blkdev.h 45.7 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>
L
Linus Torvalds 已提交
24 25 26

#include <asm/scatterlist.h>

27
struct module;
28 29
struct scsi_ioctl_command;

L
Linus Torvalds 已提交
30 31 32
struct request_queue;
struct elevator_queue;
struct request_pm_state;
33
struct blk_trace;
34 35
struct request;
struct sg_io_hdr;
M
Mike Christie 已提交
36
struct bsg_job;
T
Tejun Heo 已提交
37
struct blkcg_gq;
L
Linus Torvalds 已提交
38 39 40 41

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

T
Tejun Heo 已提交
42 43 44 45 46 47
/*
 * 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
48
struct request;
49
typedef void (rq_end_io_fn)(struct request *, int);
L
Linus Torvalds 已提交
50

51 52 53
#define BLK_RL_SYNCFULL		(1U << 0)
#define BLK_RL_ASYNCFULL	(1U << 1)

L
Linus Torvalds 已提交
54
struct request_list {
55
	struct request_queue	*q;	/* the queue this rl belongs to */
56 57 58
#ifdef CONFIG_BLK_CGROUP
	struct blkcg_gq		*blkg;	/* blkg this request pool belongs to */
#endif
59 60 61 62
	/*
	 * count[], starved[], and wait[] are indexed by
	 * BLK_RW_SYNC/BLK_RW_ASYNC
	 */
63 64 65 66
	int			count[2];
	int			starved[2];
	mempool_t		*rq_pool;
	wait_queue_head_t	wait[2];
67
	unsigned int		flags;
L
Linus Torvalds 已提交
68 69
};

70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86
/*
 * 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,
87
	REQ_TYPE_ATA_PC,
88 89
};

L
Linus Torvalds 已提交
90 91 92
#define BLK_MAX_CDB	16

/*
J
Jens Axboe 已提交
93
 * try to put the fields that are referenced together in the same cacheline.
94
 * if you modify this structure, be sure to check block/blk-core.c:blk_rq_init()
J
Jens Axboe 已提交
95
 * as well!
L
Linus Torvalds 已提交
96 97
 */
struct request {
98 99 100 101 102 103 104 105
	union {
		struct list_head queuelist;
		struct llist_node ll_list;
	};
	union {
		struct call_single_data csd;
		struct work_struct mq_flush_data;
	};
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
	struct hlist_node hash;	/* merge hash */
124 125 126
	/*
	 * The rb_node is only used inside the io scheduler, requests
	 * are pruned when moved to the dispatch queue. So let the
127
	 * completion_data share space with the rb_node.
128 129 130
	 */
	union {
		struct rb_node rb_node;	/* sort/lookup */
131
		void *completion_data;
132
	};
133

134
	/*
135
	 * Three pointers are available for the IO schedulers, if they need
136 137
	 * more they have to dynamically allocate it.  Flush requests are
	 * never put on the IO scheduler. So let the flush fields share
138
	 * space with the elevator data.
139
	 */
140
	union {
141 142 143 144 145
		struct {
			struct io_cq		*icq;
			void			*priv[2];
		} elv;

146 147 148
		struct {
			unsigned int		seq;
			struct list_head	list;
149
			rq_end_io_fn		*saved_end_io;
150 151
		} flush;
	};
152

153
	struct gendisk *rq_disk;
154
	struct hd_struct *part;
L
Linus Torvalds 已提交
155
	unsigned long start_time;
156
#ifdef CONFIG_BLK_CGROUP
157
	struct request_list *rl;		/* rl this rq is alloced from */
158 159 160
	unsigned long long start_time_ns;
	unsigned long long io_start_time_ns;    /* when passed to hardware */
#endif
L
Linus Torvalds 已提交
161 162 163 164
	/* Number of scatter-gather DMA addr+len pairs after
	 * physical address coalescing is performed.
	 */
	unsigned short nr_phys_segments;
165 166 167
#if defined(CONFIG_BLK_DEV_INTEGRITY)
	unsigned short nr_integrity_segments;
#endif
L
Linus Torvalds 已提交
168

169 170
	unsigned short ioprio;

T
Tejun Heo 已提交
171 172
	void *special;		/* opaque pointer available for LLD use */
	char *buffer;		/* kaddr of the current segment if available */
L
Linus Torvalds 已提交
173

174 175 176
	int tag;
	int errors;

L
Linus Torvalds 已提交
177 178 179
	/*
	 * when request is used as a packet command carrier
	 */
180 181
	unsigned char __cmd[BLK_MAX_CDB];
	unsigned char *cmd;
182
	unsigned short cmd_len;
L
Linus Torvalds 已提交
183

184
	unsigned int extra_len;	/* length of alignment and padding */
L
Linus Torvalds 已提交
185
	unsigned int sense_len;
T
Tejun Heo 已提交
186
	unsigned int resid_len;	/* residual count */
L
Linus Torvalds 已提交
187 188
	void *sense;

J
Jens Axboe 已提交
189 190
	unsigned long deadline;
	struct list_head timeout_list;
L
Linus Torvalds 已提交
191
	unsigned int timeout;
192
	int retries;
L
Linus Torvalds 已提交
193 194

	/*
195
	 * completion callback.
L
Linus Torvalds 已提交
196 197 198
	 */
	rq_end_io_fn *end_io;
	void *end_io_data;
199 200 201

	/* for bidi */
	struct request *next_rq;
L
Linus Torvalds 已提交
202 203
};

204 205 206 207 208
static inline unsigned short req_get_ioprio(struct request *req)
{
	return req->ioprio;
}

L
Linus Torvalds 已提交
209
/*
210
 * State information carried for REQ_TYPE_PM_SUSPEND and REQ_TYPE_PM_RESUME
L
Linus Torvalds 已提交
211 212 213 214 215 216 217 218 219 220 221 222 223
 * 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>

224 225
struct blk_queue_ctx;

226
typedef void (request_fn_proc) (struct request_queue *q);
227
typedef void (make_request_fn) (struct request_queue *q, struct bio *bio);
228
typedef int (prep_rq_fn) (struct request_queue *, struct request *);
229
typedef void (unprep_rq_fn) (struct request_queue *, struct request *);
L
Linus Torvalds 已提交
230 231

struct bio_vec;
232 233 234 235 236 237 238 239
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 *);
240
typedef void (softirq_done_fn)(struct request *);
241
typedef int (dma_drain_needed_fn)(struct request *);
242
typedef int (lld_busy_fn) (struct request_queue *q);
M
Mike Christie 已提交
243
typedef int (bsg_job_fn) (struct bsg_job *);
L
Linus Torvalds 已提交
244

J
Jens Axboe 已提交
245 246 247 248 249 250 251 252
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
253 254 255 256 257 258 259 260 261 262
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 */
263
	int real_max_depth;		/* what the array can hold */
L
Linus Torvalds 已提交
264 265 266
	atomic_t refcnt;		/* map can be shared */
};

267 268 269
#define BLK_SCSI_MAX_CMDS	(256)
#define BLK_SCSI_CMD_PER_LONG	(BLK_SCSI_MAX_CMDS / (sizeof(long) * 8))

270 271 272 273 274 275 276
struct queue_limits {
	unsigned long		bounce_pfn;
	unsigned long		seg_boundary_mask;

	unsigned int		max_hw_sectors;
	unsigned int		max_sectors;
	unsigned int		max_segment_size;
277 278 279 280
	unsigned int		physical_block_size;
	unsigned int		alignment_offset;
	unsigned int		io_min;
	unsigned int		io_opt;
281
	unsigned int		max_discard_sectors;
282
	unsigned int		max_write_same_sectors;
283 284
	unsigned int		discard_granularity;
	unsigned int		discard_alignment;
285 286

	unsigned short		logical_block_size;
287
	unsigned short		max_segments;
288
	unsigned short		max_integrity_segments;
289

290
	unsigned char		misaligned;
291
	unsigned char		discard_misaligned;
292
	unsigned char		cluster;
293
	unsigned char		discard_zeroes_data;
294 295
};

296
struct request_queue {
L
Linus Torvalds 已提交
297 298 299 300 301
	/*
	 * Together with queue_head for cacheline sharing
	 */
	struct list_head	queue_head;
	struct request		*last_merge;
J
Jens Axboe 已提交
302
	struct elevator_queue	*elevator;
303 304
	int			nr_rqs[2];	/* # allocated [a]sync rqs */
	int			nr_rqs_elvpriv;	/* # allocated rqs w/ elvpriv */
L
Linus Torvalds 已提交
305 306

	/*
307 308 309 310
	 * 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
311
	 */
312
	struct request_list	root_rl;
L
Linus Torvalds 已提交
313 314 315 316

	request_fn_proc		*request_fn;
	make_request_fn		*make_request_fn;
	prep_rq_fn		*prep_rq_fn;
317
	unprep_rq_fn		*unprep_rq_fn;
L
Linus Torvalds 已提交
318
	merge_bvec_fn		*merge_bvec_fn;
319
	softirq_done_fn		*softirq_done_fn;
J
Jens Axboe 已提交
320
	rq_timed_out_fn		*rq_timed_out_fn;
321
	dma_drain_needed_fn	*dma_drain_needed;
322
	lld_busy_fn		*lld_busy_fn;
L
Linus Torvalds 已提交
323

324 325 326 327 328 329 330 331 332 333 334 335
	struct blk_mq_ops	*mq_ops;

	unsigned int		*mq_map;

	/* sw queues */
	struct blk_mq_ctx	*queue_ctx;
	unsigned int		nr_queues;

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

336 337 338
	/*
	 * Dispatch queue sorting
	 */
J
Jens Axboe 已提交
339
	sector_t		end_sector;
340 341
	struct request		*boundary_rq;

L
Linus Torvalds 已提交
342
	/*
343
	 * Delayed queue handling
L
Linus Torvalds 已提交
344
	 */
345
	struct delayed_work	delay_work;
L
Linus Torvalds 已提交
346 347 348 349 350 351 352 353 354 355

	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;

	/*
356
	 * various queue flags, see QUEUE_* below
L
Linus Torvalds 已提交
357
	 */
358
	unsigned long		queue_flags;
L
Linus Torvalds 已提交
359

360 361 362 363 364 365
	/*
	 * ida allocated id for this queue.  Used to index queues from
	 * ioctx.
	 */
	int			id;

L
Linus Torvalds 已提交
366
	/*
367
	 * queue needs bounce pages for pages above this limit
L
Linus Torvalds 已提交
368
	 */
369
	gfp_t			bounce_gfp;
L
Linus Torvalds 已提交
370 371

	/*
已提交
372 373 374
	 * 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
375
	 */
已提交
376
	spinlock_t		__queue_lock;
L
Linus Torvalds 已提交
377 378 379 380 381 382 383
	spinlock_t		*queue_lock;

	/*
	 * queue kobject
	 */
	struct kobject kobj;

384 385 386 387 388
	/*
	 * mq queue kobject
	 */
	struct kobject mq_kobj;

L
Lin Ming 已提交
389 390 391 392 393 394
#ifdef CONFIG_PM_RUNTIME
	struct device		*dev;
	int			rpm_status;
	unsigned int		nr_pending;
#endif

L
Linus Torvalds 已提交
395 396 397 398 399 400 401 402
	/*
	 * 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
403
	unsigned int		dma_drain_size;
404
	void			*dma_drain_buffer;
405
	unsigned int		dma_pad_mask;
L
Linus Torvalds 已提交
406 407 408
	unsigned int		dma_alignment;

	struct blk_queue_tag	*queue_tags;
409
	struct list_head	tag_busy_list;
L
Linus Torvalds 已提交
410

411
	unsigned int		nr_sorted;
412
	unsigned int		in_flight[2];
413 414 415 416 417 418
	/*
	 * 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
419

J
Jens Axboe 已提交
420 421 422 423
	unsigned int		rq_timeout;
	struct timer_list	timeout;
	struct list_head	timeout_list;

424
	struct list_head	icq_list;
425
#ifdef CONFIG_BLK_CGROUP
426
	DECLARE_BITMAP		(blkcg_pols, BLKCG_MAX_POLS);
T
Tejun Heo 已提交
427
	struct blkcg_gq		*root_blkg;
428
	struct list_head	blkg_list;
429
#endif
430

431 432
	struct queue_limits	limits;

L
Linus Torvalds 已提交
433 434 435 436 437
	/*
	 * sg stuff
	 */
	unsigned int		sg_timeout;
	unsigned int		sg_reserved_size;
438
	int			node;
439
#ifdef CONFIG_BLK_DEV_IO_TRACE
440
	struct blk_trace	*blk_trace;
441
#endif
L
Linus Torvalds 已提交
442
	/*
443
	 * for flush operations
L
Linus Torvalds 已提交
444
	 */
445
	unsigned int		flush_flags;
446
	unsigned int		flush_not_queueable:1;
447
	unsigned int		flush_queue_delayed:1;
448 449 450 451 452
	unsigned int		flush_pending_idx:1;
	unsigned int		flush_running_idx:1;
	unsigned long		flush_pending_since;
	struct list_head	flush_queue[2];
	struct list_head	flush_data_in_flight;
453 454 455 456 457 458 459
	union {
		struct request	flush_rq;
		struct {
			spinlock_t mq_flush_lock;
			struct work_struct mq_flush_work;
		};
	};
460 461

	struct mutex		sysfs_lock;
462

463 464
	int			bypass_depth;

465
#if defined(CONFIG_BLK_DEV_BSG)
M
Mike Christie 已提交
466 467
	bsg_job_fn		*bsg_job_fn;
	int			bsg_job_size;
468 469
	struct bsg_class_device bsg_dev;
#endif
470 471 472 473 474

#ifdef CONFIG_BLK_DEV_THROTTLING
	/* Throttle data */
	struct throtl_data *td;
#endif
T
Tejun Heo 已提交
475
	struct rcu_head		rcu_head;
476 477 478
	wait_queue_head_t	mq_freeze_wq;
	struct percpu_counter	mq_usage_counter;
	struct list_head	all_q_node;
L
Linus Torvalds 已提交
479 480 481 482
};

#define QUEUE_FLAG_QUEUED	1	/* uses generic tag queueing */
#define QUEUE_FLAG_STOPPED	2	/* queue is stopped */
483 484
#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
485
#define QUEUE_FLAG_DYING	5	/* queue being torn down */
486
#define QUEUE_FLAG_BYPASS	6	/* act as dumb FIFO queue */
487 488
#define QUEUE_FLAG_BIDI		7	/* queue supports bidi requests */
#define QUEUE_FLAG_NOMERGES     8	/* disable merge attempts */
D
Dan Williams 已提交
489
#define QUEUE_FLAG_SAME_COMP	9	/* complete on same CPU-group */
490 491 492
#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) */
493
#define QUEUE_FLAG_VIRT        QUEUE_FLAG_NONROT /* paravirt device */
494 495 496 497 498
#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
499
#define QUEUE_FLAG_SAME_FORCE  18	/* force complete on same CPU */
500
#define QUEUE_FLAG_DEAD        19	/* queue tear-down finished */
501
#define QUEUE_FLAG_INIT_DONE   20	/* queue is initialized */
502 503

#define QUEUE_FLAG_DEFAULT	((1 << QUEUE_FLAG_IO_STAT) |		\
504
				 (1 << QUEUE_FLAG_STACKABLE)	|	\
505 506
				 (1 << QUEUE_FLAG_SAME_COMP)	|	\
				 (1 << QUEUE_FLAG_ADD_RANDOM))
507

508 509 510
#define QUEUE_FLAG_MQ_DEFAULT	((1 << QUEUE_FLAG_IO_STAT) |		\
				 (1 << QUEUE_FLAG_SAME_COMP))

511
static inline void queue_lockdep_assert_held(struct request_queue *q)
512
{
513 514
	if (q->queue_lock)
		lockdep_assert_held(q->queue_lock);
515 516
}

N
Nick Piggin 已提交
517 518 519 520 521 522
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
523 524 525
static inline int queue_flag_test_and_clear(unsigned int flag,
					    struct request_queue *q)
{
526
	queue_lockdep_assert_held(q);
J
Jens Axboe 已提交
527 528 529 530 531 532 533 534 535 536 537 538

	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)
{
539
	queue_lockdep_assert_held(q);
J
Jens Axboe 已提交
540 541 542 543 544 545 546 547 548

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

	return 1;
}

N
Nick Piggin 已提交
549 550
static inline void queue_flag_set(unsigned int flag, struct request_queue *q)
{
551
	queue_lockdep_assert_held(q);
N
Nick Piggin 已提交
552 553 554 555 556 557 558 559 560
	__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);
}

561 562 563 564 565
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
566 567
static inline void queue_flag_clear(unsigned int flag, struct request_queue *q)
{
568
	queue_lockdep_assert_held(q);
N
Nick Piggin 已提交
569 570 571
	__clear_bit(flag, &q->queue_flags);
}

L
Linus Torvalds 已提交
572 573
#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
574
#define blk_queue_dying(q)	test_bit(QUEUE_FLAG_DYING, &(q)->queue_flags)
575
#define blk_queue_dead(q)	test_bit(QUEUE_FLAG_DEAD, &(q)->queue_flags)
576
#define blk_queue_bypass(q)	test_bit(QUEUE_FLAG_BYPASS, &(q)->queue_flags)
577
#define blk_queue_init_done(q)	test_bit(QUEUE_FLAG_INIT_DONE, &(q)->queue_flags)
578
#define blk_queue_nomerges(q)	test_bit(QUEUE_FLAG_NOMERGES, &(q)->queue_flags)
579 580
#define blk_queue_noxmerges(q)	\
	test_bit(QUEUE_FLAG_NOXMERGES, &(q)->queue_flags)
581
#define blk_queue_nonrot(q)	test_bit(QUEUE_FLAG_NONROT, &(q)->queue_flags)
582
#define blk_queue_io_stat(q)	test_bit(QUEUE_FLAG_IO_STAT, &(q)->queue_flags)
583
#define blk_queue_add_random(q)	test_bit(QUEUE_FLAG_ADD_RANDOM, &(q)->queue_flags)
584 585
#define blk_queue_stackable(q)	\
	test_bit(QUEUE_FLAG_STACKABLE, &(q)->queue_flags)
586
#define blk_queue_discard(q)	test_bit(QUEUE_FLAG_DISCARD, &(q)->queue_flags)
A
Adrian Hunter 已提交
587 588
#define blk_queue_secdiscard(q)	(blk_queue_discard(q) && \
	test_bit(QUEUE_FLAG_SECDISCARD, &(q)->queue_flags))
L
Linus Torvalds 已提交
589

590 591 592 593 594 595
#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) && \
596
	 ((rq)->cmd_type == REQ_TYPE_FS))
597

L
Linus Torvalds 已提交
598
#define blk_pm_request(rq)	\
599 600
	((rq)->cmd_type == REQ_TYPE_PM_SUSPEND || \
	 (rq)->cmd_type == REQ_TYPE_PM_RESUME)
L
Linus Torvalds 已提交
601

602
#define blk_rq_cpu_valid(rq)	((rq)->cpu != -1)
603
#define blk_bidi_rq(rq)		((rq)->next_rq != NULL)
604 605
/* rq->queuelist of dequeued request must be list_empty() */
#define blk_queued_rq(rq)	(!list_empty(&(rq)->queuelist))
L
Linus Torvalds 已提交
606 607 608

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

J
Jens Axboe 已提交
609
#define rq_data_dir(rq)		(((rq)->cmd_flags & 1) != 0)
L
Linus Torvalds 已提交
610

611 612 613 614 615
static inline unsigned int blk_queue_cluster(struct request_queue *q)
{
	return q->limits.cluster;
}

616
/*
617
 * We regard a request as sync, if either a read or a sync write
618
 */
619 620
static inline bool rw_is_sync(unsigned int rw_flags)
{
621
	return !(rw_flags & REQ_WRITE) || (rw_flags & REQ_SYNC);
622 623 624 625 626 627 628
}

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

629
static inline bool blk_rl_full(struct request_list *rl, bool sync)
L
Linus Torvalds 已提交
630
{
631 632 633
	unsigned int flag = sync ? BLK_RL_SYNCFULL : BLK_RL_ASYNCFULL;

	return rl->flags & flag;
L
Linus Torvalds 已提交
634 635
}

636
static inline void blk_set_rl_full(struct request_list *rl, bool sync)
L
Linus Torvalds 已提交
637
{
638 639 640
	unsigned int flag = sync ? BLK_RL_SYNCFULL : BLK_RL_ASYNCFULL;

	rl->flags |= flag;
L
Linus Torvalds 已提交
641 642
}

643
static inline void blk_clear_rl_full(struct request_list *rl, bool sync)
L
Linus Torvalds 已提交
644
{
645 646 647
	unsigned int flag = sync ? BLK_RL_SYNCFULL : BLK_RL_ASYNCFULL;

	rl->flags &= ~flag;
L
Linus Torvalds 已提交
648 649
}

650 651 652 653
static inline bool rq_mergeable(struct request *rq)
{
	if (rq->cmd_type != REQ_TYPE_FS)
		return false;
L
Linus Torvalds 已提交
654

655 656 657 658 659
	if (rq->cmd_flags & REQ_NOMERGE_FLAGS)
		return false;

	return true;
}
L
Linus Torvalds 已提交
660

661 662 663 664 665 666 667 668 669
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;

670 671 672
	if ((flags1 & REQ_WRITE_SAME) != (flags2 & REQ_WRITE_SAME))
		return false;

673 674 675
	return true;
}

676 677 678 679 680 681 682 683
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
684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699
/*
 * 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
 */
700 701

#if BITS_PER_LONG == 32
L
Linus Torvalds 已提交
702
#define BLK_BOUNCE_HIGH		((u64)blk_max_low_pfn << PAGE_SHIFT)
703 704 705 706
#else
#define BLK_BOUNCE_HIGH		-1ULL
#endif
#define BLK_BOUNCE_ANY		(-1ULL)
707
#define BLK_BOUNCE_ISA		(DMA_BIT_MASK(24))
L
Linus Torvalds 已提交
708

709 710 711 712
/*
 * default timeout for SG_IO if none specified
 */
#define BLK_DEFAULT_SG_TIMEOUT	(60 * HZ)
L
Linus Torvalds 已提交
713
#define BLK_MIN_SG_TIMEOUT	(7 * HZ)
714

715
#ifdef CONFIG_BOUNCE
L
Linus Torvalds 已提交
716
extern int init_emergency_isa_pool(void);
717
extern void blk_queue_bounce(struct request_queue *q, struct bio **bio);
L
Linus Torvalds 已提交
718 719 720 721 722
#else
static inline int init_emergency_isa_pool(void)
{
	return 0;
}
723
static inline void blk_queue_bounce(struct request_queue *q, struct bio **bio)
L
Linus Torvalds 已提交
724 725 726 727
{
}
#endif /* CONFIG_MMU */

728 729 730 731
struct rq_map_data {
	struct page **pages;
	int page_order;
	int nr_entries;
732
	unsigned long offset;
733
	int null_mapped;
734
	int from_user;
735 736
};

737
struct req_iterator {
738
	struct bvec_iter iter;
739 740 741 742
	struct bio *bio;
};

/* This should not be used directly - use rq_for_each_segment */
743 744
#define for_each_bio(_bio)		\
	for (; _bio; _bio = _bio->bi_next)
745
#define __rq_for_each_bio(_bio, rq)	\
L
Linus Torvalds 已提交
746 747 748
	if ((rq->bio))			\
		for (_bio = (rq)->bio; _bio; _bio = _bio->bi_next)

749 750
#define rq_for_each_segment(bvl, _rq, _iter)			\
	__rq_for_each_bio(_iter.bio, _rq)			\
751
		bio_for_each_segment(bvl, _iter.bio, _iter.iter)
752

K
Kent Overstreet 已提交
753
#define rq_iter_last(bvec, _iter)				\
754
		(_iter.bio->bi_next == NULL &&			\
K
Kent Overstreet 已提交
755
		 bio_iter_last(bvec, _iter.iter))
756

757 758 759 760 761 762 763 764 765 766 767
#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
768 769 770
extern int blk_register_queue(struct gendisk *disk);
extern void blk_unregister_queue(struct gendisk *disk);
extern void generic_make_request(struct bio *bio);
771
extern void blk_rq_init(struct request_queue *q, struct request *rq);
L
Linus Torvalds 已提交
772
extern void blk_put_request(struct request *);
773 774
extern void __blk_put_request(struct request_queue *, struct request *);
extern struct request *blk_get_request(struct request_queue *, int, gfp_t);
775 776
extern struct request *blk_make_request(struct request_queue *, struct bio *,
					gfp_t);
777
extern void blk_requeue_request(struct request_queue *, struct request *);
778 779
extern void blk_add_request_payload(struct request *rq, struct page *page,
		unsigned int len);
780
extern int blk_rq_check_limits(struct request_queue *q, struct request *rq);
781
extern int blk_lld_busy(struct request_queue *q);
782 783 784 785 786
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);
787 788
extern int blk_insert_cloned_request(struct request_queue *q,
				     struct request *rq);
789
extern void blk_delay_queue(struct request_queue *, unsigned long);
790
extern void blk_recount_segments(struct request_queue *, struct bio *);
791
extern int scsi_verify_blk_ioctl(struct block_device *, unsigned int);
792 793
extern int scsi_cmd_blk_ioctl(struct block_device *, fmode_t,
			      unsigned int, void __user *);
794 795
extern int scsi_cmd_ioctl(struct request_queue *, struct gendisk *, fmode_t,
			  unsigned int, void __user *);
796 797
extern int sg_scsi_ioctl(struct request_queue *, struct gendisk *, fmode_t,
			 struct scsi_ioctl_command __user *);
798

799
extern void blk_queue_bio(struct request_queue *q, struct bio *bio);
C
Christoph Hellwig 已提交
800

801 802 803 804 805
/*
 * 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.
 */
806
static inline void blk_clear_queue_congested(struct request_queue *q, int sync)
807
{
808
	clear_bdi_congested(&q->backing_dev_info, sync);
809 810 811 812 813 814
}

/*
 * A queue has just entered congestion.  Flag that in the queue's VM-visible
 * state flags and increment the global gounter of congested queues.
 */
815
static inline void blk_set_queue_congested(struct request_queue *q, int sync)
816
{
817
	set_bdi_congested(&q->backing_dev_info, sync);
818 819
}

820 821
extern void blk_start_queue(struct request_queue *q);
extern void blk_stop_queue(struct request_queue *q);
L
Linus Torvalds 已提交
822
extern void blk_sync_queue(struct request_queue *q);
823
extern void __blk_stop_queue(struct request_queue *q);
824
extern void __blk_run_queue(struct request_queue *q);
825
extern void blk_run_queue(struct request_queue *);
826
extern void blk_run_queue_async(struct request_queue *q);
827
extern int blk_rq_map_user(struct request_queue *, struct request *,
828 829
			   struct rq_map_data *, void __user *, unsigned long,
			   gfp_t);
J
Jens Axboe 已提交
830
extern int blk_rq_unmap_user(struct bio *);
831 832
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 *,
833 834
			       struct rq_map_data *, struct sg_iovec *, int,
			       unsigned int, gfp_t);
835
extern int blk_execute_rq(struct request_queue *, struct gendisk *,
836
			  struct request *, int);
837
extern void blk_execute_rq_nowait(struct request_queue *, struct gendisk *,
838
				  struct request *, int, rq_end_io_fn *);
839

840
static inline struct request_queue *bdev_get_queue(struct block_device *bdev)
L
Linus Torvalds 已提交
841 842 843 844
{
	return bdev->bd_disk->queue;
}

845
/*
846 847 848 849 850 851
 * 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
852
 */
853 854
static inline sector_t blk_rq_pos(const struct request *rq)
{
855
	return rq->__sector;
856 857 858 859
}

static inline unsigned int blk_rq_bytes(const struct request *rq)
{
860
	return rq->__data_len;
861 862
}

863 864 865 866
static inline int blk_rq_cur_bytes(const struct request *rq)
{
	return rq->bio ? bio_cur_bytes(rq->bio) : 0;
}
867

868 869
extern unsigned int blk_rq_err_bytes(const struct request *rq);

870 871
static inline unsigned int blk_rq_sectors(const struct request *rq)
{
872
	return blk_rq_bytes(rq) >> 9;
873 874 875 876
}

static inline unsigned int blk_rq_cur_sectors(const struct request *rq)
{
877
	return blk_rq_cur_bytes(rq) >> 9;
878 879
}

880 881 882 883
static inline unsigned int blk_queue_get_max_sectors(struct request_queue *q,
						     unsigned int cmd_flags)
{
	if (unlikely(cmd_flags & REQ_DISCARD))
884
		return min(q->limits.max_discard_sectors, UINT_MAX >> 9);
885

886 887 888
	if (unlikely(cmd_flags & REQ_WRITE_SAME))
		return q->limits.max_write_same_sectors;

889 890 891 892 893 894 895 896 897 898 899 900 901
	return q->limits.max_sectors;
}

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;

	return blk_queue_get_max_sectors(q, rq->cmd_flags);
}

902 903 904 905 906 907 908 909 910 911 912
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;
}

913 914 915 916 917 918 919
/*
 * 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
920
/*
921 922 923 924 925
 * Request completion related functions.
 *
 * blk_update_request() completes given number of bytes and updates
 * the request without completing it.
 *
926 927
 * blk_end_request() and friends.  __blk_end_request() must be called
 * with the request queue spinlock acquired.
L
Linus Torvalds 已提交
928 929
 *
 * Several drivers define their own end_request and call
930 931
 * blk_end_request() for parts of the original function.
 * This prevents code duplication in drivers.
L
Linus Torvalds 已提交
932
 */
933 934
extern bool blk_update_request(struct request *rq, int error,
			       unsigned int nr_bytes);
935 936 937 938
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);
939
extern bool blk_end_request_err(struct request *rq, int error);
940 941 942 943
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);
944
extern bool __blk_end_request_err(struct request *rq, int error);
945

946
extern void blk_complete_request(struct request *);
J
Jens Axboe 已提交
947 948
extern void __blk_complete_request(struct request *);
extern void blk_abort_request(struct request *);
949
extern void blk_unprep_request(struct request *);
950

L
Linus Torvalds 已提交
951 952 953
/*
 * Access functions for manipulating queue properties
 */
954
extern struct request_queue *blk_init_queue_node(request_fn_proc *rfn,
955
					spinlock_t *lock, int node_id);
956
extern struct request_queue *blk_init_queue(request_fn_proc *, spinlock_t *);
957 958
extern struct request_queue *blk_init_allocated_queue(struct request_queue *,
						      request_fn_proc *, spinlock_t *);
959 960 961
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);
962
extern void blk_limits_max_hw_sectors(struct queue_limits *, unsigned int);
963
extern void blk_queue_max_hw_sectors(struct request_queue *, unsigned int);
964
extern void blk_queue_max_segments(struct request_queue *, unsigned short);
965
extern void blk_queue_max_segment_size(struct request_queue *, unsigned int);
966 967
extern void blk_queue_max_discard_sectors(struct request_queue *q,
		unsigned int max_discard_sectors);
968 969
extern void blk_queue_max_write_same_sectors(struct request_queue *q,
		unsigned int max_write_same_sectors);
970
extern void blk_queue_logical_block_size(struct request_queue *, unsigned short);
971
extern void blk_queue_physical_block_size(struct request_queue *, unsigned int);
972 973
extern void blk_queue_alignment_offset(struct request_queue *q,
				       unsigned int alignment);
974
extern void blk_limits_io_min(struct queue_limits *limits, unsigned int min);
975
extern void blk_queue_io_min(struct request_queue *q, unsigned int min);
976
extern void blk_limits_io_opt(struct queue_limits *limits, unsigned int opt);
977
extern void blk_queue_io_opt(struct request_queue *q, unsigned int opt);
978
extern void blk_set_default_limits(struct queue_limits *lim);
979
extern void blk_set_stacking_limits(struct queue_limits *lim);
980 981
extern int blk_stack_limits(struct queue_limits *t, struct queue_limits *b,
			    sector_t offset);
982 983
extern int bdev_stack_limits(struct queue_limits *t, struct block_device *bdev,
			    sector_t offset);
984 985
extern void disk_stack_limits(struct gendisk *disk, struct block_device *bdev,
			      sector_t offset);
986
extern void blk_queue_stack_limits(struct request_queue *t, struct request_queue *b);
987
extern void blk_queue_dma_pad(struct request_queue *, unsigned int);
988
extern void blk_queue_update_dma_pad(struct request_queue *, unsigned int);
989 990 991
extern int blk_queue_dma_drain(struct request_queue *q,
			       dma_drain_needed_fn *dma_drain_needed,
			       void *buf, unsigned int size);
992
extern void blk_queue_lld_busy(struct request_queue *q, lld_busy_fn *fn);
993 994
extern void blk_queue_segment_boundary(struct request_queue *, unsigned long);
extern void blk_queue_prep_rq(struct request_queue *, prep_rq_fn *pfn);
995
extern void blk_queue_unprep_rq(struct request_queue *, unprep_rq_fn *ufn);
996 997
extern void blk_queue_merge_bvec(struct request_queue *, merge_bvec_fn *);
extern void blk_queue_dma_alignment(struct request_queue *, int);
998
extern void blk_queue_update_dma_alignment(struct request_queue *, int);
999
extern void blk_queue_softirq_done(struct request_queue *, softirq_done_fn *);
J
Jens Axboe 已提交
1000 1001
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);
1002
extern void blk_queue_flush(struct request_queue *q, unsigned int flush);
1003
extern void blk_queue_flush_queueable(struct request_queue *q, bool queueable);
L
Linus Torvalds 已提交
1004 1005
extern struct backing_dev_info *blk_get_backing_dev_info(struct block_device *bdev);

1006
extern int blk_rq_map_sg(struct request_queue *, struct request *, struct scatterlist *);
A
Asias He 已提交
1007 1008
extern int blk_bio_map_sg(struct request_queue *q, struct bio *bio,
			  struct scatterlist *sglist);
L
Linus Torvalds 已提交
1009 1010 1011
extern void blk_dump_rq_flags(struct request *, char *);
extern long nr_blockdev_pages(void);

T
Tejun Heo 已提交
1012
bool __must_check blk_get_queue(struct request_queue *);
1013 1014 1015
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
1016

L
Lin Ming 已提交
1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037
/*
 * block layer runtime pm functions
 */
#ifdef CONFIG_PM_RUNTIME
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

1038
/*
S
Suresh Jayaraman 已提交
1039 1040 1041 1042 1043 1044 1045 1046 1047 1048
 * 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.
1049
 */
1050
struct blk_plug {
S
Suresh Jayaraman 已提交
1051 1052
	unsigned long magic; /* detect uninitialized use-cases */
	struct list_head list; /* requests */
1053
	struct list_head mq_list; /* blk-mq requests */
S
Suresh Jayaraman 已提交
1054
	struct list_head cb_list; /* md requires an unplug callback */
1055
};
1056 1057
#define BLK_MAX_REQUEST_COUNT 16

1058
struct blk_plug_cb;
1059
typedef void (*blk_plug_cb_fn)(struct blk_plug_cb *, bool);
1060 1061
struct blk_plug_cb {
	struct list_head list;
1062 1063
	blk_plug_cb_fn callback;
	void *data;
1064
};
1065 1066
extern struct blk_plug_cb *blk_check_plugged(blk_plug_cb_fn unplug,
					     void *data, int size);
1067 1068
extern void blk_start_plug(struct blk_plug *);
extern void blk_finish_plug(struct blk_plug *);
1069
extern void blk_flush_plug_list(struct blk_plug *, bool);
1070 1071 1072 1073 1074

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

1075 1076 1077 1078 1079 1080 1081 1082
	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;

1083
	if (plug)
1084
		blk_flush_plug_list(plug, true);
1085 1086 1087 1088 1089 1090
}

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

1091 1092 1093 1094
	return plug &&
		(!list_empty(&plug->list) ||
		 !list_empty(&plug->mq_list) ||
		 !list_empty(&plug->cb_list));
1095 1096
}

L
Linus Torvalds 已提交
1097 1098 1099
/*
 * tag stuff
 */
1100
#define blk_rq_tagged(rq)		((rq)->cmd_flags & REQ_QUEUED)
1101 1102 1103 1104 1105 1106 1107
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 *);
extern int blk_queue_init_tags(struct request_queue *, int, struct blk_queue_tag *);
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 *);
1108 1109
extern struct blk_queue_tag *blk_init_tags(int);
extern void blk_free_tags(struct blk_queue_tag *);
L
Linus Torvalds 已提交
1110

1111 1112 1113 1114 1115 1116 1117
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];
}
1118 1119 1120 1121

#define BLKDEV_DISCARD_SECURE  0x01    /* secure discard */

extern int blkdev_issue_flush(struct block_device *, gfp_t, sector_t *);
1122 1123
extern int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
		sector_t nr_sects, gfp_t gfp_mask, unsigned long flags);
1124 1125
extern int blkdev_issue_write_same(struct block_device *bdev, sector_t sector,
		sector_t nr_sects, gfp_t gfp_mask, struct page *page);
1126
extern int blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
1127
			sector_t nr_sects, gfp_t gfp_mask);
1128 1129
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
1130
{
1131 1132 1133
	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
1134
}
1135
static inline int sb_issue_zeroout(struct super_block *sb, sector_t block,
1136
		sector_t nr_blocks, gfp_t gfp_mask)
1137 1138 1139 1140
{
	return blkdev_issue_zeroout(sb->s_bdev,
				    block << (sb->s_blocksize_bits - 9),
				    nr_blocks << (sb->s_blocksize_bits - 9),
1141
				    gfp_mask);
1142
}
L
Linus Torvalds 已提交
1143

1144
extern int blk_verify_command(unsigned char *cmd, fmode_t has_write_perm);
1145

1146 1147 1148 1149 1150 1151 1152
enum blk_default_limits {
	BLK_MAX_SEGMENTS	= 128,
	BLK_SAFE_MAX_SECTORS	= 255,
	BLK_DEF_MAX_SECTORS	= 1024,
	BLK_MAX_SEGMENT_SIZE	= 65536,
	BLK_SEG_BOUNDARY_MASK	= 0xFFFFFFFFUL,
};
1153

L
Linus Torvalds 已提交
1154 1155
#define blkdev_entry_to_request(entry) list_entry((entry), struct request, queuelist)

1156 1157
static inline unsigned long queue_bounce_pfn(struct request_queue *q)
{
1158
	return q->limits.bounce_pfn;
1159 1160 1161 1162
}

static inline unsigned long queue_segment_boundary(struct request_queue *q)
{
1163
	return q->limits.seg_boundary_mask;
1164 1165 1166 1167
}

static inline unsigned int queue_max_sectors(struct request_queue *q)
{
1168
	return q->limits.max_sectors;
1169 1170 1171 1172
}

static inline unsigned int queue_max_hw_sectors(struct request_queue *q)
{
1173
	return q->limits.max_hw_sectors;
1174 1175
}

1176
static inline unsigned short queue_max_segments(struct request_queue *q)
1177
{
1178
	return q->limits.max_segments;
1179 1180 1181 1182
}

static inline unsigned int queue_max_segment_size(struct request_queue *q)
{
1183
	return q->limits.max_segment_size;
1184 1185
}

1186
static inline unsigned short queue_logical_block_size(struct request_queue *q)
L
Linus Torvalds 已提交
1187 1188 1189
{
	int retval = 512;

1190 1191
	if (q && q->limits.logical_block_size)
		retval = q->limits.logical_block_size;
L
Linus Torvalds 已提交
1192 1193 1194 1195

	return retval;
}

1196
static inline unsigned short bdev_logical_block_size(struct block_device *bdev)
L
Linus Torvalds 已提交
1197
{
1198
	return queue_logical_block_size(bdev_get_queue(bdev));
L
Linus Torvalds 已提交
1199 1200
}

1201 1202 1203 1204 1205
static inline unsigned int queue_physical_block_size(struct request_queue *q)
{
	return q->limits.physical_block_size;
}

1206
static inline unsigned int bdev_physical_block_size(struct block_device *bdev)
M
Martin K. Petersen 已提交
1207 1208 1209 1210
{
	return queue_physical_block_size(bdev_get_queue(bdev));
}

1211 1212 1213 1214 1215
static inline unsigned int queue_io_min(struct request_queue *q)
{
	return q->limits.io_min;
}

M
Martin K. Petersen 已提交
1216 1217 1218 1219 1220
static inline int bdev_io_min(struct block_device *bdev)
{
	return queue_io_min(bdev_get_queue(bdev));
}

1221 1222 1223 1224 1225
static inline unsigned int queue_io_opt(struct request_queue *q)
{
	return q->limits.io_opt;
}

M
Martin K. Petersen 已提交
1226 1227 1228 1229 1230
static inline int bdev_io_opt(struct block_device *bdev)
{
	return queue_io_opt(bdev_get_queue(bdev));
}

1231 1232
static inline int queue_alignment_offset(struct request_queue *q)
{
M
Martin K. Petersen 已提交
1233
	if (q->limits.misaligned)
1234 1235
		return -1;

M
Martin K. Petersen 已提交
1236
	return q->limits.alignment_offset;
1237 1238
}

1239
static inline int queue_limit_alignment_offset(struct queue_limits *lim, sector_t sector)
1240 1241
{
	unsigned int granularity = max(lim->physical_block_size, lim->io_min);
1242
	unsigned int alignment = (sector << 9) & (granularity - 1);
1243

1244 1245
	return (granularity + lim->alignment_offset - alignment)
		& (granularity - 1);
1246 1247
}

M
Martin K. Petersen 已提交
1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260
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;
}

1261 1262 1263 1264 1265 1266 1267 1268
static inline int queue_discard_alignment(struct request_queue *q)
{
	if (q->limits.discard_misaligned)
		return -1;

	return q->limits.discard_alignment;
}

1269
static inline int queue_limit_discard_alignment(struct queue_limits *lim, sector_t sector)
1270
{
1271
	unsigned int alignment, granularity, offset;
1272

1273 1274 1275
	if (!lim->max_discard_sectors)
		return 0;

1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289
	/* 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;
1290 1291
}

1292 1293 1294 1295 1296 1297 1298 1299 1300 1301
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;
}

1302 1303
static inline unsigned int queue_discard_zeroes_data(struct request_queue *q)
{
1304
	if (q->limits.max_discard_sectors && q->limits.discard_zeroes_data == 1)
1305 1306 1307 1308 1309 1310 1311 1312 1313 1314
		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));
}

1315 1316 1317 1318 1319 1320 1321 1322 1323 1324
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;
}

1325
static inline int queue_dma_alignment(struct request_queue *q)
L
Linus Torvalds 已提交
1326
{
1327
	return q ? q->dma_alignment : 511;
L
Linus Torvalds 已提交
1328 1329
}

1330
static inline int blk_rq_aligned(struct request_queue *q, unsigned long addr,
1331 1332 1333
				 unsigned int len)
{
	unsigned int alignment = queue_dma_alignment(q) | q->dma_pad_mask;
1334
	return !(addr & alignment) && !(len & alignment);
1335 1336
}

L
Linus Torvalds 已提交
1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347
/* 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;
}

1348
static inline unsigned int block_size(struct block_device *bdev)
L
Linus Torvalds 已提交
1349 1350 1351 1352
{
	return bdev->bd_block_size;
}

1353 1354 1355 1356 1357
static inline bool queue_flush_queueable(struct request_queue *q)
{
	return !q->flush_not_queueable;
}

L
Linus Torvalds 已提交
1358 1359 1360 1361 1362 1363 1364 1365 1366 1367
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;
1368
int kblockd_schedule_work(struct request_queue *q, struct work_struct *work);
1369
int kblockd_schedule_delayed_work(struct request_queue *q, struct delayed_work *dwork, unsigned long delay);
L
Linus Torvalds 已提交
1370

1371
#ifdef CONFIG_BLK_CGROUP
1372 1373 1374 1375 1376
/*
 * 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().
 */
1377 1378
static inline void set_start_time_ns(struct request *req)
{
1379
	preempt_disable();
1380
	req->start_time_ns = sched_clock();
1381
	preempt_enable();
1382 1383 1384 1385
}

static inline void set_io_start_time_ns(struct request *req)
{
1386
	preempt_disable();
1387
	req->io_start_time_ns = sched_clock();
1388
	preempt_enable();
1389
}
1390 1391 1392 1393 1394 1395 1396 1397 1398 1399

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;
}
1400 1401 1402
#else
static inline void set_start_time_ns(struct request *req) {}
static inline void set_io_start_time_ns(struct request *req) {}
1403 1404 1405 1406 1407 1408 1409 1410
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;
}
1411 1412
#endif

L
Linus Torvalds 已提交
1413 1414 1415 1416 1417
#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) "-*")

1418 1419
#if defined(CONFIG_BLK_DEV_INTEGRITY)

1420 1421
#define INTEGRITY_FLAG_READ	2	/* verify data integrity on read */
#define INTEGRITY_FLAG_WRITE	4	/* generate data integrity on write */
1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452

struct blk_integrity_exchg {
	void			*prot_buf;
	void			*data_buf;
	sector_t		sector;
	unsigned int		data_size;
	unsigned short		sector_size;
	const char		*disk_name;
};

typedef void (integrity_gen_fn) (struct blk_integrity_exchg *);
typedef int (integrity_vrfy_fn) (struct blk_integrity_exchg *);
typedef void (integrity_set_tag_fn) (void *, void *, unsigned int);
typedef void (integrity_get_tag_fn) (void *, void *, unsigned int);

struct blk_integrity {
	integrity_gen_fn	*generate_fn;
	integrity_vrfy_fn	*verify_fn;
	integrity_set_tag_fn	*set_tag_fn;
	integrity_get_tag_fn	*get_tag_fn;

	unsigned short		flags;
	unsigned short		tuple_size;
	unsigned short		sector_size;
	unsigned short		tag_size;

	const char		*name;

	struct kobject		kobj;
};

1453
extern bool blk_integrity_is_initialized(struct gendisk *);
1454 1455
extern int blk_integrity_register(struct gendisk *, struct blk_integrity *);
extern void blk_integrity_unregister(struct gendisk *);
1456
extern int blk_integrity_compare(struct gendisk *, struct gendisk *);
1457 1458 1459 1460 1461 1462 1463
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 *);
extern int blk_integrity_merge_rq(struct request_queue *, struct request *,
				  struct request *);
extern int blk_integrity_merge_bio(struct request_queue *, struct request *,
				   struct bio *);
1464

1465 1466 1467 1468 1469 1470
static inline
struct blk_integrity *bdev_get_integrity(struct block_device *bdev)
{
	return bdev->bd_disk->integrity;
}

1471 1472 1473 1474 1475
static inline struct blk_integrity *blk_get_integrity(struct gendisk *disk)
{
	return disk->integrity;
}

1476 1477
static inline int blk_integrity_rq(struct request *rq)
{
1478 1479 1480
	if (rq->bio == NULL)
		return 0;

1481 1482 1483
	return bio_integrity(rq->bio);
}

1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495
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;
}

1496 1497
#else /* CONFIG_BLK_DEV_INTEGRITY */

1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561
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)
{
	return 0;
}
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;
}
static inline int blk_integrity_merge_rq(struct request_queue *rq,
					 struct request *r1,
					 struct request *r2)
{
	return 0;
}
static inline int blk_integrity_merge_bio(struct request_queue *rq,
					  struct request *r,
					  struct bio *b)
{
	return 0;
}
static inline bool blk_integrity_is_initialized(struct gendisk *g)
{
	return 0;
}
1562 1563 1564

#endif /* CONFIG_BLK_DEV_INTEGRITY */

1565
struct block_device_operations {
A
Al Viro 已提交
1566
	int (*open) (struct block_device *, fmode_t);
1567
	void (*release) (struct gendisk *, fmode_t);
A
Al Viro 已提交
1568 1569
	int (*ioctl) (struct block_device *, fmode_t, unsigned, unsigned long);
	int (*compat_ioctl) (struct block_device *, fmode_t, unsigned, unsigned long);
1570 1571
	int (*direct_access) (struct block_device *, sector_t,
						void **, unsigned long *);
1572 1573 1574
	unsigned int (*check_events) (struct gendisk *disk,
				      unsigned int clearing);
	/* ->media_changed() is DEPRECATED, use ->check_events() instead */
1575
	int (*media_changed) (struct gendisk *);
1576
	void (*unlock_native_capacity) (struct gendisk *);
1577 1578
	int (*revalidate_disk) (struct gendisk *);
	int (*getgeo)(struct block_device *, struct hd_geometry *);
1579 1580
	/* this callback is with swap_lock and sometimes page table lock held */
	void (*swap_slot_free_notify) (struct block_device *, unsigned long);
1581 1582 1583
	struct module *owner;
};

1584 1585
extern int __blkdev_driver_ioctl(struct block_device *, fmode_t, unsigned int,
				 unsigned long);
1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596
#else /* CONFIG_BLOCK */
/*
 * stubs for when the block layer is configured out
 */
#define buffer_heads_over_limit 0

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

1597 1598 1599 1600
struct blk_plug {
};

static inline void blk_start_plug(struct blk_plug *plug)
1601 1602 1603
{
}

1604
static inline void blk_finish_plug(struct blk_plug *plug)
1605 1606 1607
{
}

1608
static inline void blk_flush_plug(struct task_struct *task)
1609 1610 1611
{
}

1612 1613 1614 1615 1616
static inline void blk_schedule_flush_plug(struct task_struct *task)
{
}


1617 1618 1619 1620 1621
static inline bool blk_needs_flush_plug(struct task_struct *tsk)
{
	return false;
}

1622 1623
#endif /* CONFIG_BLOCK */

L
Linus Torvalds 已提交
1624
#endif