blkdev.h 44.6 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 11 12 13 14 15 16 17 18
#include <linux/major.h>
#include <linux/genhd.h>
#include <linux/list.h>
#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>
19
#include <linux/gfp.h>
20
#include <linux/bsg.h>
21
#include <linux/smp.h>
T
Tejun Heo 已提交
22
#include <linux/rcupdate.h>
L
Linus Torvalds 已提交
23 24 25

#include <asm/scatterlist.h>

26
struct module;
27 28
struct scsi_ioctl_command;

L
Linus Torvalds 已提交
29 30 31
struct request_queue;
struct elevator_queue;
struct request_pm_state;
32
struct blk_trace;
33 34
struct request;
struct sg_io_hdr;
M
Mike Christie 已提交
35
struct bsg_job;
T
Tejun Heo 已提交
36
struct blkcg_gq;
L
Linus Torvalds 已提交
37 38 39 40

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

T
Tejun Heo 已提交
41 42 43 44 45 46
/*
 * 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
47
struct request;
48
typedef void (rq_end_io_fn)(struct request *, int);
L
Linus Torvalds 已提交
49

50 51 52
#define BLK_RL_SYNCFULL		(1U << 0)
#define BLK_RL_ASYNCFULL	(1U << 1)

L
Linus Torvalds 已提交
53
struct request_list {
54
	struct request_queue	*q;	/* the queue this rl belongs to */
55 56 57
#ifdef CONFIG_BLK_CGROUP
	struct blkcg_gq		*blkg;	/* blkg this request pool belongs to */
#endif
58 59 60 61
	/*
	 * count[], starved[], and wait[] are indexed by
	 * BLK_RW_SYNC/BLK_RW_ASYNC
	 */
62 63 64 65
	int			count[2];
	int			starved[2];
	mempool_t		*rq_pool;
	wait_queue_head_t	wait[2];
66
	unsigned int		flags;
L
Linus Torvalds 已提交
67 68
};

69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85
/*
 * 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,
86
	REQ_TYPE_ATA_PC,
87 88
};

L
Linus Torvalds 已提交
89 90 91
#define BLK_MAX_CDB	16

/*
J
Jens Axboe 已提交
92
 * try to put the fields that are referenced together in the same cacheline.
93
 * if you modify this structure, be sure to check block/blk-core.c:blk_rq_init()
J
Jens Axboe 已提交
94
 * as well!
L
Linus Torvalds 已提交
95 96
 */
struct request {
97
	struct list_head queuelist;
98
	struct call_single_data csd;
99

100
	struct request_queue *q;
101

102 103
	unsigned int cmd_flags;
	enum rq_cmd_type_bits cmd_type;
J
Jens Axboe 已提交
104
	unsigned long atomic_flags;
L
Linus Torvalds 已提交
105

106 107
	int cpu;

108 109
	/* the following two fields are internal, NEVER access directly */
	unsigned int __data_len;	/* total data len */
110
	sector_t __sector;		/* sector cursor */
L
Linus Torvalds 已提交
111 112 113 114

	struct bio *bio;
	struct bio *biotail;

115
	struct hlist_node hash;	/* merge hash */
116 117 118
	/*
	 * The rb_node is only used inside the io scheduler, requests
	 * are pruned when moved to the dispatch queue. So let the
119
	 * completion_data share space with the rb_node.
120 121 122
	 */
	union {
		struct rb_node rb_node;	/* sort/lookup */
123
		void *completion_data;
124
	};
125

126
	/*
127
	 * Three pointers are available for the IO schedulers, if they need
128 129
	 * more they have to dynamically allocate it.  Flush requests are
	 * never put on the IO scheduler. So let the flush fields share
130
	 * space with the elevator data.
131
	 */
132
	union {
133 134 135 136 137
		struct {
			struct io_cq		*icq;
			void			*priv[2];
		} elv;

138 139 140
		struct {
			unsigned int		seq;
			struct list_head	list;
141
			rq_end_io_fn		*saved_end_io;
142 143
		} flush;
	};
144

145
	struct gendisk *rq_disk;
146
	struct hd_struct *part;
L
Linus Torvalds 已提交
147
	unsigned long start_time;
148
#ifdef CONFIG_BLK_CGROUP
149
	struct request_list *rl;		/* rl this rq is alloced from */
150 151 152
	unsigned long long start_time_ns;
	unsigned long long io_start_time_ns;    /* when passed to hardware */
#endif
L
Linus Torvalds 已提交
153 154 155 156
	/* Number of scatter-gather DMA addr+len pairs after
	 * physical address coalescing is performed.
	 */
	unsigned short nr_phys_segments;
157 158 159
#if defined(CONFIG_BLK_DEV_INTEGRITY)
	unsigned short nr_integrity_segments;
#endif
L
Linus Torvalds 已提交
160

161 162
	unsigned short ioprio;

163 164
	int ref_count;

T
Tejun Heo 已提交
165 166
	void *special;		/* opaque pointer available for LLD use */
	char *buffer;		/* kaddr of the current segment if available */
L
Linus Torvalds 已提交
167

168 169 170
	int tag;
	int errors;

L
Linus Torvalds 已提交
171 172 173
	/*
	 * when request is used as a packet command carrier
	 */
174 175
	unsigned char __cmd[BLK_MAX_CDB];
	unsigned char *cmd;
176
	unsigned short cmd_len;
L
Linus Torvalds 已提交
177

178
	unsigned int extra_len;	/* length of alignment and padding */
L
Linus Torvalds 已提交
179
	unsigned int sense_len;
T
Tejun Heo 已提交
180
	unsigned int resid_len;	/* residual count */
L
Linus Torvalds 已提交
181 182
	void *sense;

J
Jens Axboe 已提交
183 184
	unsigned long deadline;
	struct list_head timeout_list;
L
Linus Torvalds 已提交
185
	unsigned int timeout;
186
	int retries;
L
Linus Torvalds 已提交
187 188

	/*
189
	 * completion callback.
L
Linus Torvalds 已提交
190 191 192
	 */
	rq_end_io_fn *end_io;
	void *end_io_data;
193 194 195

	/* for bidi */
	struct request *next_rq;
L
Linus Torvalds 已提交
196 197
};

198 199 200 201 202
static inline unsigned short req_get_ioprio(struct request *req)
{
	return req->ioprio;
}

L
Linus Torvalds 已提交
203
/*
204
 * State information carried for REQ_TYPE_PM_SUSPEND and REQ_TYPE_PM_RESUME
L
Linus Torvalds 已提交
205 206 207 208 209 210 211 212 213 214 215 216 217
 * 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>

218
typedef void (request_fn_proc) (struct request_queue *q);
219
typedef void (make_request_fn) (struct request_queue *q, struct bio *bio);
220
typedef int (prep_rq_fn) (struct request_queue *, struct request *);
221
typedef void (unprep_rq_fn) (struct request_queue *, struct request *);
L
Linus Torvalds 已提交
222 223

struct bio_vec;
224 225 226 227 228 229 230 231
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 *);
232
typedef void (softirq_done_fn)(struct request *);
233
typedef int (dma_drain_needed_fn)(struct request *);
234
typedef int (lld_busy_fn) (struct request_queue *q);
M
Mike Christie 已提交
235
typedef int (bsg_job_fn) (struct bsg_job *);
L
Linus Torvalds 已提交
236

J
Jens Axboe 已提交
237 238 239 240 241 242 243 244
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
245 246 247 248 249 250 251 252 253 254
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 */
255
	int real_max_depth;		/* what the array can hold */
L
Linus Torvalds 已提交
256 257 258
	atomic_t refcnt;		/* map can be shared */
};

259 260 261
#define BLK_SCSI_MAX_CMDS	(256)
#define BLK_SCSI_CMD_PER_LONG	(BLK_SCSI_MAX_CMDS / (sizeof(long) * 8))

262 263 264 265 266 267 268
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;
269 270 271 272
	unsigned int		physical_block_size;
	unsigned int		alignment_offset;
	unsigned int		io_min;
	unsigned int		io_opt;
273
	unsigned int		max_discard_sectors;
274
	unsigned int		max_write_same_sectors;
275 276
	unsigned int		discard_granularity;
	unsigned int		discard_alignment;
277 278

	unsigned short		logical_block_size;
279
	unsigned short		max_segments;
280
	unsigned short		max_integrity_segments;
281

282
	unsigned char		misaligned;
283
	unsigned char		discard_misaligned;
284
	unsigned char		cluster;
285
	unsigned char		discard_zeroes_data;
286 287
};

288
struct request_queue {
L
Linus Torvalds 已提交
289 290 291 292 293
	/*
	 * Together with queue_head for cacheline sharing
	 */
	struct list_head	queue_head;
	struct request		*last_merge;
J
Jens Axboe 已提交
294
	struct elevator_queue	*elevator;
295 296
	int			nr_rqs[2];	/* # allocated [a]sync rqs */
	int			nr_rqs_elvpriv;	/* # allocated rqs w/ elvpriv */
L
Linus Torvalds 已提交
297 298

	/*
299 300 301 302
	 * 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
303
	 */
304
	struct request_list	root_rl;
L
Linus Torvalds 已提交
305 306 307 308

	request_fn_proc		*request_fn;
	make_request_fn		*make_request_fn;
	prep_rq_fn		*prep_rq_fn;
309
	unprep_rq_fn		*unprep_rq_fn;
L
Linus Torvalds 已提交
310
	merge_bvec_fn		*merge_bvec_fn;
311
	softirq_done_fn		*softirq_done_fn;
J
Jens Axboe 已提交
312
	rq_timed_out_fn		*rq_timed_out_fn;
313
	dma_drain_needed_fn	*dma_drain_needed;
314
	lld_busy_fn		*lld_busy_fn;
L
Linus Torvalds 已提交
315

316 317 318
	/*
	 * Dispatch queue sorting
	 */
J
Jens Axboe 已提交
319
	sector_t		end_sector;
320 321
	struct request		*boundary_rq;

L
Linus Torvalds 已提交
322
	/*
323
	 * Delayed queue handling
L
Linus Torvalds 已提交
324
	 */
325
	struct delayed_work	delay_work;
L
Linus Torvalds 已提交
326 327 328 329 330 331 332 333 334 335

	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;

	/*
336
	 * various queue flags, see QUEUE_* below
L
Linus Torvalds 已提交
337
	 */
338
	unsigned long		queue_flags;
L
Linus Torvalds 已提交
339

340 341 342 343 344 345
	/*
	 * ida allocated id for this queue.  Used to index queues from
	 * ioctx.
	 */
	int			id;

L
Linus Torvalds 已提交
346
	/*
347
	 * queue needs bounce pages for pages above this limit
L
Linus Torvalds 已提交
348
	 */
349
	gfp_t			bounce_gfp;
L
Linus Torvalds 已提交
350 351

	/*
已提交
352 353 354
	 * 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
355
	 */
已提交
356
	spinlock_t		__queue_lock;
L
Linus Torvalds 已提交
357 358 359 360 361 362 363
	spinlock_t		*queue_lock;

	/*
	 * queue kobject
	 */
	struct kobject kobj;

L
Lin Ming 已提交
364 365 366 367 368 369
#ifdef CONFIG_PM_RUNTIME
	struct device		*dev;
	int			rpm_status;
	unsigned int		nr_pending;
#endif

L
Linus Torvalds 已提交
370 371 372 373 374 375 376 377
	/*
	 * 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
378
	unsigned int		dma_drain_size;
379
	void			*dma_drain_buffer;
380
	unsigned int		dma_pad_mask;
L
Linus Torvalds 已提交
381 382 383
	unsigned int		dma_alignment;

	struct blk_queue_tag	*queue_tags;
384
	struct list_head	tag_busy_list;
L
Linus Torvalds 已提交
385

386
	unsigned int		nr_sorted;
387
	unsigned int		in_flight[2];
388 389 390 391 392 393
	/*
	 * 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
394

J
Jens Axboe 已提交
395 396 397 398
	unsigned int		rq_timeout;
	struct timer_list	timeout;
	struct list_head	timeout_list;

399
	struct list_head	icq_list;
400
#ifdef CONFIG_BLK_CGROUP
401
	DECLARE_BITMAP		(blkcg_pols, BLKCG_MAX_POLS);
T
Tejun Heo 已提交
402
	struct blkcg_gq		*root_blkg;
403
	struct list_head	blkg_list;
404
#endif
405

406 407
	struct queue_limits	limits;

L
Linus Torvalds 已提交
408 409 410 411 412
	/*
	 * sg stuff
	 */
	unsigned int		sg_timeout;
	unsigned int		sg_reserved_size;
413
	int			node;
414
#ifdef CONFIG_BLK_DEV_IO_TRACE
415
	struct blk_trace	*blk_trace;
416
#endif
L
Linus Torvalds 已提交
417
	/*
418
	 * for flush operations
L
Linus Torvalds 已提交
419
	 */
420
	unsigned int		flush_flags;
421
	unsigned int		flush_not_queueable:1;
422
	unsigned int		flush_queue_delayed:1;
423 424 425 426 427
	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;
428
	struct request		flush_rq;
429 430

	struct mutex		sysfs_lock;
431

432 433
	int			bypass_depth;

434
#if defined(CONFIG_BLK_DEV_BSG)
M
Mike Christie 已提交
435 436
	bsg_job_fn		*bsg_job_fn;
	int			bsg_job_size;
437 438
	struct bsg_class_device bsg_dev;
#endif
439

440 441 442
#ifdef CONFIG_BLK_CGROUP
	struct list_head	all_q_node;
#endif
443 444 445 446
#ifdef CONFIG_BLK_DEV_THROTTLING
	/* Throttle data */
	struct throtl_data *td;
#endif
T
Tejun Heo 已提交
447
	struct rcu_head		rcu_head;
L
Linus Torvalds 已提交
448 449 450 451
};

#define QUEUE_FLAG_QUEUED	1	/* uses generic tag queueing */
#define QUEUE_FLAG_STOPPED	2	/* queue is stopped */
452 453
#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
454
#define QUEUE_FLAG_DYING	5	/* queue being torn down */
455
#define QUEUE_FLAG_BYPASS	6	/* act as dumb FIFO queue */
456 457
#define QUEUE_FLAG_BIDI		7	/* queue supports bidi requests */
#define QUEUE_FLAG_NOMERGES     8	/* disable merge attempts */
D
Dan Williams 已提交
458
#define QUEUE_FLAG_SAME_COMP	9	/* complete on same CPU-group */
459 460 461
#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) */
462
#define QUEUE_FLAG_VIRT        QUEUE_FLAG_NONROT /* paravirt device */
463 464 465 466 467
#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
468
#define QUEUE_FLAG_SAME_FORCE  18	/* force complete on same CPU */
469
#define QUEUE_FLAG_DEAD        19	/* queue tear-down finished */
470 471

#define QUEUE_FLAG_DEFAULT	((1 << QUEUE_FLAG_IO_STAT) |		\
472
				 (1 << QUEUE_FLAG_STACKABLE)	|	\
473 474
				 (1 << QUEUE_FLAG_SAME_COMP)	|	\
				 (1 << QUEUE_FLAG_ADD_RANDOM))
475

476
static inline void queue_lockdep_assert_held(struct request_queue *q)
477
{
478 479
	if (q->queue_lock)
		lockdep_assert_held(q->queue_lock);
480 481
}

N
Nick Piggin 已提交
482 483 484 485 486 487
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
488 489 490
static inline int queue_flag_test_and_clear(unsigned int flag,
					    struct request_queue *q)
{
491
	queue_lockdep_assert_held(q);
J
Jens Axboe 已提交
492 493 494 495 496 497 498 499 500 501 502 503

	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)
{
504
	queue_lockdep_assert_held(q);
J
Jens Axboe 已提交
505 506 507 508 509 510 511 512 513

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

	return 1;
}

N
Nick Piggin 已提交
514 515
static inline void queue_flag_set(unsigned int flag, struct request_queue *q)
{
516
	queue_lockdep_assert_held(q);
N
Nick Piggin 已提交
517 518 519 520 521 522 523 524 525
	__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);
}

526 527 528 529 530
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
531 532
static inline void queue_flag_clear(unsigned int flag, struct request_queue *q)
{
533
	queue_lockdep_assert_held(q);
N
Nick Piggin 已提交
534 535 536
	__clear_bit(flag, &q->queue_flags);
}

L
Linus Torvalds 已提交
537 538
#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
539
#define blk_queue_dying(q)	test_bit(QUEUE_FLAG_DYING, &(q)->queue_flags)
540
#define blk_queue_dead(q)	test_bit(QUEUE_FLAG_DEAD, &(q)->queue_flags)
541
#define blk_queue_bypass(q)	test_bit(QUEUE_FLAG_BYPASS, &(q)->queue_flags)
542
#define blk_queue_nomerges(q)	test_bit(QUEUE_FLAG_NOMERGES, &(q)->queue_flags)
543 544
#define blk_queue_noxmerges(q)	\
	test_bit(QUEUE_FLAG_NOXMERGES, &(q)->queue_flags)
545
#define blk_queue_nonrot(q)	test_bit(QUEUE_FLAG_NONROT, &(q)->queue_flags)
546
#define blk_queue_io_stat(q)	test_bit(QUEUE_FLAG_IO_STAT, &(q)->queue_flags)
547
#define blk_queue_add_random(q)	test_bit(QUEUE_FLAG_ADD_RANDOM, &(q)->queue_flags)
548 549
#define blk_queue_stackable(q)	\
	test_bit(QUEUE_FLAG_STACKABLE, &(q)->queue_flags)
550
#define blk_queue_discard(q)	test_bit(QUEUE_FLAG_DISCARD, &(q)->queue_flags)
A
Adrian Hunter 已提交
551 552
#define blk_queue_secdiscard(q)	(blk_queue_discard(q) && \
	test_bit(QUEUE_FLAG_SECDISCARD, &(q)->queue_flags))
L
Linus Torvalds 已提交
553

554 555 556 557 558 559
#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) && \
560
	 ((rq)->cmd_type == REQ_TYPE_FS))
561

L
Linus Torvalds 已提交
562
#define blk_pm_request(rq)	\
563 564
	((rq)->cmd_type == REQ_TYPE_PM_SUSPEND || \
	 (rq)->cmd_type == REQ_TYPE_PM_RESUME)
L
Linus Torvalds 已提交
565

566
#define blk_rq_cpu_valid(rq)	((rq)->cpu != -1)
567
#define blk_bidi_rq(rq)		((rq)->next_rq != NULL)
568 569
/* rq->queuelist of dequeued request must be list_empty() */
#define blk_queued_rq(rq)	(!list_empty(&(rq)->queuelist))
L
Linus Torvalds 已提交
570 571 572

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

573
#define rq_data_dir(rq)		((rq)->cmd_flags & 1)
L
Linus Torvalds 已提交
574

575 576 577 578 579
static inline unsigned int blk_queue_cluster(struct request_queue *q)
{
	return q->limits.cluster;
}

580
/*
581
 * We regard a request as sync, if either a read or a sync write
582
 */
583 584
static inline bool rw_is_sync(unsigned int rw_flags)
{
585
	return !(rw_flags & REQ_WRITE) || (rw_flags & REQ_SYNC);
586 587 588 589 590 591 592
}

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

593
static inline bool blk_rl_full(struct request_list *rl, bool sync)
L
Linus Torvalds 已提交
594
{
595 596 597
	unsigned int flag = sync ? BLK_RL_SYNCFULL : BLK_RL_ASYNCFULL;

	return rl->flags & flag;
L
Linus Torvalds 已提交
598 599
}

600
static inline void blk_set_rl_full(struct request_list *rl, bool sync)
L
Linus Torvalds 已提交
601
{
602 603 604
	unsigned int flag = sync ? BLK_RL_SYNCFULL : BLK_RL_ASYNCFULL;

	rl->flags |= flag;
L
Linus Torvalds 已提交
605 606
}

607
static inline void blk_clear_rl_full(struct request_list *rl, bool sync)
L
Linus Torvalds 已提交
608
{
609 610 611
	unsigned int flag = sync ? BLK_RL_SYNCFULL : BLK_RL_ASYNCFULL;

	rl->flags &= ~flag;
L
Linus Torvalds 已提交
612 613
}

614 615 616 617
static inline bool rq_mergeable(struct request *rq)
{
	if (rq->cmd_type != REQ_TYPE_FS)
		return false;
L
Linus Torvalds 已提交
618

619 620 621 622 623
	if (rq->cmd_flags & REQ_NOMERGE_FLAGS)
		return false;

	return true;
}
L
Linus Torvalds 已提交
624

625 626 627 628 629 630 631 632 633
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;

634 635 636
	if ((flags1 & REQ_WRITE_SAME) != (flags2 & REQ_WRITE_SAME))
		return false;

637 638 639
	return true;
}

640 641 642 643 644 645 646 647
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
648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663
/*
 * 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
 */
664 665

#if BITS_PER_LONG == 32
L
Linus Torvalds 已提交
666
#define BLK_BOUNCE_HIGH		((u64)blk_max_low_pfn << PAGE_SHIFT)
667 668 669 670
#else
#define BLK_BOUNCE_HIGH		-1ULL
#endif
#define BLK_BOUNCE_ANY		(-1ULL)
671
#define BLK_BOUNCE_ISA		(DMA_BIT_MASK(24))
L
Linus Torvalds 已提交
672

673 674 675 676
/*
 * default timeout for SG_IO if none specified
 */
#define BLK_DEFAULT_SG_TIMEOUT	(60 * HZ)
L
Linus Torvalds 已提交
677
#define BLK_MIN_SG_TIMEOUT	(7 * HZ)
678

679
#ifdef CONFIG_BOUNCE
L
Linus Torvalds 已提交
680
extern int init_emergency_isa_pool(void);
681
extern void blk_queue_bounce(struct request_queue *q, struct bio **bio);
L
Linus Torvalds 已提交
682 683 684 685 686
#else
static inline int init_emergency_isa_pool(void)
{
	return 0;
}
687
static inline void blk_queue_bounce(struct request_queue *q, struct bio **bio)
L
Linus Torvalds 已提交
688 689 690 691
{
}
#endif /* CONFIG_MMU */

692 693 694 695
struct rq_map_data {
	struct page **pages;
	int page_order;
	int nr_entries;
696
	unsigned long offset;
697
	int null_mapped;
698
	int from_user;
699 700
};

701 702 703 704 705 706
struct req_iterator {
	int i;
	struct bio *bio;
};

/* This should not be used directly - use rq_for_each_segment */
707 708
#define for_each_bio(_bio)		\
	for (; _bio; _bio = _bio->bi_next)
709
#define __rq_for_each_bio(_bio, rq)	\
L
Linus Torvalds 已提交
710 711 712
	if ((rq->bio))			\
		for (_bio = (rq)->bio; _bio; _bio = _bio->bi_next)

713 714 715 716 717 718 719
#define rq_for_each_segment(bvl, _rq, _iter)			\
	__rq_for_each_bio(_iter.bio, _rq)			\
		bio_for_each_segment(bvl, _iter.bio, _iter.i)

#define rq_iter_last(rq, _iter)					\
		(_iter.bio->bi_next == NULL && _iter.i == _iter.bio->bi_vcnt-1)

720 721 722 723 724 725 726 727 728 729 730
#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
731 732 733
extern int blk_register_queue(struct gendisk *disk);
extern void blk_unregister_queue(struct gendisk *disk);
extern void generic_make_request(struct bio *bio);
734
extern void blk_rq_init(struct request_queue *q, struct request *rq);
L
Linus Torvalds 已提交
735
extern void blk_put_request(struct request *);
736 737
extern void __blk_put_request(struct request_queue *, struct request *);
extern struct request *blk_get_request(struct request_queue *, int, gfp_t);
738 739
extern struct request *blk_make_request(struct request_queue *, struct bio *,
					gfp_t);
740
extern void blk_requeue_request(struct request_queue *, struct request *);
741 742
extern void blk_add_request_payload(struct request *rq, struct page *page,
		unsigned int len);
743
extern int blk_rq_check_limits(struct request_queue *q, struct request *rq);
744
extern int blk_lld_busy(struct request_queue *q);
745 746 747 748 749
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);
750 751
extern int blk_insert_cloned_request(struct request_queue *q,
				     struct request *rq);
752
extern void blk_delay_queue(struct request_queue *, unsigned long);
753
extern void blk_recount_segments(struct request_queue *, struct bio *);
754
extern int scsi_verify_blk_ioctl(struct block_device *, unsigned int);
755 756
extern int scsi_cmd_blk_ioctl(struct block_device *, fmode_t,
			      unsigned int, void __user *);
757 758
extern int scsi_cmd_ioctl(struct request_queue *, struct gendisk *, fmode_t,
			  unsigned int, void __user *);
759 760
extern int sg_scsi_ioctl(struct request_queue *, struct gendisk *, fmode_t,
			 struct scsi_ioctl_command __user *);
761

762
extern void blk_queue_bio(struct request_queue *q, struct bio *bio);
C
Christoph Hellwig 已提交
763

764 765 766 767 768
/*
 * 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.
 */
769
static inline void blk_clear_queue_congested(struct request_queue *q, int sync)
770
{
771
	clear_bdi_congested(&q->backing_dev_info, sync);
772 773 774 775 776 777
}

/*
 * A queue has just entered congestion.  Flag that in the queue's VM-visible
 * state flags and increment the global gounter of congested queues.
 */
778
static inline void blk_set_queue_congested(struct request_queue *q, int sync)
779
{
780
	set_bdi_congested(&q->backing_dev_info, sync);
781 782
}

783 784
extern void blk_start_queue(struct request_queue *q);
extern void blk_stop_queue(struct request_queue *q);
L
Linus Torvalds 已提交
785
extern void blk_sync_queue(struct request_queue *q);
786
extern void __blk_stop_queue(struct request_queue *q);
787
extern void __blk_run_queue(struct request_queue *q);
788
extern void blk_run_queue(struct request_queue *);
789
extern void blk_run_queue_async(struct request_queue *q);
790
extern int blk_rq_map_user(struct request_queue *, struct request *,
791 792
			   struct rq_map_data *, void __user *, unsigned long,
			   gfp_t);
J
Jens Axboe 已提交
793
extern int blk_rq_unmap_user(struct bio *);
794 795
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 *,
796 797
			       struct rq_map_data *, struct sg_iovec *, int,
			       unsigned int, gfp_t);
798
extern int blk_execute_rq(struct request_queue *, struct gendisk *,
799
			  struct request *, int);
800
extern void blk_execute_rq_nowait(struct request_queue *, struct gendisk *,
801
				  struct request *, int, rq_end_io_fn *);
802

803
static inline struct request_queue *bdev_get_queue(struct block_device *bdev)
L
Linus Torvalds 已提交
804 805 806 807
{
	return bdev->bd_disk->queue;
}

808
/*
809 810 811 812 813 814
 * 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
815
 */
816 817
static inline sector_t blk_rq_pos(const struct request *rq)
{
818
	return rq->__sector;
819 820 821 822
}

static inline unsigned int blk_rq_bytes(const struct request *rq)
{
823
	return rq->__data_len;
824 825
}

826 827 828 829
static inline int blk_rq_cur_bytes(const struct request *rq)
{
	return rq->bio ? bio_cur_bytes(rq->bio) : 0;
}
830

831 832
extern unsigned int blk_rq_err_bytes(const struct request *rq);

833 834
static inline unsigned int blk_rq_sectors(const struct request *rq)
{
835
	return blk_rq_bytes(rq) >> 9;
836 837 838 839
}

static inline unsigned int blk_rq_cur_sectors(const struct request *rq)
{
840
	return blk_rq_cur_bytes(rq) >> 9;
841 842
}

843 844 845 846
static inline unsigned int blk_queue_get_max_sectors(struct request_queue *q,
						     unsigned int cmd_flags)
{
	if (unlikely(cmd_flags & REQ_DISCARD))
847
		return min(q->limits.max_discard_sectors, UINT_MAX >> 9);
848

849 850 851
	if (unlikely(cmd_flags & REQ_WRITE_SAME))
		return q->limits.max_write_same_sectors;

852 853 854 855 856 857 858 859 860 861 862 863 864
	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);
}

865 866 867 868 869 870 871
/*
 * 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
872
/*
873 874 875 876 877
 * Request completion related functions.
 *
 * blk_update_request() completes given number of bytes and updates
 * the request without completing it.
 *
878 879
 * blk_end_request() and friends.  __blk_end_request() must be called
 * with the request queue spinlock acquired.
L
Linus Torvalds 已提交
880 881
 *
 * Several drivers define their own end_request and call
882 883
 * blk_end_request() for parts of the original function.
 * This prevents code duplication in drivers.
L
Linus Torvalds 已提交
884
 */
885 886
extern bool blk_update_request(struct request *rq, int error,
			       unsigned int nr_bytes);
887 888 889 890
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);
891
extern bool blk_end_request_err(struct request *rq, int error);
892 893 894 895
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);
896
extern bool __blk_end_request_err(struct request *rq, int error);
897

898
extern void blk_complete_request(struct request *);
J
Jens Axboe 已提交
899 900
extern void __blk_complete_request(struct request *);
extern void blk_abort_request(struct request *);
901
extern void blk_unprep_request(struct request *);
902

L
Linus Torvalds 已提交
903 904 905
/*
 * Access functions for manipulating queue properties
 */
906
extern struct request_queue *blk_init_queue_node(request_fn_proc *rfn,
907
					spinlock_t *lock, int node_id);
908
extern struct request_queue *blk_init_queue(request_fn_proc *, spinlock_t *);
909 910
extern struct request_queue *blk_init_allocated_queue(struct request_queue *,
						      request_fn_proc *, spinlock_t *);
911 912 913
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);
914
extern void blk_limits_max_hw_sectors(struct queue_limits *, unsigned int);
915
extern void blk_queue_max_hw_sectors(struct request_queue *, unsigned int);
916
extern void blk_queue_max_segments(struct request_queue *, unsigned short);
917
extern void blk_queue_max_segment_size(struct request_queue *, unsigned int);
918 919
extern void blk_queue_max_discard_sectors(struct request_queue *q,
		unsigned int max_discard_sectors);
920 921
extern void blk_queue_max_write_same_sectors(struct request_queue *q,
		unsigned int max_write_same_sectors);
922
extern void blk_queue_logical_block_size(struct request_queue *, unsigned short);
923
extern void blk_queue_physical_block_size(struct request_queue *, unsigned int);
924 925
extern void blk_queue_alignment_offset(struct request_queue *q,
				       unsigned int alignment);
926
extern void blk_limits_io_min(struct queue_limits *limits, unsigned int min);
927
extern void blk_queue_io_min(struct request_queue *q, unsigned int min);
928
extern void blk_limits_io_opt(struct queue_limits *limits, unsigned int opt);
929
extern void blk_queue_io_opt(struct request_queue *q, unsigned int opt);
930
extern void blk_set_default_limits(struct queue_limits *lim);
931
extern void blk_set_stacking_limits(struct queue_limits *lim);
932 933
extern int blk_stack_limits(struct queue_limits *t, struct queue_limits *b,
			    sector_t offset);
934 935
extern int bdev_stack_limits(struct queue_limits *t, struct block_device *bdev,
			    sector_t offset);
936 937
extern void disk_stack_limits(struct gendisk *disk, struct block_device *bdev,
			      sector_t offset);
938
extern void blk_queue_stack_limits(struct request_queue *t, struct request_queue *b);
939
extern void blk_queue_dma_pad(struct request_queue *, unsigned int);
940
extern void blk_queue_update_dma_pad(struct request_queue *, unsigned int);
941 942 943
extern int blk_queue_dma_drain(struct request_queue *q,
			       dma_drain_needed_fn *dma_drain_needed,
			       void *buf, unsigned int size);
944
extern void blk_queue_lld_busy(struct request_queue *q, lld_busy_fn *fn);
945 946
extern void blk_queue_segment_boundary(struct request_queue *, unsigned long);
extern void blk_queue_prep_rq(struct request_queue *, prep_rq_fn *pfn);
947
extern void blk_queue_unprep_rq(struct request_queue *, unprep_rq_fn *ufn);
948 949
extern void blk_queue_merge_bvec(struct request_queue *, merge_bvec_fn *);
extern void blk_queue_dma_alignment(struct request_queue *, int);
950
extern void blk_queue_update_dma_alignment(struct request_queue *, int);
951
extern void blk_queue_softirq_done(struct request_queue *, softirq_done_fn *);
J
Jens Axboe 已提交
952 953
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);
954
extern void blk_queue_flush(struct request_queue *q, unsigned int flush);
955
extern void blk_queue_flush_queueable(struct request_queue *q, bool queueable);
L
Linus Torvalds 已提交
956 957
extern struct backing_dev_info *blk_get_backing_dev_info(struct block_device *bdev);

958
extern int blk_rq_map_sg(struct request_queue *, struct request *, struct scatterlist *);
A
Asias He 已提交
959 960
extern int blk_bio_map_sg(struct request_queue *q, struct bio *bio,
			  struct scatterlist *sglist);
L
Linus Torvalds 已提交
961 962 963
extern void blk_dump_rq_flags(struct request *, char *);
extern long nr_blockdev_pages(void);

T
Tejun Heo 已提交
964
bool __must_check blk_get_queue(struct request_queue *);
965 966 967
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
968

L
Lin Ming 已提交
969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989
/*
 * 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

990
/*
S
Suresh Jayaraman 已提交
991 992 993 994 995 996 997 998 999 1000
 * 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.
1001
 */
1002
struct blk_plug {
S
Suresh Jayaraman 已提交
1003 1004 1005
	unsigned long magic; /* detect uninitialized use-cases */
	struct list_head list; /* requests */
	struct list_head cb_list; /* md requires an unplug callback */
1006
};
1007 1008
#define BLK_MAX_REQUEST_COUNT 16

1009
struct blk_plug_cb;
1010
typedef void (*blk_plug_cb_fn)(struct blk_plug_cb *, bool);
1011 1012
struct blk_plug_cb {
	struct list_head list;
1013 1014
	blk_plug_cb_fn callback;
	void *data;
1015
};
1016 1017
extern struct blk_plug_cb *blk_check_plugged(blk_plug_cb_fn unplug,
					     void *data, int size);
1018 1019
extern void blk_start_plug(struct blk_plug *);
extern void blk_finish_plug(struct blk_plug *);
1020
extern void blk_flush_plug_list(struct blk_plug *, bool);
1021 1022 1023 1024 1025

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

1026 1027 1028 1029 1030 1031 1032 1033
	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;

1034
	if (plug)
1035
		blk_flush_plug_list(plug, true);
1036 1037 1038 1039 1040 1041
}

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

1042
	return plug && (!list_empty(&plug->list) || !list_empty(&plug->cb_list));
1043 1044
}

L
Linus Torvalds 已提交
1045 1046 1047
/*
 * tag stuff
 */
1048
#define blk_rq_tagged(rq)		((rq)->cmd_flags & REQ_QUEUED)
1049 1050 1051 1052 1053 1054 1055
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 *);
1056 1057
extern struct blk_queue_tag *blk_init_tags(int);
extern void blk_free_tags(struct blk_queue_tag *);
L
Linus Torvalds 已提交
1058

1059 1060 1061 1062 1063 1064 1065
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];
}
1066 1067 1068 1069

#define BLKDEV_DISCARD_SECURE  0x01    /* secure discard */

extern int blkdev_issue_flush(struct block_device *, gfp_t, sector_t *);
1070 1071
extern int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
		sector_t nr_sects, gfp_t gfp_mask, unsigned long flags);
1072 1073
extern int blkdev_issue_write_same(struct block_device *bdev, sector_t sector,
		sector_t nr_sects, gfp_t gfp_mask, struct page *page);
1074
extern int blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
1075
			sector_t nr_sects, gfp_t gfp_mask);
1076 1077
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
1078
{
1079 1080 1081
	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
1082
}
1083
static inline int sb_issue_zeroout(struct super_block *sb, sector_t block,
1084
		sector_t nr_blocks, gfp_t gfp_mask)
1085 1086 1087 1088
{
	return blkdev_issue_zeroout(sb->s_bdev,
				    block << (sb->s_blocksize_bits - 9),
				    nr_blocks << (sb->s_blocksize_bits - 9),
1089
				    gfp_mask);
1090
}
L
Linus Torvalds 已提交
1091

1092
extern int blk_verify_command(unsigned char *cmd, fmode_t has_write_perm);
1093

1094 1095 1096 1097 1098 1099 1100
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,
};
1101

L
Linus Torvalds 已提交
1102 1103
#define blkdev_entry_to_request(entry) list_entry((entry), struct request, queuelist)

1104 1105
static inline unsigned long queue_bounce_pfn(struct request_queue *q)
{
1106
	return q->limits.bounce_pfn;
1107 1108 1109 1110
}

static inline unsigned long queue_segment_boundary(struct request_queue *q)
{
1111
	return q->limits.seg_boundary_mask;
1112 1113 1114 1115
}

static inline unsigned int queue_max_sectors(struct request_queue *q)
{
1116
	return q->limits.max_sectors;
1117 1118 1119 1120
}

static inline unsigned int queue_max_hw_sectors(struct request_queue *q)
{
1121
	return q->limits.max_hw_sectors;
1122 1123
}

1124
static inline unsigned short queue_max_segments(struct request_queue *q)
1125
{
1126
	return q->limits.max_segments;
1127 1128 1129 1130
}

static inline unsigned int queue_max_segment_size(struct request_queue *q)
{
1131
	return q->limits.max_segment_size;
1132 1133
}

1134
static inline unsigned short queue_logical_block_size(struct request_queue *q)
L
Linus Torvalds 已提交
1135 1136 1137
{
	int retval = 512;

1138 1139
	if (q && q->limits.logical_block_size)
		retval = q->limits.logical_block_size;
L
Linus Torvalds 已提交
1140 1141 1142 1143

	return retval;
}

1144
static inline unsigned short bdev_logical_block_size(struct block_device *bdev)
L
Linus Torvalds 已提交
1145
{
1146
	return queue_logical_block_size(bdev_get_queue(bdev));
L
Linus Torvalds 已提交
1147 1148
}

1149 1150 1151 1152 1153
static inline unsigned int queue_physical_block_size(struct request_queue *q)
{
	return q->limits.physical_block_size;
}

1154
static inline unsigned int bdev_physical_block_size(struct block_device *bdev)
M
Martin K. Petersen 已提交
1155 1156 1157 1158
{
	return queue_physical_block_size(bdev_get_queue(bdev));
}

1159 1160 1161 1162 1163
static inline unsigned int queue_io_min(struct request_queue *q)
{
	return q->limits.io_min;
}

M
Martin K. Petersen 已提交
1164 1165 1166 1167 1168
static inline int bdev_io_min(struct block_device *bdev)
{
	return queue_io_min(bdev_get_queue(bdev));
}

1169 1170 1171 1172 1173
static inline unsigned int queue_io_opt(struct request_queue *q)
{
	return q->limits.io_opt;
}

M
Martin K. Petersen 已提交
1174 1175 1176 1177 1178
static inline int bdev_io_opt(struct block_device *bdev)
{
	return queue_io_opt(bdev_get_queue(bdev));
}

1179 1180
static inline int queue_alignment_offset(struct request_queue *q)
{
M
Martin K. Petersen 已提交
1181
	if (q->limits.misaligned)
1182 1183
		return -1;

M
Martin K. Petersen 已提交
1184
	return q->limits.alignment_offset;
1185 1186
}

1187
static inline int queue_limit_alignment_offset(struct queue_limits *lim, sector_t sector)
1188 1189
{
	unsigned int granularity = max(lim->physical_block_size, lim->io_min);
1190
	unsigned int alignment = (sector << 9) & (granularity - 1);
1191

1192 1193
	return (granularity + lim->alignment_offset - alignment)
		& (granularity - 1);
1194 1195
}

M
Martin K. Petersen 已提交
1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208
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;
}

1209 1210 1211 1212 1213 1214 1215 1216
static inline int queue_discard_alignment(struct request_queue *q)
{
	if (q->limits.discard_misaligned)
		return -1;

	return q->limits.discard_alignment;
}

1217
static inline int queue_limit_discard_alignment(struct queue_limits *lim, sector_t sector)
1218
{
1219
	unsigned int alignment, granularity, offset;
1220

1221 1222 1223
	if (!lim->max_discard_sectors)
		return 0;

1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237
	/* 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;
1238 1239
}

1240 1241 1242 1243 1244 1245 1246 1247 1248 1249
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;
}

1250 1251
static inline unsigned int queue_discard_zeroes_data(struct request_queue *q)
{
1252
	if (q->limits.max_discard_sectors && q->limits.discard_zeroes_data == 1)
1253 1254 1255 1256 1257 1258 1259 1260 1261 1262
		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));
}

1263 1264 1265 1266 1267 1268 1269 1270 1271 1272
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;
}

1273
static inline int queue_dma_alignment(struct request_queue *q)
L
Linus Torvalds 已提交
1274
{
1275
	return q ? q->dma_alignment : 511;
L
Linus Torvalds 已提交
1276 1277
}

1278
static inline int blk_rq_aligned(struct request_queue *q, unsigned long addr,
1279 1280 1281
				 unsigned int len)
{
	unsigned int alignment = queue_dma_alignment(q) | q->dma_pad_mask;
1282
	return !(addr & alignment) && !(len & alignment);
1283 1284
}

L
Linus Torvalds 已提交
1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295
/* 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;
}

1296
static inline unsigned int block_size(struct block_device *bdev)
L
Linus Torvalds 已提交
1297 1298 1299 1300
{
	return bdev->bd_block_size;
}

1301 1302 1303 1304 1305
static inline bool queue_flush_queueable(struct request_queue *q)
{
	return !q->flush_not_queueable;
}

L
Linus Torvalds 已提交
1306 1307 1308 1309 1310 1311 1312 1313 1314 1315
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;
1316
int kblockd_schedule_work(struct request_queue *q, struct work_struct *work);
L
Linus Torvalds 已提交
1317

1318
#ifdef CONFIG_BLK_CGROUP
1319 1320 1321 1322 1323
/*
 * 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().
 */
1324 1325
static inline void set_start_time_ns(struct request *req)
{
1326
	preempt_disable();
1327
	req->start_time_ns = sched_clock();
1328
	preempt_enable();
1329 1330 1331 1332
}

static inline void set_io_start_time_ns(struct request *req)
{
1333
	preempt_disable();
1334
	req->io_start_time_ns = sched_clock();
1335
	preempt_enable();
1336
}
1337 1338 1339 1340 1341 1342 1343 1344 1345 1346

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;
}
1347 1348 1349
#else
static inline void set_start_time_ns(struct request *req) {}
static inline void set_io_start_time_ns(struct request *req) {}
1350 1351 1352 1353 1354 1355 1356 1357
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;
}
1358 1359
#endif

L
Linus Torvalds 已提交
1360 1361 1362 1363 1364
#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) "-*")

1365 1366
#if defined(CONFIG_BLK_DEV_INTEGRITY)

1367 1368
#define INTEGRITY_FLAG_READ	2	/* verify data integrity on read */
#define INTEGRITY_FLAG_WRITE	4	/* generate data integrity on write */
1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399

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;
};

1400
extern bool blk_integrity_is_initialized(struct gendisk *);
1401 1402
extern int blk_integrity_register(struct gendisk *, struct blk_integrity *);
extern void blk_integrity_unregister(struct gendisk *);
1403
extern int blk_integrity_compare(struct gendisk *, struct gendisk *);
1404 1405 1406 1407 1408 1409 1410
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 *);
1411

1412 1413 1414 1415 1416 1417
static inline
struct blk_integrity *bdev_get_integrity(struct block_device *bdev)
{
	return bdev->bd_disk->integrity;
}

1418 1419 1420 1421 1422
static inline struct blk_integrity *blk_get_integrity(struct gendisk *disk)
{
	return disk->integrity;
}

1423 1424
static inline int blk_integrity_rq(struct request *rq)
{
1425 1426 1427
	if (rq->bio == NULL)
		return 0;

1428 1429 1430
	return bio_integrity(rq->bio);
}

1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442
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;
}

1443 1444
#else /* CONFIG_BLK_DEV_INTEGRITY */

1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508
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;
}
1509 1510 1511

#endif /* CONFIG_BLK_DEV_INTEGRITY */

1512
struct block_device_operations {
A
Al Viro 已提交
1513
	int (*open) (struct block_device *, fmode_t);
1514
	void (*release) (struct gendisk *, fmode_t);
A
Al Viro 已提交
1515 1516
	int (*ioctl) (struct block_device *, fmode_t, unsigned, unsigned long);
	int (*compat_ioctl) (struct block_device *, fmode_t, unsigned, unsigned long);
1517 1518
	int (*direct_access) (struct block_device *, sector_t,
						void **, unsigned long *);
1519 1520 1521
	unsigned int (*check_events) (struct gendisk *disk,
				      unsigned int clearing);
	/* ->media_changed() is DEPRECATED, use ->check_events() instead */
1522
	int (*media_changed) (struct gendisk *);
1523
	void (*unlock_native_capacity) (struct gendisk *);
1524 1525
	int (*revalidate_disk) (struct gendisk *);
	int (*getgeo)(struct block_device *, struct hd_geometry *);
1526 1527
	/* this callback is with swap_lock and sometimes page table lock held */
	void (*swap_slot_free_notify) (struct block_device *, unsigned long);
1528 1529 1530
	struct module *owner;
};

1531 1532
extern int __blkdev_driver_ioctl(struct block_device *, fmode_t, unsigned int,
				 unsigned long);
1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543
#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;
}

1544 1545 1546 1547
struct blk_plug {
};

static inline void blk_start_plug(struct blk_plug *plug)
1548 1549 1550
{
}

1551
static inline void blk_finish_plug(struct blk_plug *plug)
1552 1553 1554
{
}

1555
static inline void blk_flush_plug(struct task_struct *task)
1556 1557 1558
{
}

1559 1560 1561 1562 1563
static inline void blk_schedule_flush_plug(struct task_struct *task)
{
}


1564 1565 1566 1567 1568
static inline bool blk_needs_flush_plug(struct task_struct *tsk)
{
	return false;
}

1569 1570
#endif /* CONFIG_BLOCK */

L
Linus Torvalds 已提交
1571
#endif