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
	struct list_head queuelist;
99 100
	union {
		struct call_single_data csd;
101
		struct work_struct mq_flush_work;
102
		unsigned long fifo_time;
103
	};
104

105
	struct request_queue *q;
106
	struct blk_mq_ctx *mq_ctx;
107

J
Jens Axboe 已提交
108
	u64 cmd_flags;
109
	enum rq_cmd_type_bits cmd_type;
J
Jens Axboe 已提交
110
	unsigned long atomic_flags;
L
Linus Torvalds 已提交
111

112 113
	int cpu;

114 115
	/* the following two fields are internal, NEVER access directly */
	unsigned int __data_len;	/* total data len */
116
	sector_t __sector;		/* sector cursor */
L
Linus Torvalds 已提交
117 118 119 120

	struct bio *bio;
	struct bio *biotail;

121
	struct hlist_node hash;	/* merge hash */
122 123 124
	/*
	 * The rb_node is only used inside the io scheduler, requests
	 * are pruned when moved to the dispatch queue. So let the
125
	 * completion_data share space with the rb_node.
126 127 128
	 */
	union {
		struct rb_node rb_node;	/* sort/lookup */
129
		void *completion_data;
130
	};
131

132
	/*
133
	 * Three pointers are available for the IO schedulers, if they need
134 135
	 * more they have to dynamically allocate it.  Flush requests are
	 * never put on the IO scheduler. So let the flush fields share
136
	 * space with the elevator data.
137
	 */
138
	union {
139 140 141 142 143
		struct {
			struct io_cq		*icq;
			void			*priv[2];
		} elv;

144 145 146
		struct {
			unsigned int		seq;
			struct list_head	list;
147
			rq_end_io_fn		*saved_end_io;
148 149
		} flush;
	};
150

151
	struct gendisk *rq_disk;
152
	struct hd_struct *part;
L
Linus Torvalds 已提交
153
	unsigned long start_time;
154
#ifdef CONFIG_BLK_CGROUP
155
	struct request_list *rl;		/* rl this rq is alloced from */
156 157 158
	unsigned long long start_time_ns;
	unsigned long long io_start_time_ns;    /* when passed to hardware */
#endif
L
Linus Torvalds 已提交
159 160 161 162
	/* Number of scatter-gather DMA addr+len pairs after
	 * physical address coalescing is performed.
	 */
	unsigned short nr_phys_segments;
163 164 165
#if defined(CONFIG_BLK_DEV_INTEGRITY)
	unsigned short nr_integrity_segments;
#endif
L
Linus Torvalds 已提交
166

167 168
	unsigned short ioprio;

T
Tejun Heo 已提交
169 170
	void *special;		/* opaque pointer available for LLD use */
	char *buffer;		/* kaddr of the current segment if available */
L
Linus Torvalds 已提交
171

172 173 174
	int tag;
	int errors;

L
Linus Torvalds 已提交
175 176 177
	/*
	 * when request is used as a packet command carrier
	 */
178 179
	unsigned char __cmd[BLK_MAX_CDB];
	unsigned char *cmd;
180
	unsigned short cmd_len;
L
Linus Torvalds 已提交
181

182
	unsigned int extra_len;	/* length of alignment and padding */
L
Linus Torvalds 已提交
183
	unsigned int sense_len;
T
Tejun Heo 已提交
184
	unsigned int resid_len;	/* residual count */
L
Linus Torvalds 已提交
185 186
	void *sense;

J
Jens Axboe 已提交
187 188
	unsigned long deadline;
	struct list_head timeout_list;
L
Linus Torvalds 已提交
189
	unsigned int timeout;
190
	int retries;
L
Linus Torvalds 已提交
191 192

	/*
193
	 * completion callback.
L
Linus Torvalds 已提交
194 195 196
	 */
	rq_end_io_fn *end_io;
	void *end_io_data;
197 198 199

	/* for bidi */
	struct request *next_rq;
L
Linus Torvalds 已提交
200 201
};

202 203 204 205 206
static inline unsigned short req_get_ioprio(struct request *req)
{
	return req->ioprio;
}

L
Linus Torvalds 已提交
207
/*
208
 * State information carried for REQ_TYPE_PM_SUSPEND and REQ_TYPE_PM_RESUME
L
Linus Torvalds 已提交
209 210 211 212 213 214 215 216 217 218 219 220 221
 * 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>

222 223
struct blk_queue_ctx;

224
typedef void (request_fn_proc) (struct request_queue *q);
225
typedef void (make_request_fn) (struct request_queue *q, struct bio *bio);
226
typedef int (prep_rq_fn) (struct request_queue *, struct request *);
227
typedef void (unprep_rq_fn) (struct request_queue *, struct request *);
L
Linus Torvalds 已提交
228 229

struct bio_vec;
230 231 232 233 234 235 236 237
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 *);
238
typedef void (softirq_done_fn)(struct request *);
239
typedef int (dma_drain_needed_fn)(struct request *);
240
typedef int (lld_busy_fn) (struct request_queue *q);
M
Mike Christie 已提交
241
typedef int (bsg_job_fn) (struct bsg_job *);
L
Linus Torvalds 已提交
242

J
Jens Axboe 已提交
243 244 245 246 247 248 249 250
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
251 252 253 254 255 256 257 258 259 260
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 */
261
	int real_max_depth;		/* what the array can hold */
L
Linus Torvalds 已提交
262 263 264
	atomic_t refcnt;		/* map can be shared */
};

265 266 267
#define BLK_SCSI_MAX_CMDS	(256)
#define BLK_SCSI_CMD_PER_LONG	(BLK_SCSI_MAX_CMDS / (sizeof(long) * 8))

268 269 270 271 272 273 274
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;
275 276 277 278
	unsigned int		physical_block_size;
	unsigned int		alignment_offset;
	unsigned int		io_min;
	unsigned int		io_opt;
279
	unsigned int		max_discard_sectors;
280
	unsigned int		max_write_same_sectors;
281 282
	unsigned int		discard_granularity;
	unsigned int		discard_alignment;
283 284

	unsigned short		logical_block_size;
285
	unsigned short		max_segments;
286
	unsigned short		max_integrity_segments;
287

288
	unsigned char		misaligned;
289
	unsigned char		discard_misaligned;
290
	unsigned char		cluster;
291
	unsigned char		discard_zeroes_data;
292
	unsigned char		raid_partial_stripes_expensive;
293 294
};

295
struct request_queue {
L
Linus Torvalds 已提交
296 297 298 299 300
	/*
	 * Together with queue_head for cacheline sharing
	 */
	struct list_head	queue_head;
	struct request		*last_merge;
J
Jens Axboe 已提交
301
	struct elevator_queue	*elevator;
302 303
	int			nr_rqs[2];	/* # allocated [a]sync rqs */
	int			nr_rqs_elvpriv;	/* # allocated rqs w/ elvpriv */
L
Linus Torvalds 已提交
304 305

	/*
306 307 308 309
	 * 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
310
	 */
311
	struct request_list	root_rl;
L
Linus Torvalds 已提交
312 313 314 315

	request_fn_proc		*request_fn;
	make_request_fn		*make_request_fn;
	prep_rq_fn		*prep_rq_fn;
316
	unprep_rq_fn		*unprep_rq_fn;
L
Linus Torvalds 已提交
317
	merge_bvec_fn		*merge_bvec_fn;
318
	softirq_done_fn		*softirq_done_fn;
J
Jens Axboe 已提交
319
	rq_timed_out_fn		*rq_timed_out_fn;
320
	dma_drain_needed_fn	*dma_drain_needed;
321
	lld_busy_fn		*lld_busy_fn;
L
Linus Torvalds 已提交
322

323 324 325 326 327 328 329 330 331 332 333 334
	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;

335 336 337
	/*
	 * Dispatch queue sorting
	 */
J
Jens Axboe 已提交
338
	sector_t		end_sector;
339 340
	struct request		*boundary_rq;

L
Linus Torvalds 已提交
341
	/*
342
	 * Delayed queue handling
L
Linus Torvalds 已提交
343
	 */
344
	struct delayed_work	delay_work;
L
Linus Torvalds 已提交
345 346 347 348 349 350 351 352 353 354

	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;

	/*
355
	 * various queue flags, see QUEUE_* below
L
Linus Torvalds 已提交
356
	 */
357
	unsigned long		queue_flags;
L
Linus Torvalds 已提交
358

359 360 361 362 363 364
	/*
	 * ida allocated id for this queue.  Used to index queues from
	 * ioctx.
	 */
	int			id;

L
Linus Torvalds 已提交
365
	/*
366
	 * queue needs bounce pages for pages above this limit
L
Linus Torvalds 已提交
367
	 */
368
	gfp_t			bounce_gfp;
L
Linus Torvalds 已提交
369 370

	/*
已提交
371 372 373
	 * 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
374
	 */
已提交
375
	spinlock_t		__queue_lock;
L
Linus Torvalds 已提交
376 377 378 379 380 381 382
	spinlock_t		*queue_lock;

	/*
	 * queue kobject
	 */
	struct kobject kobj;

383 384 385 386 387
	/*
	 * mq queue kobject
	 */
	struct kobject mq_kobj;

L
Lin Ming 已提交
388 389 390 391 392 393
#ifdef CONFIG_PM_RUNTIME
	struct device		*dev;
	int			rpm_status;
	unsigned int		nr_pending;
#endif

L
Linus Torvalds 已提交
394 395 396 397 398 399 400 401
	/*
	 * 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
402
	unsigned int		dma_drain_size;
403
	void			*dma_drain_buffer;
404
	unsigned int		dma_pad_mask;
L
Linus Torvalds 已提交
405 406 407
	unsigned int		dma_alignment;

	struct blk_queue_tag	*queue_tags;
408
	struct list_head	tag_busy_list;
L
Linus Torvalds 已提交
409

410
	unsigned int		nr_sorted;
411
	unsigned int		in_flight[2];
412 413 414 415 416 417
	/*
	 * 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
418

J
Jens Axboe 已提交
419 420 421 422
	unsigned int		rq_timeout;
	struct timer_list	timeout;
	struct list_head	timeout_list;

423
	struct list_head	icq_list;
424
#ifdef CONFIG_BLK_CGROUP
425
	DECLARE_BITMAP		(blkcg_pols, BLKCG_MAX_POLS);
T
Tejun Heo 已提交
426
	struct blkcg_gq		*root_blkg;
427
	struct list_head	blkg_list;
428
#endif
429

430 431
	struct queue_limits	limits;

L
Linus Torvalds 已提交
432 433 434 435 436
	/*
	 * sg stuff
	 */
	unsigned int		sg_timeout;
	unsigned int		sg_reserved_size;
437
	int			node;
438
#ifdef CONFIG_BLK_DEV_IO_TRACE
439
	struct blk_trace	*blk_trace;
440
#endif
L
Linus Torvalds 已提交
441
	/*
442
	 * for flush operations
L
Linus Torvalds 已提交
443
	 */
444
	unsigned int		flush_flags;
445
	unsigned int		flush_not_queueable:1;
446
	unsigned int		flush_queue_delayed:1;
447 448 449 450 451
	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;
452 453
	struct request		*flush_rq;
	spinlock_t		mq_flush_lock;
454 455

	struct mutex		sysfs_lock;
456

457 458
	int			bypass_depth;

459
#if defined(CONFIG_BLK_DEV_BSG)
M
Mike Christie 已提交
460 461
	bsg_job_fn		*bsg_job_fn;
	int			bsg_job_size;
462 463
	struct bsg_class_device bsg_dev;
#endif
464 465 466 467 468

#ifdef CONFIG_BLK_DEV_THROTTLING
	/* Throttle data */
	struct throtl_data *td;
#endif
T
Tejun Heo 已提交
469
	struct rcu_head		rcu_head;
470 471 472
	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
473 474 475 476
};

#define QUEUE_FLAG_QUEUED	1	/* uses generic tag queueing */
#define QUEUE_FLAG_STOPPED	2	/* queue is stopped */
477 478
#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
479
#define QUEUE_FLAG_DYING	5	/* queue being torn down */
480
#define QUEUE_FLAG_BYPASS	6	/* act as dumb FIFO queue */
481 482
#define QUEUE_FLAG_BIDI		7	/* queue supports bidi requests */
#define QUEUE_FLAG_NOMERGES     8	/* disable merge attempts */
D
Dan Williams 已提交
483
#define QUEUE_FLAG_SAME_COMP	9	/* complete on same CPU-group */
484 485 486
#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) */
487
#define QUEUE_FLAG_VIRT        QUEUE_FLAG_NONROT /* paravirt device */
488 489 490 491 492
#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
493
#define QUEUE_FLAG_SAME_FORCE  18	/* force complete on same CPU */
494
#define QUEUE_FLAG_DEAD        19	/* queue tear-down finished */
495
#define QUEUE_FLAG_INIT_DONE   20	/* queue is initialized */
496 497

#define QUEUE_FLAG_DEFAULT	((1 << QUEUE_FLAG_IO_STAT) |		\
498
				 (1 << QUEUE_FLAG_STACKABLE)	|	\
499 500
				 (1 << QUEUE_FLAG_SAME_COMP)	|	\
				 (1 << QUEUE_FLAG_ADD_RANDOM))
501

502 503 504
#define QUEUE_FLAG_MQ_DEFAULT	((1 << QUEUE_FLAG_IO_STAT) |		\
				 (1 << QUEUE_FLAG_SAME_COMP))

505
static inline void queue_lockdep_assert_held(struct request_queue *q)
506
{
507 508
	if (q->queue_lock)
		lockdep_assert_held(q->queue_lock);
509 510
}

N
Nick Piggin 已提交
511 512 513 514 515 516
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
517 518 519
static inline int queue_flag_test_and_clear(unsigned int flag,
					    struct request_queue *q)
{
520
	queue_lockdep_assert_held(q);
J
Jens Axboe 已提交
521 522 523 524 525 526 527 528 529 530 531 532

	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)
{
533
	queue_lockdep_assert_held(q);
J
Jens Axboe 已提交
534 535 536 537 538 539 540 541 542

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

	return 1;
}

N
Nick Piggin 已提交
543 544
static inline void queue_flag_set(unsigned int flag, struct request_queue *q)
{
545
	queue_lockdep_assert_held(q);
N
Nick Piggin 已提交
546 547 548 549 550 551 552 553 554
	__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);
}

555 556 557 558 559
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
560 561
static inline void queue_flag_clear(unsigned int flag, struct request_queue *q)
{
562
	queue_lockdep_assert_held(q);
N
Nick Piggin 已提交
563 564 565
	__clear_bit(flag, &q->queue_flags);
}

L
Linus Torvalds 已提交
566 567
#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
568
#define blk_queue_dying(q)	test_bit(QUEUE_FLAG_DYING, &(q)->queue_flags)
569
#define blk_queue_dead(q)	test_bit(QUEUE_FLAG_DEAD, &(q)->queue_flags)
570
#define blk_queue_bypass(q)	test_bit(QUEUE_FLAG_BYPASS, &(q)->queue_flags)
571
#define blk_queue_init_done(q)	test_bit(QUEUE_FLAG_INIT_DONE, &(q)->queue_flags)
572
#define blk_queue_nomerges(q)	test_bit(QUEUE_FLAG_NOMERGES, &(q)->queue_flags)
573 574
#define blk_queue_noxmerges(q)	\
	test_bit(QUEUE_FLAG_NOXMERGES, &(q)->queue_flags)
575
#define blk_queue_nonrot(q)	test_bit(QUEUE_FLAG_NONROT, &(q)->queue_flags)
576
#define blk_queue_io_stat(q)	test_bit(QUEUE_FLAG_IO_STAT, &(q)->queue_flags)
577
#define blk_queue_add_random(q)	test_bit(QUEUE_FLAG_ADD_RANDOM, &(q)->queue_flags)
578 579
#define blk_queue_stackable(q)	\
	test_bit(QUEUE_FLAG_STACKABLE, &(q)->queue_flags)
580
#define blk_queue_discard(q)	test_bit(QUEUE_FLAG_DISCARD, &(q)->queue_flags)
A
Adrian Hunter 已提交
581 582
#define blk_queue_secdiscard(q)	(blk_queue_discard(q) && \
	test_bit(QUEUE_FLAG_SECDISCARD, &(q)->queue_flags))
L
Linus Torvalds 已提交
583

584 585 586 587 588 589
#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) && \
590
	 ((rq)->cmd_type == REQ_TYPE_FS))
591

L
Linus Torvalds 已提交
592
#define blk_pm_request(rq)	\
593 594
	((rq)->cmd_type == REQ_TYPE_PM_SUSPEND || \
	 (rq)->cmd_type == REQ_TYPE_PM_RESUME)
L
Linus Torvalds 已提交
595

596
#define blk_rq_cpu_valid(rq)	((rq)->cpu != -1)
597
#define blk_bidi_rq(rq)		((rq)->next_rq != NULL)
598 599
/* rq->queuelist of dequeued request must be list_empty() */
#define blk_queued_rq(rq)	(!list_empty(&(rq)->queuelist))
L
Linus Torvalds 已提交
600 601 602

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

J
Jens Axboe 已提交
603
#define rq_data_dir(rq)		(((rq)->cmd_flags & 1) != 0)
L
Linus Torvalds 已提交
604

605 606 607 608 609
static inline unsigned int blk_queue_cluster(struct request_queue *q)
{
	return q->limits.cluster;
}

610
/*
611
 * We regard a request as sync, if either a read or a sync write
612
 */
613 614
static inline bool rw_is_sync(unsigned int rw_flags)
{
615
	return !(rw_flags & REQ_WRITE) || (rw_flags & REQ_SYNC);
616 617 618 619 620 621 622
}

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

623
static inline bool blk_rl_full(struct request_list *rl, bool sync)
L
Linus Torvalds 已提交
624
{
625 626 627
	unsigned int flag = sync ? BLK_RL_SYNCFULL : BLK_RL_ASYNCFULL;

	return rl->flags & flag;
L
Linus Torvalds 已提交
628 629
}

630
static inline void blk_set_rl_full(struct request_list *rl, bool sync)
L
Linus Torvalds 已提交
631
{
632 633 634
	unsigned int flag = sync ? BLK_RL_SYNCFULL : BLK_RL_ASYNCFULL;

	rl->flags |= flag;
L
Linus Torvalds 已提交
635 636
}

637
static inline void blk_clear_rl_full(struct request_list *rl, bool sync)
L
Linus Torvalds 已提交
638
{
639 640 641
	unsigned int flag = sync ? BLK_RL_SYNCFULL : BLK_RL_ASYNCFULL;

	rl->flags &= ~flag;
L
Linus Torvalds 已提交
642 643
}

644 645 646 647
static inline bool rq_mergeable(struct request *rq)
{
	if (rq->cmd_type != REQ_TYPE_FS)
		return false;
L
Linus Torvalds 已提交
648

649 650 651 652 653
	if (rq->cmd_flags & REQ_NOMERGE_FLAGS)
		return false;

	return true;
}
L
Linus Torvalds 已提交
654

655 656 657 658 659 660 661 662 663
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;

664 665 666
	if ((flags1 & REQ_WRITE_SAME) != (flags2 & REQ_WRITE_SAME))
		return false;

667 668 669
	return true;
}

670 671 672 673 674 675 676 677
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
678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693
/*
 * 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
 */
694 695

#if BITS_PER_LONG == 32
L
Linus Torvalds 已提交
696
#define BLK_BOUNCE_HIGH		((u64)blk_max_low_pfn << PAGE_SHIFT)
697 698 699 700
#else
#define BLK_BOUNCE_HIGH		-1ULL
#endif
#define BLK_BOUNCE_ANY		(-1ULL)
701
#define BLK_BOUNCE_ISA		(DMA_BIT_MASK(24))
L
Linus Torvalds 已提交
702

703 704 705 706
/*
 * default timeout for SG_IO if none specified
 */
#define BLK_DEFAULT_SG_TIMEOUT	(60 * HZ)
L
Linus Torvalds 已提交
707
#define BLK_MIN_SG_TIMEOUT	(7 * HZ)
708

709
#ifdef CONFIG_BOUNCE
L
Linus Torvalds 已提交
710
extern int init_emergency_isa_pool(void);
711
extern void blk_queue_bounce(struct request_queue *q, struct bio **bio);
L
Linus Torvalds 已提交
712 713 714 715 716
#else
static inline int init_emergency_isa_pool(void)
{
	return 0;
}
717
static inline void blk_queue_bounce(struct request_queue *q, struct bio **bio)
L
Linus Torvalds 已提交
718 719 720 721
{
}
#endif /* CONFIG_MMU */

722 723 724 725
struct rq_map_data {
	struct page **pages;
	int page_order;
	int nr_entries;
726
	unsigned long offset;
727
	int null_mapped;
728
	int from_user;
729 730
};

731
struct req_iterator {
732
	struct bvec_iter iter;
733 734 735 736
	struct bio *bio;
};

/* This should not be used directly - use rq_for_each_segment */
737 738
#define for_each_bio(_bio)		\
	for (; _bio; _bio = _bio->bi_next)
739
#define __rq_for_each_bio(_bio, rq)	\
L
Linus Torvalds 已提交
740 741 742
	if ((rq->bio))			\
		for (_bio = (rq)->bio; _bio; _bio = _bio->bi_next)

743 744
#define rq_for_each_segment(bvl, _rq, _iter)			\
	__rq_for_each_bio(_iter.bio, _rq)			\
745
		bio_for_each_segment(bvl, _iter.bio, _iter.iter)
746

K
Kent Overstreet 已提交
747
#define rq_iter_last(bvec, _iter)				\
748
		(_iter.bio->bi_next == NULL &&			\
K
Kent Overstreet 已提交
749
		 bio_iter_last(bvec, _iter.iter))
750

751 752 753 754 755 756 757 758 759 760 761
#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
762 763 764
extern int blk_register_queue(struct gendisk *disk);
extern void blk_unregister_queue(struct gendisk *disk);
extern void generic_make_request(struct bio *bio);
765
extern void blk_rq_init(struct request_queue *q, struct request *rq);
L
Linus Torvalds 已提交
766
extern void blk_put_request(struct request *);
767 768
extern void __blk_put_request(struct request_queue *, struct request *);
extern struct request *blk_get_request(struct request_queue *, int, gfp_t);
769 770
extern struct request *blk_make_request(struct request_queue *, struct bio *,
					gfp_t);
771
extern void blk_requeue_request(struct request_queue *, struct request *);
772 773
extern void blk_add_request_payload(struct request *rq, struct page *page,
		unsigned int len);
774
extern int blk_rq_check_limits(struct request_queue *q, struct request *rq);
775
extern int blk_lld_busy(struct request_queue *q);
776 777 778 779 780
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);
781 782
extern int blk_insert_cloned_request(struct request_queue *q,
				     struct request *rq);
783
extern void blk_delay_queue(struct request_queue *, unsigned long);
784
extern void blk_recount_segments(struct request_queue *, struct bio *);
785
extern int scsi_verify_blk_ioctl(struct block_device *, unsigned int);
786 787
extern int scsi_cmd_blk_ioctl(struct block_device *, fmode_t,
			      unsigned int, void __user *);
788 789
extern int scsi_cmd_ioctl(struct request_queue *, struct gendisk *, fmode_t,
			  unsigned int, void __user *);
790 791
extern int sg_scsi_ioctl(struct request_queue *, struct gendisk *, fmode_t,
			 struct scsi_ioctl_command __user *);
792

793
extern void blk_queue_bio(struct request_queue *q, struct bio *bio);
C
Christoph Hellwig 已提交
794

795 796 797 798 799
/*
 * 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.
 */
800
static inline void blk_clear_queue_congested(struct request_queue *q, int sync)
801
{
802
	clear_bdi_congested(&q->backing_dev_info, sync);
803 804 805 806 807 808
}

/*
 * A queue has just entered congestion.  Flag that in the queue's VM-visible
 * state flags and increment the global gounter of congested queues.
 */
809
static inline void blk_set_queue_congested(struct request_queue *q, int sync)
810
{
811
	set_bdi_congested(&q->backing_dev_info, sync);
812 813
}

814 815
extern void blk_start_queue(struct request_queue *q);
extern void blk_stop_queue(struct request_queue *q);
L
Linus Torvalds 已提交
816
extern void blk_sync_queue(struct request_queue *q);
817
extern void __blk_stop_queue(struct request_queue *q);
818
extern void __blk_run_queue(struct request_queue *q);
819
extern void blk_run_queue(struct request_queue *);
820
extern void blk_run_queue_async(struct request_queue *q);
821
extern int blk_rq_map_user(struct request_queue *, struct request *,
822 823
			   struct rq_map_data *, void __user *, unsigned long,
			   gfp_t);
J
Jens Axboe 已提交
824
extern int blk_rq_unmap_user(struct bio *);
825 826
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 *,
827 828
			       struct rq_map_data *, struct sg_iovec *, int,
			       unsigned int, gfp_t);
829
extern int blk_execute_rq(struct request_queue *, struct gendisk *,
830
			  struct request *, int);
831
extern void blk_execute_rq_nowait(struct request_queue *, struct gendisk *,
832
				  struct request *, int, rq_end_io_fn *);
833

834
static inline struct request_queue *bdev_get_queue(struct block_device *bdev)
L
Linus Torvalds 已提交
835 836 837 838
{
	return bdev->bd_disk->queue;
}

839
/*
840 841 842 843 844 845
 * 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
846
 */
847 848
static inline sector_t blk_rq_pos(const struct request *rq)
{
849
	return rq->__sector;
850 851 852 853
}

static inline unsigned int blk_rq_bytes(const struct request *rq)
{
854
	return rq->__data_len;
855 856
}

857 858 859 860
static inline int blk_rq_cur_bytes(const struct request *rq)
{
	return rq->bio ? bio_cur_bytes(rq->bio) : 0;
}
861

862 863
extern unsigned int blk_rq_err_bytes(const struct request *rq);

864 865
static inline unsigned int blk_rq_sectors(const struct request *rq)
{
866
	return blk_rq_bytes(rq) >> 9;
867 868 869 870
}

static inline unsigned int blk_rq_cur_sectors(const struct request *rq)
{
871
	return blk_rq_cur_bytes(rq) >> 9;
872 873
}

874 875 876 877
static inline unsigned int blk_queue_get_max_sectors(struct request_queue *q,
						     unsigned int cmd_flags)
{
	if (unlikely(cmd_flags & REQ_DISCARD))
878
		return min(q->limits.max_discard_sectors, UINT_MAX >> 9);
879

880 881 882
	if (unlikely(cmd_flags & REQ_WRITE_SAME))
		return q->limits.max_write_same_sectors;

883 884 885 886 887 888 889 890 891 892 893 894 895
	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);
}

896 897 898 899 900 901 902 903 904 905 906
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;
}

907 908 909 910 911 912 913
/*
 * 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
914
/*
915 916 917 918 919
 * Request completion related functions.
 *
 * blk_update_request() completes given number of bytes and updates
 * the request without completing it.
 *
920 921
 * blk_end_request() and friends.  __blk_end_request() must be called
 * with the request queue spinlock acquired.
L
Linus Torvalds 已提交
922 923
 *
 * Several drivers define their own end_request and call
924 925
 * blk_end_request() for parts of the original function.
 * This prevents code duplication in drivers.
L
Linus Torvalds 已提交
926
 */
927 928
extern bool blk_update_request(struct request *rq, int error,
			       unsigned int nr_bytes);
929 930 931 932
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);
933
extern bool blk_end_request_err(struct request *rq, int error);
934 935 936 937
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);
938
extern bool __blk_end_request_err(struct request *rq, int error);
939

940
extern void blk_complete_request(struct request *);
J
Jens Axboe 已提交
941 942
extern void __blk_complete_request(struct request *);
extern void blk_abort_request(struct request *);
943
extern void blk_unprep_request(struct request *);
944

L
Linus Torvalds 已提交
945 946 947
/*
 * Access functions for manipulating queue properties
 */
948
extern struct request_queue *blk_init_queue_node(request_fn_proc *rfn,
949
					spinlock_t *lock, int node_id);
950
extern struct request_queue *blk_init_queue(request_fn_proc *, spinlock_t *);
951 952
extern struct request_queue *blk_init_allocated_queue(struct request_queue *,
						      request_fn_proc *, spinlock_t *);
953 954 955
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);
956
extern void blk_limits_max_hw_sectors(struct queue_limits *, unsigned int);
957
extern void blk_queue_max_hw_sectors(struct request_queue *, unsigned int);
958
extern void blk_queue_max_segments(struct request_queue *, unsigned short);
959
extern void blk_queue_max_segment_size(struct request_queue *, unsigned int);
960 961
extern void blk_queue_max_discard_sectors(struct request_queue *q,
		unsigned int max_discard_sectors);
962 963
extern void blk_queue_max_write_same_sectors(struct request_queue *q,
		unsigned int max_write_same_sectors);
964
extern void blk_queue_logical_block_size(struct request_queue *, unsigned short);
965
extern void blk_queue_physical_block_size(struct request_queue *, unsigned int);
966 967
extern void blk_queue_alignment_offset(struct request_queue *q,
				       unsigned int alignment);
968
extern void blk_limits_io_min(struct queue_limits *limits, unsigned int min);
969
extern void blk_queue_io_min(struct request_queue *q, unsigned int min);
970
extern void blk_limits_io_opt(struct queue_limits *limits, unsigned int opt);
971
extern void blk_queue_io_opt(struct request_queue *q, unsigned int opt);
972
extern void blk_set_default_limits(struct queue_limits *lim);
973
extern void blk_set_stacking_limits(struct queue_limits *lim);
974 975
extern int blk_stack_limits(struct queue_limits *t, struct queue_limits *b,
			    sector_t offset);
976 977
extern int bdev_stack_limits(struct queue_limits *t, struct block_device *bdev,
			    sector_t offset);
978 979
extern void disk_stack_limits(struct gendisk *disk, struct block_device *bdev,
			      sector_t offset);
980
extern void blk_queue_stack_limits(struct request_queue *t, struct request_queue *b);
981
extern void blk_queue_dma_pad(struct request_queue *, unsigned int);
982
extern void blk_queue_update_dma_pad(struct request_queue *, unsigned int);
983 984 985
extern int blk_queue_dma_drain(struct request_queue *q,
			       dma_drain_needed_fn *dma_drain_needed,
			       void *buf, unsigned int size);
986
extern void blk_queue_lld_busy(struct request_queue *q, lld_busy_fn *fn);
987 988
extern void blk_queue_segment_boundary(struct request_queue *, unsigned long);
extern void blk_queue_prep_rq(struct request_queue *, prep_rq_fn *pfn);
989
extern void blk_queue_unprep_rq(struct request_queue *, unprep_rq_fn *ufn);
990 991
extern void blk_queue_merge_bvec(struct request_queue *, merge_bvec_fn *);
extern void blk_queue_dma_alignment(struct request_queue *, int);
992
extern void blk_queue_update_dma_alignment(struct request_queue *, int);
993
extern void blk_queue_softirq_done(struct request_queue *, softirq_done_fn *);
J
Jens Axboe 已提交
994 995
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);
996
extern void blk_queue_flush(struct request_queue *q, unsigned int flush);
997
extern void blk_queue_flush_queueable(struct request_queue *q, bool queueable);
L
Linus Torvalds 已提交
998 999
extern struct backing_dev_info *blk_get_backing_dev_info(struct block_device *bdev);

1000
extern int blk_rq_map_sg(struct request_queue *, struct request *, struct scatterlist *);
A
Asias He 已提交
1001 1002
extern int blk_bio_map_sg(struct request_queue *q, struct bio *bio,
			  struct scatterlist *sglist);
L
Linus Torvalds 已提交
1003 1004 1005
extern void blk_dump_rq_flags(struct request *, char *);
extern long nr_blockdev_pages(void);

T
Tejun Heo 已提交
1006
bool __must_check blk_get_queue(struct request_queue *);
1007 1008 1009
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
1010

L
Lin Ming 已提交
1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031
/*
 * 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

1032
/*
S
Suresh Jayaraman 已提交
1033 1034 1035 1036 1037 1038 1039 1040 1041 1042
 * 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.
1043
 */
1044
struct blk_plug {
S
Suresh Jayaraman 已提交
1045 1046
	unsigned long magic; /* detect uninitialized use-cases */
	struct list_head list; /* requests */
1047
	struct list_head mq_list; /* blk-mq requests */
S
Suresh Jayaraman 已提交
1048
	struct list_head cb_list; /* md requires an unplug callback */
1049
};
1050 1051
#define BLK_MAX_REQUEST_COUNT 16

1052
struct blk_plug_cb;
1053
typedef void (*blk_plug_cb_fn)(struct blk_plug_cb *, bool);
1054 1055
struct blk_plug_cb {
	struct list_head list;
1056 1057
	blk_plug_cb_fn callback;
	void *data;
1058
};
1059 1060
extern struct blk_plug_cb *blk_check_plugged(blk_plug_cb_fn unplug,
					     void *data, int size);
1061 1062
extern void blk_start_plug(struct blk_plug *);
extern void blk_finish_plug(struct blk_plug *);
1063
extern void blk_flush_plug_list(struct blk_plug *, bool);
1064 1065 1066 1067 1068

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

1069 1070 1071 1072 1073 1074 1075 1076
	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;

1077
	if (plug)
1078
		blk_flush_plug_list(plug, true);
1079 1080 1081 1082 1083 1084
}

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

1085 1086 1087 1088
	return plug &&
		(!list_empty(&plug->list) ||
		 !list_empty(&plug->mq_list) ||
		 !list_empty(&plug->cb_list));
1089 1090
}

L
Linus Torvalds 已提交
1091 1092 1093
/*
 * tag stuff
 */
1094
#define blk_rq_tagged(rq)		((rq)->cmd_flags & REQ_QUEUED)
1095 1096 1097 1098 1099 1100 1101
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 *);
1102 1103
extern struct blk_queue_tag *blk_init_tags(int);
extern void blk_free_tags(struct blk_queue_tag *);
L
Linus Torvalds 已提交
1104

1105 1106 1107 1108 1109 1110 1111
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];
}
1112 1113 1114 1115

#define BLKDEV_DISCARD_SECURE  0x01    /* secure discard */

extern int blkdev_issue_flush(struct block_device *, gfp_t, sector_t *);
1116 1117
extern int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
		sector_t nr_sects, gfp_t gfp_mask, unsigned long flags);
1118 1119
extern int blkdev_issue_write_same(struct block_device *bdev, sector_t sector,
		sector_t nr_sects, gfp_t gfp_mask, struct page *page);
1120
extern int blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
1121
			sector_t nr_sects, gfp_t gfp_mask);
1122 1123
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
1124
{
1125 1126 1127
	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
1128
}
1129
static inline int sb_issue_zeroout(struct super_block *sb, sector_t block,
1130
		sector_t nr_blocks, gfp_t gfp_mask)
1131 1132 1133 1134
{
	return blkdev_issue_zeroout(sb->s_bdev,
				    block << (sb->s_blocksize_bits - 9),
				    nr_blocks << (sb->s_blocksize_bits - 9),
1135
				    gfp_mask);
1136
}
L
Linus Torvalds 已提交
1137

1138
extern int blk_verify_command(unsigned char *cmd, fmode_t has_write_perm);
1139

1140 1141 1142 1143 1144 1145 1146
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,
};
1147

L
Linus Torvalds 已提交
1148 1149
#define blkdev_entry_to_request(entry) list_entry((entry), struct request, queuelist)

1150 1151
static inline unsigned long queue_bounce_pfn(struct request_queue *q)
{
1152
	return q->limits.bounce_pfn;
1153 1154 1155 1156
}

static inline unsigned long queue_segment_boundary(struct request_queue *q)
{
1157
	return q->limits.seg_boundary_mask;
1158 1159 1160 1161
}

static inline unsigned int queue_max_sectors(struct request_queue *q)
{
1162
	return q->limits.max_sectors;
1163 1164 1165 1166
}

static inline unsigned int queue_max_hw_sectors(struct request_queue *q)
{
1167
	return q->limits.max_hw_sectors;
1168 1169
}

1170
static inline unsigned short queue_max_segments(struct request_queue *q)
1171
{
1172
	return q->limits.max_segments;
1173 1174 1175 1176
}

static inline unsigned int queue_max_segment_size(struct request_queue *q)
{
1177
	return q->limits.max_segment_size;
1178 1179
}

1180
static inline unsigned short queue_logical_block_size(struct request_queue *q)
L
Linus Torvalds 已提交
1181 1182 1183
{
	int retval = 512;

1184 1185
	if (q && q->limits.logical_block_size)
		retval = q->limits.logical_block_size;
L
Linus Torvalds 已提交
1186 1187 1188 1189

	return retval;
}

1190
static inline unsigned short bdev_logical_block_size(struct block_device *bdev)
L
Linus Torvalds 已提交
1191
{
1192
	return queue_logical_block_size(bdev_get_queue(bdev));
L
Linus Torvalds 已提交
1193 1194
}

1195 1196 1197 1198 1199
static inline unsigned int queue_physical_block_size(struct request_queue *q)
{
	return q->limits.physical_block_size;
}

1200
static inline unsigned int bdev_physical_block_size(struct block_device *bdev)
M
Martin K. Petersen 已提交
1201 1202 1203 1204
{
	return queue_physical_block_size(bdev_get_queue(bdev));
}

1205 1206 1207 1208 1209
static inline unsigned int queue_io_min(struct request_queue *q)
{
	return q->limits.io_min;
}

M
Martin K. Petersen 已提交
1210 1211 1212 1213 1214
static inline int bdev_io_min(struct block_device *bdev)
{
	return queue_io_min(bdev_get_queue(bdev));
}

1215 1216 1217 1218 1219
static inline unsigned int queue_io_opt(struct request_queue *q)
{
	return q->limits.io_opt;
}

M
Martin K. Petersen 已提交
1220 1221 1222 1223 1224
static inline int bdev_io_opt(struct block_device *bdev)
{
	return queue_io_opt(bdev_get_queue(bdev));
}

1225 1226
static inline int queue_alignment_offset(struct request_queue *q)
{
M
Martin K. Petersen 已提交
1227
	if (q->limits.misaligned)
1228 1229
		return -1;

M
Martin K. Petersen 已提交
1230
	return q->limits.alignment_offset;
1231 1232
}

1233
static inline int queue_limit_alignment_offset(struct queue_limits *lim, sector_t sector)
1234 1235
{
	unsigned int granularity = max(lim->physical_block_size, lim->io_min);
1236
	unsigned int alignment = (sector << 9) & (granularity - 1);
1237

1238 1239
	return (granularity + lim->alignment_offset - alignment)
		& (granularity - 1);
1240 1241
}

M
Martin K. Petersen 已提交
1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254
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;
}

1255 1256 1257 1258 1259 1260 1261 1262
static inline int queue_discard_alignment(struct request_queue *q)
{
	if (q->limits.discard_misaligned)
		return -1;

	return q->limits.discard_alignment;
}

1263
static inline int queue_limit_discard_alignment(struct queue_limits *lim, sector_t sector)
1264
{
1265
	unsigned int alignment, granularity, offset;
1266

1267 1268 1269
	if (!lim->max_discard_sectors)
		return 0;

1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283
	/* 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;
1284 1285
}

1286 1287 1288 1289 1290 1291 1292 1293 1294 1295
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;
}

1296 1297
static inline unsigned int queue_discard_zeroes_data(struct request_queue *q)
{
1298
	if (q->limits.max_discard_sectors && q->limits.discard_zeroes_data == 1)
1299 1300 1301 1302 1303 1304 1305 1306 1307 1308
		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));
}

1309 1310 1311 1312 1313 1314 1315 1316 1317 1318
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;
}

1319
static inline int queue_dma_alignment(struct request_queue *q)
L
Linus Torvalds 已提交
1320
{
1321
	return q ? q->dma_alignment : 511;
L
Linus Torvalds 已提交
1322 1323
}

1324
static inline int blk_rq_aligned(struct request_queue *q, unsigned long addr,
1325 1326 1327
				 unsigned int len)
{
	unsigned int alignment = queue_dma_alignment(q) | q->dma_pad_mask;
1328
	return !(addr & alignment) && !(len & alignment);
1329 1330
}

L
Linus Torvalds 已提交
1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341
/* 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;
}

1342
static inline unsigned int block_size(struct block_device *bdev)
L
Linus Torvalds 已提交
1343 1344 1345 1346
{
	return bdev->bd_block_size;
}

1347 1348 1349 1350 1351
static inline bool queue_flush_queueable(struct request_queue *q)
{
	return !q->flush_not_queueable;
}

L
Linus Torvalds 已提交
1352 1353 1354 1355 1356 1357 1358 1359 1360 1361
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;
1362
int kblockd_schedule_work(struct request_queue *q, struct work_struct *work);
1363
int kblockd_schedule_delayed_work(struct request_queue *q, struct delayed_work *dwork, unsigned long delay);
L
Linus Torvalds 已提交
1364

1365
#ifdef CONFIG_BLK_CGROUP
1366 1367 1368 1369 1370
/*
 * 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().
 */
1371 1372
static inline void set_start_time_ns(struct request *req)
{
1373
	preempt_disable();
1374
	req->start_time_ns = sched_clock();
1375
	preempt_enable();
1376 1377 1378 1379
}

static inline void set_io_start_time_ns(struct request *req)
{
1380
	preempt_disable();
1381
	req->io_start_time_ns = sched_clock();
1382
	preempt_enable();
1383
}
1384 1385 1386 1387 1388 1389 1390 1391 1392 1393

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;
}
1394 1395 1396
#else
static inline void set_start_time_ns(struct request *req) {}
static inline void set_io_start_time_ns(struct request *req) {}
1397 1398 1399 1400 1401 1402 1403 1404
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;
}
1405 1406
#endif

L
Linus Torvalds 已提交
1407 1408 1409 1410 1411
#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) "-*")

1412 1413
#if defined(CONFIG_BLK_DEV_INTEGRITY)

1414 1415
#define INTEGRITY_FLAG_READ	2	/* verify data integrity on read */
#define INTEGRITY_FLAG_WRITE	4	/* generate data integrity on write */
1416 1417 1418 1419 1420 1421 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

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;
};

1447
extern bool blk_integrity_is_initialized(struct gendisk *);
1448 1449
extern int blk_integrity_register(struct gendisk *, struct blk_integrity *);
extern void blk_integrity_unregister(struct gendisk *);
1450
extern int blk_integrity_compare(struct gendisk *, struct gendisk *);
1451 1452 1453 1454 1455 1456 1457
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 *);
1458

1459 1460 1461 1462 1463 1464
static inline
struct blk_integrity *bdev_get_integrity(struct block_device *bdev)
{
	return bdev->bd_disk->integrity;
}

1465 1466 1467 1468 1469
static inline struct blk_integrity *blk_get_integrity(struct gendisk *disk)
{
	return disk->integrity;
}

1470 1471
static inline int blk_integrity_rq(struct request *rq)
{
1472 1473 1474
	if (rq->bio == NULL)
		return 0;

1475 1476 1477
	return bio_integrity(rq->bio);
}

1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489
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;
}

1490 1491
#else /* CONFIG_BLK_DEV_INTEGRITY */

1492 1493 1494 1495 1496 1497 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
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;
}
1556 1557 1558

#endif /* CONFIG_BLK_DEV_INTEGRITY */

1559
struct block_device_operations {
A
Al Viro 已提交
1560
	int (*open) (struct block_device *, fmode_t);
1561
	void (*release) (struct gendisk *, fmode_t);
A
Al Viro 已提交
1562 1563
	int (*ioctl) (struct block_device *, fmode_t, unsigned, unsigned long);
	int (*compat_ioctl) (struct block_device *, fmode_t, unsigned, unsigned long);
1564 1565
	int (*direct_access) (struct block_device *, sector_t,
						void **, unsigned long *);
1566 1567 1568
	unsigned int (*check_events) (struct gendisk *disk,
				      unsigned int clearing);
	/* ->media_changed() is DEPRECATED, use ->check_events() instead */
1569
	int (*media_changed) (struct gendisk *);
1570
	void (*unlock_native_capacity) (struct gendisk *);
1571 1572
	int (*revalidate_disk) (struct gendisk *);
	int (*getgeo)(struct block_device *, struct hd_geometry *);
1573 1574
	/* this callback is with swap_lock and sometimes page table lock held */
	void (*swap_slot_free_notify) (struct block_device *, unsigned long);
1575 1576 1577
	struct module *owner;
};

1578 1579
extern int __blkdev_driver_ioctl(struct block_device *, fmode_t, unsigned int,
				 unsigned long);
1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590
#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;
}

1591 1592 1593 1594
struct blk_plug {
};

static inline void blk_start_plug(struct blk_plug *plug)
1595 1596 1597
{
}

1598
static inline void blk_finish_plug(struct blk_plug *plug)
1599 1600 1601
{
}

1602
static inline void blk_flush_plug(struct task_struct *task)
1603 1604 1605
{
}

1606 1607 1608 1609 1610
static inline void blk_schedule_flush_plug(struct task_struct *task)
{
}


1611 1612 1613 1614 1615
static inline bool blk_needs_flush_plug(struct task_struct *tsk)
{
	return false;
}

1616 1617
#endif /* CONFIG_BLOCK */

L
Linus Torvalds 已提交
1618
#endif