blkdev.h 44.7 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 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

J
Jens Axboe 已提交
102
	u64 cmd_flags;
103
	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;

T
Tejun Heo 已提交
163 164
	void *special;		/* opaque pointer available for LLD use */
	char *buffer;		/* kaddr of the current segment if available */
L
Linus Torvalds 已提交
165

166 167 168
	int tag;
	int errors;

L
Linus Torvalds 已提交
169 170 171
	/*
	 * when request is used as a packet command carrier
	 */
172 173
	unsigned char __cmd[BLK_MAX_CDB];
	unsigned char *cmd;
174
	unsigned short cmd_len;
L
Linus Torvalds 已提交
175

176
	unsigned int extra_len;	/* length of alignment and padding */
L
Linus Torvalds 已提交
177
	unsigned int sense_len;
T
Tejun Heo 已提交
178
	unsigned int resid_len;	/* residual count */
L
Linus Torvalds 已提交
179 180
	void *sense;

J
Jens Axboe 已提交
181 182
	unsigned long deadline;
	struct list_head timeout_list;
L
Linus Torvalds 已提交
183
	unsigned int timeout;
184
	int retries;
L
Linus Torvalds 已提交
185 186

	/*
187
	 * completion callback.
L
Linus Torvalds 已提交
188 189 190
	 */
	rq_end_io_fn *end_io;
	void *end_io_data;
191 192 193

	/* for bidi */
	struct request *next_rq;
L
Linus Torvalds 已提交
194 195
};

196 197 198 199 200
static inline unsigned short req_get_ioprio(struct request *req)
{
	return req->ioprio;
}

L
Linus Torvalds 已提交
201
/*
202
 * State information carried for REQ_TYPE_PM_SUSPEND and REQ_TYPE_PM_RESUME
L
Linus Torvalds 已提交
203 204 205 206 207 208 209 210 211 212 213 214 215
 * 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>

216
typedef void (request_fn_proc) (struct request_queue *q);
217
typedef void (make_request_fn) (struct request_queue *q, struct bio *bio);
218
typedef int (prep_rq_fn) (struct request_queue *, struct request *);
219
typedef void (unprep_rq_fn) (struct request_queue *, struct request *);
L
Linus Torvalds 已提交
220 221

struct bio_vec;
222 223 224 225 226 227 228 229
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 *);
230
typedef void (softirq_done_fn)(struct request *);
231
typedef int (dma_drain_needed_fn)(struct request *);
232
typedef int (lld_busy_fn) (struct request_queue *q);
M
Mike Christie 已提交
233
typedef int (bsg_job_fn) (struct bsg_job *);
L
Linus Torvalds 已提交
234

J
Jens Axboe 已提交
235 236 237 238 239 240 241 242
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
243 244 245 246 247 248 249 250 251 252
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 */
253
	int real_max_depth;		/* what the array can hold */
L
Linus Torvalds 已提交
254 255 256
	atomic_t refcnt;		/* map can be shared */
};

257 258 259
#define BLK_SCSI_MAX_CMDS	(256)
#define BLK_SCSI_CMD_PER_LONG	(BLK_SCSI_MAX_CMDS / (sizeof(long) * 8))

260 261 262 263 264 265 266
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;
267 268 269 270
	unsigned int		physical_block_size;
	unsigned int		alignment_offset;
	unsigned int		io_min;
	unsigned int		io_opt;
271
	unsigned int		max_discard_sectors;
272
	unsigned int		max_write_same_sectors;
273 274
	unsigned int		discard_granularity;
	unsigned int		discard_alignment;
275 276

	unsigned short		logical_block_size;
277
	unsigned short		max_segments;
278
	unsigned short		max_integrity_segments;
279

280
	unsigned char		misaligned;
281
	unsigned char		discard_misaligned;
282
	unsigned char		cluster;
283
	unsigned char		discard_zeroes_data;
284 285
};

286
struct request_queue {
L
Linus Torvalds 已提交
287 288 289 290 291
	/*
	 * Together with queue_head for cacheline sharing
	 */
	struct list_head	queue_head;
	struct request		*last_merge;
J
Jens Axboe 已提交
292
	struct elevator_queue	*elevator;
293 294
	int			nr_rqs[2];	/* # allocated [a]sync rqs */
	int			nr_rqs_elvpriv;	/* # allocated rqs w/ elvpriv */
L
Linus Torvalds 已提交
295 296

	/*
297 298 299 300
	 * 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
301
	 */
302
	struct request_list	root_rl;
L
Linus Torvalds 已提交
303 304 305 306

	request_fn_proc		*request_fn;
	make_request_fn		*make_request_fn;
	prep_rq_fn		*prep_rq_fn;
307
	unprep_rq_fn		*unprep_rq_fn;
L
Linus Torvalds 已提交
308
	merge_bvec_fn		*merge_bvec_fn;
309
	softirq_done_fn		*softirq_done_fn;
J
Jens Axboe 已提交
310
	rq_timed_out_fn		*rq_timed_out_fn;
311
	dma_drain_needed_fn	*dma_drain_needed;
312
	lld_busy_fn		*lld_busy_fn;
L
Linus Torvalds 已提交
313

314 315 316
	/*
	 * Dispatch queue sorting
	 */
J
Jens Axboe 已提交
317
	sector_t		end_sector;
318 319
	struct request		*boundary_rq;

L
Linus Torvalds 已提交
320
	/*
321
	 * Delayed queue handling
L
Linus Torvalds 已提交
322
	 */
323
	struct delayed_work	delay_work;
L
Linus Torvalds 已提交
324 325 326 327 328 329 330 331 332 333

	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;

	/*
334
	 * various queue flags, see QUEUE_* below
L
Linus Torvalds 已提交
335
	 */
336
	unsigned long		queue_flags;
L
Linus Torvalds 已提交
337

338 339 340 341 342 343
	/*
	 * ida allocated id for this queue.  Used to index queues from
	 * ioctx.
	 */
	int			id;

L
Linus Torvalds 已提交
344
	/*
345
	 * queue needs bounce pages for pages above this limit
L
Linus Torvalds 已提交
346
	 */
347
	gfp_t			bounce_gfp;
L
Linus Torvalds 已提交
348 349

	/*
已提交
350 351 352
	 * 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
353
	 */
已提交
354
	spinlock_t		__queue_lock;
L
Linus Torvalds 已提交
355 356 357 358 359 360 361
	spinlock_t		*queue_lock;

	/*
	 * queue kobject
	 */
	struct kobject kobj;

L
Lin Ming 已提交
362 363 364 365 366 367
#ifdef CONFIG_PM_RUNTIME
	struct device		*dev;
	int			rpm_status;
	unsigned int		nr_pending;
#endif

L
Linus Torvalds 已提交
368 369 370 371 372 373 374 375
	/*
	 * 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
376
	unsigned int		dma_drain_size;
377
	void			*dma_drain_buffer;
378
	unsigned int		dma_pad_mask;
L
Linus Torvalds 已提交
379 380 381
	unsigned int		dma_alignment;

	struct blk_queue_tag	*queue_tags;
382
	struct list_head	tag_busy_list;
L
Linus Torvalds 已提交
383

384
	unsigned int		nr_sorted;
385
	unsigned int		in_flight[2];
386 387 388 389 390 391
	/*
	 * 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
392

J
Jens Axboe 已提交
393 394 395 396
	unsigned int		rq_timeout;
	struct timer_list	timeout;
	struct list_head	timeout_list;

397
	struct list_head	icq_list;
398
#ifdef CONFIG_BLK_CGROUP
399
	DECLARE_BITMAP		(blkcg_pols, BLKCG_MAX_POLS);
T
Tejun Heo 已提交
400
	struct blkcg_gq		*root_blkg;
401
	struct list_head	blkg_list;
402
#endif
403

404 405
	struct queue_limits	limits;

L
Linus Torvalds 已提交
406 407 408 409 410
	/*
	 * sg stuff
	 */
	unsigned int		sg_timeout;
	unsigned int		sg_reserved_size;
411
	int			node;
412
#ifdef CONFIG_BLK_DEV_IO_TRACE
413
	struct blk_trace	*blk_trace;
414
#endif
L
Linus Torvalds 已提交
415
	/*
416
	 * for flush operations
L
Linus Torvalds 已提交
417
	 */
418
	unsigned int		flush_flags;
419
	unsigned int		flush_not_queueable:1;
420
	unsigned int		flush_queue_delayed:1;
421 422 423 424 425
	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;
426
	struct request		flush_rq;
427 428

	struct mutex		sysfs_lock;
429

430 431
	int			bypass_depth;

432
#if defined(CONFIG_BLK_DEV_BSG)
M
Mike Christie 已提交
433 434
	bsg_job_fn		*bsg_job_fn;
	int			bsg_job_size;
435 436
	struct bsg_class_device bsg_dev;
#endif
437

438 439 440
#ifdef CONFIG_BLK_CGROUP
	struct list_head	all_q_node;
#endif
441 442 443 444
#ifdef CONFIG_BLK_DEV_THROTTLING
	/* Throttle data */
	struct throtl_data *td;
#endif
T
Tejun Heo 已提交
445
	struct rcu_head		rcu_head;
L
Linus Torvalds 已提交
446 447 448 449
};

#define QUEUE_FLAG_QUEUED	1	/* uses generic tag queueing */
#define QUEUE_FLAG_STOPPED	2	/* queue is stopped */
450 451
#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
452
#define QUEUE_FLAG_DYING	5	/* queue being torn down */
453
#define QUEUE_FLAG_BYPASS	6	/* act as dumb FIFO queue */
454 455
#define QUEUE_FLAG_BIDI		7	/* queue supports bidi requests */
#define QUEUE_FLAG_NOMERGES     8	/* disable merge attempts */
D
Dan Williams 已提交
456
#define QUEUE_FLAG_SAME_COMP	9	/* complete on same CPU-group */
457 458 459
#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) */
460
#define QUEUE_FLAG_VIRT        QUEUE_FLAG_NONROT /* paravirt device */
461 462 463 464 465
#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
466
#define QUEUE_FLAG_SAME_FORCE  18	/* force complete on same CPU */
467
#define QUEUE_FLAG_DEAD        19	/* queue tear-down finished */
468 469

#define QUEUE_FLAG_DEFAULT	((1 << QUEUE_FLAG_IO_STAT) |		\
470
				 (1 << QUEUE_FLAG_STACKABLE)	|	\
471 472
				 (1 << QUEUE_FLAG_SAME_COMP)	|	\
				 (1 << QUEUE_FLAG_ADD_RANDOM))
473

474
static inline void queue_lockdep_assert_held(struct request_queue *q)
475
{
476 477
	if (q->queue_lock)
		lockdep_assert_held(q->queue_lock);
478 479
}

N
Nick Piggin 已提交
480 481 482 483 484 485
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
486 487 488
static inline int queue_flag_test_and_clear(unsigned int flag,
					    struct request_queue *q)
{
489
	queue_lockdep_assert_held(q);
J
Jens Axboe 已提交
490 491 492 493 494 495 496 497 498 499 500 501

	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)
{
502
	queue_lockdep_assert_held(q);
J
Jens Axboe 已提交
503 504 505 506 507 508 509 510 511

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

	return 1;
}

N
Nick Piggin 已提交
512 513
static inline void queue_flag_set(unsigned int flag, struct request_queue *q)
{
514
	queue_lockdep_assert_held(q);
N
Nick Piggin 已提交
515 516 517 518 519 520 521 522 523
	__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);
}

524 525 526 527 528
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
529 530
static inline void queue_flag_clear(unsigned int flag, struct request_queue *q)
{
531
	queue_lockdep_assert_held(q);
N
Nick Piggin 已提交
532 533 534
	__clear_bit(flag, &q->queue_flags);
}

L
Linus Torvalds 已提交
535 536
#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
537
#define blk_queue_dying(q)	test_bit(QUEUE_FLAG_DYING, &(q)->queue_flags)
538
#define blk_queue_dead(q)	test_bit(QUEUE_FLAG_DEAD, &(q)->queue_flags)
539
#define blk_queue_bypass(q)	test_bit(QUEUE_FLAG_BYPASS, &(q)->queue_flags)
540
#define blk_queue_nomerges(q)	test_bit(QUEUE_FLAG_NOMERGES, &(q)->queue_flags)
541 542
#define blk_queue_noxmerges(q)	\
	test_bit(QUEUE_FLAG_NOXMERGES, &(q)->queue_flags)
543
#define blk_queue_nonrot(q)	test_bit(QUEUE_FLAG_NONROT, &(q)->queue_flags)
544
#define blk_queue_io_stat(q)	test_bit(QUEUE_FLAG_IO_STAT, &(q)->queue_flags)
545
#define blk_queue_add_random(q)	test_bit(QUEUE_FLAG_ADD_RANDOM, &(q)->queue_flags)
546 547
#define blk_queue_stackable(q)	\
	test_bit(QUEUE_FLAG_STACKABLE, &(q)->queue_flags)
548
#define blk_queue_discard(q)	test_bit(QUEUE_FLAG_DISCARD, &(q)->queue_flags)
A
Adrian Hunter 已提交
549 550
#define blk_queue_secdiscard(q)	(blk_queue_discard(q) && \
	test_bit(QUEUE_FLAG_SECDISCARD, &(q)->queue_flags))
L
Linus Torvalds 已提交
551

552 553 554 555 556 557
#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) && \
558
	 ((rq)->cmd_type == REQ_TYPE_FS))
559

L
Linus Torvalds 已提交
560
#define blk_pm_request(rq)	\
561 562
	((rq)->cmd_type == REQ_TYPE_PM_SUSPEND || \
	 (rq)->cmd_type == REQ_TYPE_PM_RESUME)
L
Linus Torvalds 已提交
563

564
#define blk_rq_cpu_valid(rq)	((rq)->cpu != -1)
565
#define blk_bidi_rq(rq)		((rq)->next_rq != NULL)
566 567
/* rq->queuelist of dequeued request must be list_empty() */
#define blk_queued_rq(rq)	(!list_empty(&(rq)->queuelist))
L
Linus Torvalds 已提交
568 569 570

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

J
Jens Axboe 已提交
571
#define rq_data_dir(rq)		(((rq)->cmd_flags & 1) != 0)
L
Linus Torvalds 已提交
572

573 574 575 576 577
static inline unsigned int blk_queue_cluster(struct request_queue *q)
{
	return q->limits.cluster;
}

578
/*
579
 * We regard a request as sync, if either a read or a sync write
580
 */
581 582
static inline bool rw_is_sync(unsigned int rw_flags)
{
583
	return !(rw_flags & REQ_WRITE) || (rw_flags & REQ_SYNC);
584 585 586 587 588 589 590
}

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

591
static inline bool blk_rl_full(struct request_list *rl, bool sync)
L
Linus Torvalds 已提交
592
{
593 594 595
	unsigned int flag = sync ? BLK_RL_SYNCFULL : BLK_RL_ASYNCFULL;

	return rl->flags & flag;
L
Linus Torvalds 已提交
596 597
}

598
static inline void blk_set_rl_full(struct request_list *rl, bool sync)
L
Linus Torvalds 已提交
599
{
600 601 602
	unsigned int flag = sync ? BLK_RL_SYNCFULL : BLK_RL_ASYNCFULL;

	rl->flags |= flag;
L
Linus Torvalds 已提交
603 604
}

605
static inline void blk_clear_rl_full(struct request_list *rl, bool sync)
L
Linus Torvalds 已提交
606
{
607 608 609
	unsigned int flag = sync ? BLK_RL_SYNCFULL : BLK_RL_ASYNCFULL;

	rl->flags &= ~flag;
L
Linus Torvalds 已提交
610 611
}

612 613 614 615
static inline bool rq_mergeable(struct request *rq)
{
	if (rq->cmd_type != REQ_TYPE_FS)
		return false;
L
Linus Torvalds 已提交
616

617 618 619 620 621
	if (rq->cmd_flags & REQ_NOMERGE_FLAGS)
		return false;

	return true;
}
L
Linus Torvalds 已提交
622

623 624 625 626 627 628 629 630 631
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;

632 633 634
	if ((flags1 & REQ_WRITE_SAME) != (flags2 & REQ_WRITE_SAME))
		return false;

635 636 637
	return true;
}

638 639 640 641 642 643 644 645
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
646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661
/*
 * 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
 */
662 663

#if BITS_PER_LONG == 32
L
Linus Torvalds 已提交
664
#define BLK_BOUNCE_HIGH		((u64)blk_max_low_pfn << PAGE_SHIFT)
665 666 667 668
#else
#define BLK_BOUNCE_HIGH		-1ULL
#endif
#define BLK_BOUNCE_ANY		(-1ULL)
669
#define BLK_BOUNCE_ISA		(DMA_BIT_MASK(24))
L
Linus Torvalds 已提交
670

671 672 673 674
/*
 * default timeout for SG_IO if none specified
 */
#define BLK_DEFAULT_SG_TIMEOUT	(60 * HZ)
L
Linus Torvalds 已提交
675
#define BLK_MIN_SG_TIMEOUT	(7 * HZ)
676

677
#ifdef CONFIG_BOUNCE
L
Linus Torvalds 已提交
678
extern int init_emergency_isa_pool(void);
679
extern void blk_queue_bounce(struct request_queue *q, struct bio **bio);
L
Linus Torvalds 已提交
680 681 682 683 684
#else
static inline int init_emergency_isa_pool(void)
{
	return 0;
}
685
static inline void blk_queue_bounce(struct request_queue *q, struct bio **bio)
L
Linus Torvalds 已提交
686 687 688 689
{
}
#endif /* CONFIG_MMU */

690 691 692 693
struct rq_map_data {
	struct page **pages;
	int page_order;
	int nr_entries;
694
	unsigned long offset;
695
	int null_mapped;
696
	int from_user;
697 698
};

699 700 701 702 703 704
struct req_iterator {
	int i;
	struct bio *bio;
};

/* This should not be used directly - use rq_for_each_segment */
705 706
#define for_each_bio(_bio)		\
	for (; _bio; _bio = _bio->bi_next)
707
#define __rq_for_each_bio(_bio, rq)	\
L
Linus Torvalds 已提交
708 709 710
	if ((rq->bio))			\
		for (_bio = (rq)->bio; _bio; _bio = _bio->bi_next)

711 712 713 714 715 716 717
#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)

718 719 720 721 722 723 724 725 726 727 728
#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
729 730 731
extern int blk_register_queue(struct gendisk *disk);
extern void blk_unregister_queue(struct gendisk *disk);
extern void generic_make_request(struct bio *bio);
732
extern void blk_rq_init(struct request_queue *q, struct request *rq);
L
Linus Torvalds 已提交
733
extern void blk_put_request(struct request *);
734 735
extern void __blk_put_request(struct request_queue *, struct request *);
extern struct request *blk_get_request(struct request_queue *, int, gfp_t);
736 737
extern struct request *blk_make_request(struct request_queue *, struct bio *,
					gfp_t);
738
extern void blk_requeue_request(struct request_queue *, struct request *);
739 740
extern void blk_add_request_payload(struct request *rq, struct page *page,
		unsigned int len);
741
extern int blk_rq_check_limits(struct request_queue *q, struct request *rq);
742
extern int blk_lld_busy(struct request_queue *q);
743 744 745 746 747
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);
748 749
extern int blk_insert_cloned_request(struct request_queue *q,
				     struct request *rq);
750
extern void blk_delay_queue(struct request_queue *, unsigned long);
751
extern void blk_recount_segments(struct request_queue *, struct bio *);
752
extern int scsi_verify_blk_ioctl(struct block_device *, unsigned int);
753 754
extern int scsi_cmd_blk_ioctl(struct block_device *, fmode_t,
			      unsigned int, void __user *);
755 756
extern int scsi_cmd_ioctl(struct request_queue *, struct gendisk *, fmode_t,
			  unsigned int, void __user *);
757 758
extern int sg_scsi_ioctl(struct request_queue *, struct gendisk *, fmode_t,
			 struct scsi_ioctl_command __user *);
759

760
extern void blk_queue_bio(struct request_queue *q, struct bio *bio);
C
Christoph Hellwig 已提交
761

762 763 764 765 766
/*
 * 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.
 */
767
static inline void blk_clear_queue_congested(struct request_queue *q, int sync)
768
{
769
	clear_bdi_congested(&q->backing_dev_info, sync);
770 771 772 773 774 775
}

/*
 * A queue has just entered congestion.  Flag that in the queue's VM-visible
 * state flags and increment the global gounter of congested queues.
 */
776
static inline void blk_set_queue_congested(struct request_queue *q, int sync)
777
{
778
	set_bdi_congested(&q->backing_dev_info, sync);
779 780
}

781 782
extern void blk_start_queue(struct request_queue *q);
extern void blk_stop_queue(struct request_queue *q);
L
Linus Torvalds 已提交
783
extern void blk_sync_queue(struct request_queue *q);
784
extern void __blk_stop_queue(struct request_queue *q);
785
extern void __blk_run_queue(struct request_queue *q);
786
extern void blk_run_queue(struct request_queue *);
787
extern void blk_run_queue_async(struct request_queue *q);
788
extern int blk_rq_map_user(struct request_queue *, struct request *,
789 790
			   struct rq_map_data *, void __user *, unsigned long,
			   gfp_t);
J
Jens Axboe 已提交
791
extern int blk_rq_unmap_user(struct bio *);
792 793
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 *,
794 795
			       struct rq_map_data *, struct sg_iovec *, int,
			       unsigned int, gfp_t);
796
extern int blk_execute_rq(struct request_queue *, struct gendisk *,
797
			  struct request *, int);
798
extern void blk_execute_rq_nowait(struct request_queue *, struct gendisk *,
799
				  struct request *, int, rq_end_io_fn *);
800

801
static inline struct request_queue *bdev_get_queue(struct block_device *bdev)
L
Linus Torvalds 已提交
802 803 804 805
{
	return bdev->bd_disk->queue;
}

806
/*
807 808 809 810 811 812
 * 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
813
 */
814 815
static inline sector_t blk_rq_pos(const struct request *rq)
{
816
	return rq->__sector;
817 818 819 820
}

static inline unsigned int blk_rq_bytes(const struct request *rq)
{
821
	return rq->__data_len;
822 823
}

824 825 826 827
static inline int blk_rq_cur_bytes(const struct request *rq)
{
	return rq->bio ? bio_cur_bytes(rq->bio) : 0;
}
828

829 830
extern unsigned int blk_rq_err_bytes(const struct request *rq);

831 832
static inline unsigned int blk_rq_sectors(const struct request *rq)
{
833
	return blk_rq_bytes(rq) >> 9;
834 835 836 837
}

static inline unsigned int blk_rq_cur_sectors(const struct request *rq)
{
838
	return blk_rq_cur_bytes(rq) >> 9;
839 840
}

841 842 843 844
static inline unsigned int blk_queue_get_max_sectors(struct request_queue *q,
						     unsigned int cmd_flags)
{
	if (unlikely(cmd_flags & REQ_DISCARD))
845
		return min(q->limits.max_discard_sectors, UINT_MAX >> 9);
846

847 848 849
	if (unlikely(cmd_flags & REQ_WRITE_SAME))
		return q->limits.max_write_same_sectors;

850 851 852 853 854 855 856 857 858 859 860 861 862
	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);
}

863 864 865 866 867 868 869 870 871 872 873
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;
}

874 875 876 877 878 879 880
/*
 * 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
881
/*
882 883 884 885 886
 * Request completion related functions.
 *
 * blk_update_request() completes given number of bytes and updates
 * the request without completing it.
 *
887 888
 * blk_end_request() and friends.  __blk_end_request() must be called
 * with the request queue spinlock acquired.
L
Linus Torvalds 已提交
889 890
 *
 * Several drivers define their own end_request and call
891 892
 * blk_end_request() for parts of the original function.
 * This prevents code duplication in drivers.
L
Linus Torvalds 已提交
893
 */
894 895
extern bool blk_update_request(struct request *rq, int error,
			       unsigned int nr_bytes);
896 897 898 899
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);
900
extern bool blk_end_request_err(struct request *rq, int error);
901 902 903 904
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);
905
extern bool __blk_end_request_err(struct request *rq, int error);
906

907
extern void blk_complete_request(struct request *);
J
Jens Axboe 已提交
908 909
extern void __blk_complete_request(struct request *);
extern void blk_abort_request(struct request *);
910
extern void blk_unprep_request(struct request *);
911

L
Linus Torvalds 已提交
912 913 914
/*
 * Access functions for manipulating queue properties
 */
915
extern struct request_queue *blk_init_queue_node(request_fn_proc *rfn,
916
					spinlock_t *lock, int node_id);
917
extern struct request_queue *blk_init_queue(request_fn_proc *, spinlock_t *);
918 919
extern struct request_queue *blk_init_allocated_queue(struct request_queue *,
						      request_fn_proc *, spinlock_t *);
920 921 922
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);
923
extern void blk_limits_max_hw_sectors(struct queue_limits *, unsigned int);
924
extern void blk_queue_max_hw_sectors(struct request_queue *, unsigned int);
925
extern void blk_queue_max_segments(struct request_queue *, unsigned short);
926
extern void blk_queue_max_segment_size(struct request_queue *, unsigned int);
927 928
extern void blk_queue_max_discard_sectors(struct request_queue *q,
		unsigned int max_discard_sectors);
929 930
extern void blk_queue_max_write_same_sectors(struct request_queue *q,
		unsigned int max_write_same_sectors);
931
extern void blk_queue_logical_block_size(struct request_queue *, unsigned short);
932
extern void blk_queue_physical_block_size(struct request_queue *, unsigned int);
933 934
extern void blk_queue_alignment_offset(struct request_queue *q,
				       unsigned int alignment);
935
extern void blk_limits_io_min(struct queue_limits *limits, unsigned int min);
936
extern void blk_queue_io_min(struct request_queue *q, unsigned int min);
937
extern void blk_limits_io_opt(struct queue_limits *limits, unsigned int opt);
938
extern void blk_queue_io_opt(struct request_queue *q, unsigned int opt);
939
extern void blk_set_default_limits(struct queue_limits *lim);
940
extern void blk_set_stacking_limits(struct queue_limits *lim);
941 942
extern int blk_stack_limits(struct queue_limits *t, struct queue_limits *b,
			    sector_t offset);
943 944
extern int bdev_stack_limits(struct queue_limits *t, struct block_device *bdev,
			    sector_t offset);
945 946
extern void disk_stack_limits(struct gendisk *disk, struct block_device *bdev,
			      sector_t offset);
947
extern void blk_queue_stack_limits(struct request_queue *t, struct request_queue *b);
948
extern void blk_queue_dma_pad(struct request_queue *, unsigned int);
949
extern void blk_queue_update_dma_pad(struct request_queue *, unsigned int);
950 951 952
extern int blk_queue_dma_drain(struct request_queue *q,
			       dma_drain_needed_fn *dma_drain_needed,
			       void *buf, unsigned int size);
953
extern void blk_queue_lld_busy(struct request_queue *q, lld_busy_fn *fn);
954 955
extern void blk_queue_segment_boundary(struct request_queue *, unsigned long);
extern void blk_queue_prep_rq(struct request_queue *, prep_rq_fn *pfn);
956
extern void blk_queue_unprep_rq(struct request_queue *, unprep_rq_fn *ufn);
957 958
extern void blk_queue_merge_bvec(struct request_queue *, merge_bvec_fn *);
extern void blk_queue_dma_alignment(struct request_queue *, int);
959
extern void blk_queue_update_dma_alignment(struct request_queue *, int);
960
extern void blk_queue_softirq_done(struct request_queue *, softirq_done_fn *);
J
Jens Axboe 已提交
961 962
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);
963
extern void blk_queue_flush(struct request_queue *q, unsigned int flush);
964
extern void blk_queue_flush_queueable(struct request_queue *q, bool queueable);
L
Linus Torvalds 已提交
965 966
extern struct backing_dev_info *blk_get_backing_dev_info(struct block_device *bdev);

967
extern int blk_rq_map_sg(struct request_queue *, struct request *, struct scatterlist *);
A
Asias He 已提交
968 969
extern int blk_bio_map_sg(struct request_queue *q, struct bio *bio,
			  struct scatterlist *sglist);
L
Linus Torvalds 已提交
970 971 972
extern void blk_dump_rq_flags(struct request *, char *);
extern long nr_blockdev_pages(void);

T
Tejun Heo 已提交
973
bool __must_check blk_get_queue(struct request_queue *);
974 975 976
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
977

L
Lin Ming 已提交
978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998
/*
 * 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

999
/*
S
Suresh Jayaraman 已提交
1000 1001 1002 1003 1004 1005 1006 1007 1008 1009
 * 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.
1010
 */
1011
struct blk_plug {
S
Suresh Jayaraman 已提交
1012 1013 1014
	unsigned long magic; /* detect uninitialized use-cases */
	struct list_head list; /* requests */
	struct list_head cb_list; /* md requires an unplug callback */
1015
};
1016 1017
#define BLK_MAX_REQUEST_COUNT 16

1018
struct blk_plug_cb;
1019
typedef void (*blk_plug_cb_fn)(struct blk_plug_cb *, bool);
1020 1021
struct blk_plug_cb {
	struct list_head list;
1022 1023
	blk_plug_cb_fn callback;
	void *data;
1024
};
1025 1026
extern struct blk_plug_cb *blk_check_plugged(blk_plug_cb_fn unplug,
					     void *data, int size);
1027 1028
extern void blk_start_plug(struct blk_plug *);
extern void blk_finish_plug(struct blk_plug *);
1029
extern void blk_flush_plug_list(struct blk_plug *, bool);
1030 1031 1032 1033 1034

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

1035 1036 1037 1038 1039 1040 1041 1042
	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;

1043
	if (plug)
1044
		blk_flush_plug_list(plug, true);
1045 1046 1047 1048 1049 1050
}

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

1051
	return plug && (!list_empty(&plug->list) || !list_empty(&plug->cb_list));
1052 1053
}

L
Linus Torvalds 已提交
1054 1055 1056
/*
 * tag stuff
 */
1057
#define blk_rq_tagged(rq)		((rq)->cmd_flags & REQ_QUEUED)
1058 1059 1060 1061 1062 1063 1064
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 *);
1065 1066
extern struct blk_queue_tag *blk_init_tags(int);
extern void blk_free_tags(struct blk_queue_tag *);
L
Linus Torvalds 已提交
1067

1068 1069 1070 1071 1072 1073 1074
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];
}
1075 1076 1077 1078

#define BLKDEV_DISCARD_SECURE  0x01    /* secure discard */

extern int blkdev_issue_flush(struct block_device *, gfp_t, sector_t *);
1079 1080
extern int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
		sector_t nr_sects, gfp_t gfp_mask, unsigned long flags);
1081 1082
extern int blkdev_issue_write_same(struct block_device *bdev, sector_t sector,
		sector_t nr_sects, gfp_t gfp_mask, struct page *page);
1083
extern int blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
1084
			sector_t nr_sects, gfp_t gfp_mask);
1085 1086
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
1087
{
1088 1089 1090
	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
1091
}
1092
static inline int sb_issue_zeroout(struct super_block *sb, sector_t block,
1093
		sector_t nr_blocks, gfp_t gfp_mask)
1094 1095 1096 1097
{
	return blkdev_issue_zeroout(sb->s_bdev,
				    block << (sb->s_blocksize_bits - 9),
				    nr_blocks << (sb->s_blocksize_bits - 9),
1098
				    gfp_mask);
1099
}
L
Linus Torvalds 已提交
1100

1101
extern int blk_verify_command(unsigned char *cmd, fmode_t has_write_perm);
1102

1103 1104 1105 1106 1107 1108 1109
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,
};
1110

L
Linus Torvalds 已提交
1111 1112
#define blkdev_entry_to_request(entry) list_entry((entry), struct request, queuelist)

1113 1114
static inline unsigned long queue_bounce_pfn(struct request_queue *q)
{
1115
	return q->limits.bounce_pfn;
1116 1117 1118 1119
}

static inline unsigned long queue_segment_boundary(struct request_queue *q)
{
1120
	return q->limits.seg_boundary_mask;
1121 1122 1123 1124
}

static inline unsigned int queue_max_sectors(struct request_queue *q)
{
1125
	return q->limits.max_sectors;
1126 1127 1128 1129
}

static inline unsigned int queue_max_hw_sectors(struct request_queue *q)
{
1130
	return q->limits.max_hw_sectors;
1131 1132
}

1133
static inline unsigned short queue_max_segments(struct request_queue *q)
1134
{
1135
	return q->limits.max_segments;
1136 1137 1138 1139
}

static inline unsigned int queue_max_segment_size(struct request_queue *q)
{
1140
	return q->limits.max_segment_size;
1141 1142
}

1143
static inline unsigned short queue_logical_block_size(struct request_queue *q)
L
Linus Torvalds 已提交
1144 1145 1146
{
	int retval = 512;

1147 1148
	if (q && q->limits.logical_block_size)
		retval = q->limits.logical_block_size;
L
Linus Torvalds 已提交
1149 1150 1151 1152

	return retval;
}

1153
static inline unsigned short bdev_logical_block_size(struct block_device *bdev)
L
Linus Torvalds 已提交
1154
{
1155
	return queue_logical_block_size(bdev_get_queue(bdev));
L
Linus Torvalds 已提交
1156 1157
}

1158 1159 1160 1161 1162
static inline unsigned int queue_physical_block_size(struct request_queue *q)
{
	return q->limits.physical_block_size;
}

1163
static inline unsigned int bdev_physical_block_size(struct block_device *bdev)
M
Martin K. Petersen 已提交
1164 1165 1166 1167
{
	return queue_physical_block_size(bdev_get_queue(bdev));
}

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

M
Martin K. Petersen 已提交
1173 1174 1175 1176 1177
static inline int bdev_io_min(struct block_device *bdev)
{
	return queue_io_min(bdev_get_queue(bdev));
}

1178 1179 1180 1181 1182
static inline unsigned int queue_io_opt(struct request_queue *q)
{
	return q->limits.io_opt;
}

M
Martin K. Petersen 已提交
1183 1184 1185 1186 1187
static inline int bdev_io_opt(struct block_device *bdev)
{
	return queue_io_opt(bdev_get_queue(bdev));
}

1188 1189
static inline int queue_alignment_offset(struct request_queue *q)
{
M
Martin K. Petersen 已提交
1190
	if (q->limits.misaligned)
1191 1192
		return -1;

M
Martin K. Petersen 已提交
1193
	return q->limits.alignment_offset;
1194 1195
}

1196
static inline int queue_limit_alignment_offset(struct queue_limits *lim, sector_t sector)
1197 1198
{
	unsigned int granularity = max(lim->physical_block_size, lim->io_min);
1199
	unsigned int alignment = (sector << 9) & (granularity - 1);
1200

1201 1202
	return (granularity + lim->alignment_offset - alignment)
		& (granularity - 1);
1203 1204
}

M
Martin K. Petersen 已提交
1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217
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;
}

1218 1219 1220 1221 1222 1223 1224 1225
static inline int queue_discard_alignment(struct request_queue *q)
{
	if (q->limits.discard_misaligned)
		return -1;

	return q->limits.discard_alignment;
}

1226
static inline int queue_limit_discard_alignment(struct queue_limits *lim, sector_t sector)
1227
{
1228
	unsigned int alignment, granularity, offset;
1229

1230 1231 1232
	if (!lim->max_discard_sectors)
		return 0;

1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246
	/* 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;
1247 1248
}

1249 1250 1251 1252 1253 1254 1255 1256 1257 1258
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;
}

1259 1260
static inline unsigned int queue_discard_zeroes_data(struct request_queue *q)
{
1261
	if (q->limits.max_discard_sectors && q->limits.discard_zeroes_data == 1)
1262 1263 1264 1265 1266 1267 1268 1269 1270 1271
		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));
}

1272 1273 1274 1275 1276 1277 1278 1279 1280 1281
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;
}

1282
static inline int queue_dma_alignment(struct request_queue *q)
L
Linus Torvalds 已提交
1283
{
1284
	return q ? q->dma_alignment : 511;
L
Linus Torvalds 已提交
1285 1286
}

1287
static inline int blk_rq_aligned(struct request_queue *q, unsigned long addr,
1288 1289 1290
				 unsigned int len)
{
	unsigned int alignment = queue_dma_alignment(q) | q->dma_pad_mask;
1291
	return !(addr & alignment) && !(len & alignment);
1292 1293
}

L
Linus Torvalds 已提交
1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304
/* 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;
}

1305
static inline unsigned int block_size(struct block_device *bdev)
L
Linus Torvalds 已提交
1306 1307 1308 1309
{
	return bdev->bd_block_size;
}

1310 1311 1312 1313 1314
static inline bool queue_flush_queueable(struct request_queue *q)
{
	return !q->flush_not_queueable;
}

L
Linus Torvalds 已提交
1315 1316 1317 1318 1319 1320 1321 1322 1323 1324
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;
1325
int kblockd_schedule_work(struct request_queue *q, struct work_struct *work);
L
Linus Torvalds 已提交
1326

1327
#ifdef CONFIG_BLK_CGROUP
1328 1329 1330 1331 1332
/*
 * 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().
 */
1333 1334
static inline void set_start_time_ns(struct request *req)
{
1335
	preempt_disable();
1336
	req->start_time_ns = sched_clock();
1337
	preempt_enable();
1338 1339 1340 1341
}

static inline void set_io_start_time_ns(struct request *req)
{
1342
	preempt_disable();
1343
	req->io_start_time_ns = sched_clock();
1344
	preempt_enable();
1345
}
1346 1347 1348 1349 1350 1351 1352 1353 1354 1355

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;
}
1356 1357 1358
#else
static inline void set_start_time_ns(struct request *req) {}
static inline void set_io_start_time_ns(struct request *req) {}
1359 1360 1361 1362 1363 1364 1365 1366
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;
}
1367 1368
#endif

L
Linus Torvalds 已提交
1369 1370 1371 1372 1373
#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) "-*")

1374 1375
#if defined(CONFIG_BLK_DEV_INTEGRITY)

1376 1377
#define INTEGRITY_FLAG_READ	2	/* verify data integrity on read */
#define INTEGRITY_FLAG_WRITE	4	/* generate data integrity on write */
1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408

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;
};

1409
extern bool blk_integrity_is_initialized(struct gendisk *);
1410 1411
extern int blk_integrity_register(struct gendisk *, struct blk_integrity *);
extern void blk_integrity_unregister(struct gendisk *);
1412
extern int blk_integrity_compare(struct gendisk *, struct gendisk *);
1413 1414 1415 1416 1417 1418 1419
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 *);
1420

1421 1422 1423 1424 1425 1426
static inline
struct blk_integrity *bdev_get_integrity(struct block_device *bdev)
{
	return bdev->bd_disk->integrity;
}

1427 1428 1429 1430 1431
static inline struct blk_integrity *blk_get_integrity(struct gendisk *disk)
{
	return disk->integrity;
}

1432 1433
static inline int blk_integrity_rq(struct request *rq)
{
1434 1435 1436
	if (rq->bio == NULL)
		return 0;

1437 1438 1439
	return bio_integrity(rq->bio);
}

1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451
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;
}

1452 1453
#else /* CONFIG_BLK_DEV_INTEGRITY */

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 1509 1510 1511 1512 1513 1514 1515 1516 1517
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;
}
1518 1519 1520

#endif /* CONFIG_BLK_DEV_INTEGRITY */

1521
struct block_device_operations {
A
Al Viro 已提交
1522
	int (*open) (struct block_device *, fmode_t);
1523
	void (*release) (struct gendisk *, fmode_t);
A
Al Viro 已提交
1524 1525
	int (*ioctl) (struct block_device *, fmode_t, unsigned, unsigned long);
	int (*compat_ioctl) (struct block_device *, fmode_t, unsigned, unsigned long);
1526 1527
	int (*direct_access) (struct block_device *, sector_t,
						void **, unsigned long *);
1528 1529 1530
	unsigned int (*check_events) (struct gendisk *disk,
				      unsigned int clearing);
	/* ->media_changed() is DEPRECATED, use ->check_events() instead */
1531
	int (*media_changed) (struct gendisk *);
1532
	void (*unlock_native_capacity) (struct gendisk *);
1533 1534
	int (*revalidate_disk) (struct gendisk *);
	int (*getgeo)(struct block_device *, struct hd_geometry *);
1535 1536
	/* this callback is with swap_lock and sometimes page table lock held */
	void (*swap_slot_free_notify) (struct block_device *, unsigned long);
1537 1538 1539
	struct module *owner;
};

1540 1541
extern int __blkdev_driver_ioctl(struct block_device *, fmode_t, unsigned int,
				 unsigned long);
1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552
#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;
}

1553 1554 1555 1556
struct blk_plug {
};

static inline void blk_start_plug(struct blk_plug *plug)
1557 1558 1559
{
}

1560
static inline void blk_finish_plug(struct blk_plug *plug)
1561 1562 1563
{
}

1564
static inline void blk_flush_plug(struct task_struct *task)
1565 1566 1567
{
}

1568 1569 1570 1571 1572
static inline void blk_schedule_flush_plug(struct task_struct *task)
{
}


1573 1574 1575 1576 1577
static inline bool blk_needs_flush_plug(struct task_struct *tsk)
{
	return false;
}

1578 1579
#endif /* CONFIG_BLOCK */

L
Linus Torvalds 已提交
1580
#endif