK	= 0xFFFFFFFFUL,
};
1195

L
Linus Torvalds 已提交
1196 1197
#define blkdev_entry_to_request(entry) list_entry((entry), struct request, queuelist)

1198 1199
static inline unsigned long queue_bounce_pfn(struct request_queue *q)
{
1200
	return q->limits.bounce_pfn;
1201 1202 1203 1204
}

static inline unsigned long queue_segment_boundary(struct request_queue *q)
{
1205
	return q->limits.seg_boundary_mask;
1206 1207 1208 1209
}

static inline unsigned int queue_max_sectors(struct request_queue *q)
{
1210
	return q->limits.max_sectors;
1211 1212 1213 1214
}

static inline unsigned int queue_max_hw_sectors(struct request_queue *q)
{
1215
	return q->limits.max_hw_sectors;
1216 1217
}

1218
static inline unsigned short queue_max_segments(struct request_queue *q)
1219
{
1220
	return q->limits.max_segments;
1221 1222 1223 1224
}

static inline unsigned int queue_max_segment_size(struct request_queue *q)
{
1225
	return q->limits.max_segment_size;
1226 1227
}

1228
static inline unsigned short queue_logical_block_size(struct request_queue *q)
L
Linus Torvalds 已提交
1229 1230 1231
{
	int retval = 512;

1232 1233
	if (q && q->limits.logical_block_size)
		retval = q->limits.logical_block_size;
L
Linus Torvalds 已提交
1234 1235 1236 1237

	return retval;
}

1238
static inline unsigned short bdev_logical_block_size(struct block_device *bdev)
L
Linus Torvalds 已提交
1239
{
1240
	return queue_logical_block_size(bdev_get_queue(bdev));
L
Linus Torvalds 已提交
1241 1242
}

1243 1244 1245 1246 1247
static inline unsigned int queue_physical_block_size(struct request_queue *q)
{
	return q->limits.physical_block_size;
}

1248
static inline unsigned int bdev_physical_block_size(struct block_device *bdev)
M
Martin K. Petersen 已提交
1249 1250 1251 1252
{
	return queue_physical_block_size(bdev_get_queue(bdev));
}

1253 1254 1255 1256 1257
static inline unsigned int queue_io_min(struct request_queue *q)
{
	return q->limits.io_min;
}

M
Martin K. Petersen 已提交
1258 1259 1260 1261 1262
static inline int bdev_io_min(struct block_device *bdev)
{
	return queue_io_min(bdev_get_queue(bdev));
}

1263 1264 1265 1266 1267
static inline unsigned int queue_io_opt(struct request_queue *q)
{
	return q->limits.io_opt;
}

M
Martin K. Petersen 已提交
1268 1269 1270 1271 1272
static inline int bdev_io_opt(struct block_device *bdev)
{
	return queue_io_opt(bdev_get_queue(bdev));
}

1273 1274
static inline int queue_alignment_offset(struct request_queue *q)
{
M
Martin K. Petersen 已提交
1275
	if (q->limits.misaligned)
1276 1277
		return -1;

M
Martin K. Petersen 已提交
1278
	return q->limits.alignment_offset;
1279 1280
}

1281
static inline int queue_limit_alignment_offset(struct queue_limits *lim, sector_t sector)
1282 1283
{
	unsigned int granularity = max(lim->physical_block_size, lim->io_min);
1284
	unsigned int alignment = sector_div(sector, granularity >> 9) << 9;
1285

1286
	return (granularity + lim->alignment_offset - alignment) % granularity;
1287 1288
}

M
Martin K. Petersen 已提交
1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301
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;
}

1302 1303 1304 1305 1306 1307 1308 1309
static inline int queue_discard_alignment(struct request_queue *q)
{
	if (q->limits.discard_misaligned)
		return -1;

	return q->limits.discard_alignment;
}

1310
static inline int queue_limit_discard_alignment(struct queue_limits *lim, sector_t sector)
1311
{
1312
	unsigned int alignment, granularity, offset;
1313

1314 1315 1316
	if (!lim->max_discard_sectors)
		return 0;

1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330
	/* 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;
1331 1332
}

1333 1334 1335 1336 1337 1338 1339 1340 1341 1342
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;
}

1343 1344
static inline unsigned int queue_discard_zeroes_data(struct request_queue *q)
{
1345
	if (q->limits.max_discard_sectors && q->limits.discard_zeroes_data == 1)
1346 1347 1348 1349 1350 1351 1352 1353 1354 1355
		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));
}

1356 1357 1358 1359 1360 1361 1362 1363 1364 1365
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;
}

1366
static inline int queue_dma_alignment(struct request_queue *q)
L
Linus Torvalds 已提交
1367
{
1368
	return q ? q->dma_alignment : 511;
L
Linus Torvalds 已提交
1369 1370
}

1371
static inline int blk_rq_aligned(struct request_queue *q, unsigned long addr,
1372 1373 1374
				 unsigned int len)
{
	unsigned int alignment = queue_dma_alignment(q) | q->dma_pad_mask;
1375
	return !(addr & alignment) && !(len & alignment);
1376 1377
}

L
Linus Torvalds 已提交
1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388
/* 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;
}

1389
static inline unsigned int block_size(struct block_device *bdev)
L
Linus Torvalds 已提交
1390 1391 1392 1393
{
	return bdev->bd_block_size;
}

1394 1395 1396 1397 1398
static inline bool queue_flush_queueable(struct request_queue *q)
{
	return !q->flush_not_queueable;
}

L
Linus Torvalds 已提交
1399 1400 1401 1402 1403 1404 1405 1406 1407 1408
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)
{
	page_cache_release(p.v);
}

struct work_struct;
1409 1410
int kblockd_schedule_work(struct work_struct *work);
int kblockd_schedule_delayed_work(struct delayed_work *dwork, unsigned long delay);
1411
int kblockd_schedule_delayed_work_on(int cpu, struct delayed_work *dwork, unsigned long delay);
L
Linus Torvalds 已提交
1412

1413
#ifdef CONFIG_BLK_CGROUP
1414 1415 1416 1417 1418
/*
 * 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().
 */
1419 1420
static inline void set_start_time_ns(struct request *req)
{
1421
	preempt_disable();
1422
	req->start_time_ns = sched_clock();
1423
	preempt_enable();
1424 1425 1426 1427
}

static inline void set_io_start_time_ns(struct request *req)
{
1428
	preempt_disable();
1429
	req->io_start_time_ns = sched_clock();
1430
	preempt_enable();
1431
}
1432 1433 1434 1435 1436 1437 1438 1439 1440 1441

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;
}
1442 1443 1444
#else
static inline void set_start_time_ns(struct request *req) {}
static inline void set_io_start_time_ns(struct request *req) {}
1445 1446 1447 1448 1449 1450 1451 1452
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;
}
1453 1454
#endif

L
Linus Torvalds 已提交
1455 1456 1457 1458 1459
#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) "-*")

1460 1461
#if defined(CONFIG_BLK_DEV_INTEGRITY)

1462 1463 1464
enum blk_integrity_flags {
	BLK_INTEGRITY_VERIFY		= 1 << 0,
	BLK_INTEGRITY_GENERATE		= 1 << 1,
1465
	BLK_INTEGRITY_DEVICE_CAPABLE	= 1 << 2,
1466
	BLK_INTEGRITY_IP_CHECKSUM	= 1 << 3,
1467
};
1468

1469
struct blk_integrity_iter {
1470 1471
	void			*prot_buf;
	void			*data_buf;
1472
	sector_t		seed;
1473
	unsigned int		data_size;
1474
	unsigned short		interval;
1475 1476 1477
	const char		*disk_name;
};

1478
typedef int (integrity_processing_fn) (struct blk_integrity_iter *);
1479 1480

struct blk_integrity {
1481 1482
	integrity_processing_fn	*generate_fn;
	integrity_processing_fn	*verify_fn;
1483 1484 1485

	unsigned short		flags;
	unsigned short		tuple_size;
1486
	unsigned short		interval;
1487 1488 1489 1490 1491 1492 1493
	unsigned short		tag_size;

	const char		*name;

	struct kobject		kobj;
};

1494
extern bool blk_integrity_is_initialized(struct gendisk *);
1495 1496
extern int blk_integrity_register(struct gendisk *, struct blk_integrity *);
extern void blk_integrity_unregister(struct gendisk *);
1497
extern int blk_integrity_compare(struct gendisk *, struct gendisk *);
1498 1499 1500
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 *);
1501 1502 1503 1504
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 *);
1505

1506 1507 1508 1509 1510 1511
static inline
struct blk_integrity *bdev_get_integrity(struct block_device *bdev)
{
	return bdev->bd_disk->integrity;
}

1512 1513 1514 1515 1516
static inline struct blk_integrity *blk_get_integrity(struct gendisk *disk)
{
	return disk->integrity;
}

1517
static inline bool blk_integrity_rq(struct request *rq)
1518
{
1519
	return rq->cmd_flags & REQ_INTEGRITY;
1520 1521
}

1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533
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;
}

1534 1535
#else /* CONFIG_BLK_DEV_INTEGRITY */

1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557
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)
{
1558
	return NULL;
1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583
}
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;
}
static inline int blk_integrity_register(struct gendisk *d,
					 struct blk_integrity *b)
{
	return 0;
}
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;
}
1584 1585 1586
static inline bool blk_integrity_merge_rq(struct request_queue *rq,
					  struct request *r1,
					  struct request *r2)
1587
{
1588
	return true;
1589
}
1590 1591 1592
static inline bool blk_integrity_merge_bio(struct request_queue *rq,
					   struct request *r,
					   struct bio *b)
1593
{
1594
	return true;
1595 1596 1597 1598 1599
}
static inline bool blk_integrity_is_initialized(struct gendisk *g)
{
	return 0;
}
1600 1601 1602

#endif /* CONFIG_BLK_DEV_INTEGRITY */

1603
struct block_device_operations {
A
Al Viro 已提交
1604
	int (*open) (struct block_device *, fmode_t);
1605
	void (*release) (struct gendisk *, fmode_t);
1606
	int (*rw_page)(struct block_device *, sector_t, struct page *, int rw);
A
Al Viro 已提交
1607 1608
	int (*ioctl) (struct block_device *, fmode_t, unsigned, unsigned long);
	int (*compat_ioctl) (struct block_device *, fmode_t, unsigned, unsigned long);
1609 1610
	long (*direct_access)(struct block_device *, sector_t,
					void **, unsigned long *pfn, long size);
1611 1612 1613
	unsigned int (*check_events) (struct gendisk *disk,
				      unsigned int clearing);
	/* ->media_changed() is DEPRECATED, use ->check_events() instead */
1614
	int (*media_changed) (struct gendisk *);
1615
	void (*unlock_native_capacity) (struct gendisk *);
1616 1617
	int (*revalidate_disk) (struct gendisk *);
	int (*getgeo)(struct block_device *, struct hd_geometry *);
1618 1619
	/* this callback is with swap_lock and sometimes page table lock held */
	void (*swap_slot_free_notify) (struct block_device *, unsigned long);
1620 1621 1622
	struct module *owner;
};

1623 1624
extern int __blkdev_driver_ioctl(struct block_device *, fmode_t, unsigned int,
				 unsigned long);
1625 1626 1627
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 *);
1628 1629
extern long bdev_direct_access(struct block_device *, sector_t, void **addr,
						unsigned long *pfn, long size);
1630
#else /* CONFIG_BLOCK */
1631 1632 1633

struct block_device;

1634 1635 1636 1637 1638 1639 1640 1641 1642 1643
/*
 * stubs for when the block layer is configured out
 */
#define buffer_heads_over_limit 0

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

1644 1645 1646 1647
struct blk_plug {
};

static inline void blk_start_plug(struct blk_plug *plug)
1648 1649 1650
{
}

1651
static inline void blk_finish_plug(struct blk_plug *plug)
1652 1653 1654
{
}

1655
static inline void blk_flush_plug(struct task_struct *task)
1656 1657 1658
{
}

1659 1660 1661 1662 1663
static inline void blk_schedule_flush_plug(struct task_struct *task)
{
}


1664 1665 1666 1667 1668
static inline bool blk_needs_flush_plug(struct task_struct *tsk)
{
	return false;
}

1669 1670 1671 1672 1673 1674
static inline int blkdev_issue_flush(struct block_device *bdev, gfp_t gfp_mask,
				     sector_t *error_sector)
{
	return 0;
}

1675 1676
#endif /* CONFIG_BLOCK */

L
Linus Torvalds 已提交
1677
#endif