blkdev.h 41.1 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3
#ifndef _LINUX_BLKDEV_H
#define _LINUX_BLKDEV_H

J
Jens Axboe 已提交
4 5
#ifdef CONFIG_BLOCK

6
#include <linux/sched.h>
L
Linus Torvalds 已提交
7 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/module.h>
#include <linux/stringify.h>
19
#include <linux/gfp.h>
20
#include <linux/bsg.h>
21
#include <linux/smp.h>
L
Linus Torvalds 已提交
22 23 24

#include <asm/scatterlist.h>

25 26
struct scsi_ioctl_command;

L
Linus Torvalds 已提交
27 28 29
struct request_queue;
struct elevator_queue;
struct request_pm_state;
30
struct blk_trace;
31 32
struct request;
struct sg_io_hdr;
L
Linus Torvalds 已提交
33 34 35 36 37

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

struct request;
38
typedef void (rq_end_io_fn)(struct request *, int);
L
Linus Torvalds 已提交
39 40

struct request_list {
41 42 43 44
	/*
	 * count[], starved[], and wait[] are indexed by
	 * BLK_RW_SYNC/BLK_RW_ASYNC
	 */
L
Linus Torvalds 已提交
45 46
	int count[2];
	int starved[2];
T
Tejun Heo 已提交
47
	int elvpriv;
L
Linus Torvalds 已提交
48 49 50 51
	mempool_t *rq_pool;
	wait_queue_head_t wait[2];
};

52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69
/*
 * 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 */
	REQ_TYPE_LINUX_BLOCK,		/* generic block layer message */
	/*
	 * 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,
70
	REQ_TYPE_ATA_PC,
71 72 73 74 75 76 77 78 79 80 81 82 83
};

/*
 * For request of type REQ_TYPE_LINUX_BLOCK, rq->cmd[0] is the opcode being
 * sent down (similar to how REQ_TYPE_BLOCK_PC means that ->cmd[] holds a
 * SCSI cdb.
 *
 * 0x00 -> 0x3f are driver private, to be used for whatever purpose they need,
 * typically to differentiate REQ_TYPE_SPECIAL requests.
 *
 */
enum {
	REQ_LB_OP_EJECT	= 0x40,		/* eject request */
D
David Woodhouse 已提交
84
	REQ_LB_OP_FLUSH = 0x41,		/* flush request */
85 86 87
};

/*
88
 * request type modified bits. first four bits match BIO_RW* bits, important
89 90 91
 */
enum rq_flag_bits {
	__REQ_RW,		/* not set, read. set, write */
92 93 94
	__REQ_FAILFAST_DEV,	/* no driver retries of device errors */
	__REQ_FAILFAST_TRANSPORT, /* no driver retries of transport errors */
	__REQ_FAILFAST_DRIVER,	/* no driver retries of driver errors */
95
	/* above flags must match BIO_RW_* */
D
David Woodhouse 已提交
96
	__REQ_DISCARD,		/* request to discard sectors */
97 98 99 100 101 102 103 104 105 106 107 108 109
	__REQ_SORTED,		/* elevator knows about this request */
	__REQ_SOFTBARRIER,	/* may not be passed by ioscheduler */
	__REQ_HARDBARRIER,	/* may not be passed by drive either */
	__REQ_FUA,		/* forced unit access */
	__REQ_NOMERGE,		/* don't touch this for merging */
	__REQ_STARTED,		/* drive already may have started this one */
	__REQ_DONTPREP,		/* don't call prep for this one */
	__REQ_QUEUED,		/* uses queueing */
	__REQ_ELVPRIV,		/* elevator private data attached */
	__REQ_FAILED,		/* set if the request failed */
	__REQ_QUIET,		/* don't worry about errors */
	__REQ_PREEMPT,		/* set for "ide_preempt" requests */
	__REQ_ORDERED_COLOR,	/* is before or after barrier */
110
	__REQ_RW_SYNC,		/* request is sync (sync write or read) */
111
	__REQ_ALLOCED,		/* request came from our alloc pool */
112
	__REQ_RW_META,		/* metadata io request */
113
	__REQ_COPY_USER,	/* contains copies of user pages */
114
	__REQ_INTEGRITY,	/* integrity metadata has been remapped */
115
	__REQ_NOIDLE,		/* Don't anticipate more IO after this one */
116
	__REQ_IO_STAT,		/* account I/O stat */
117
	__REQ_MIXED_MERGE,	/* merge of different types, fail separately */
118 119 120 121
	__REQ_NR_BITS,		/* stops here */
};

#define REQ_RW		(1 << __REQ_RW)
122 123 124
#define REQ_FAILFAST_DEV	(1 << __REQ_FAILFAST_DEV)
#define REQ_FAILFAST_TRANSPORT	(1 << __REQ_FAILFAST_TRANSPORT)
#define REQ_FAILFAST_DRIVER	(1 << __REQ_FAILFAST_DRIVER)
D
David Woodhouse 已提交
125
#define REQ_DISCARD	(1 << __REQ_DISCARD)
126 127 128 129 130 131 132 133 134 135 136 137 138 139
#define REQ_SORTED	(1 << __REQ_SORTED)
#define REQ_SOFTBARRIER	(1 << __REQ_SOFTBARRIER)
#define REQ_HARDBARRIER	(1 << __REQ_HARDBARRIER)
#define REQ_FUA		(1 << __REQ_FUA)
#define REQ_NOMERGE	(1 << __REQ_NOMERGE)
#define REQ_STARTED	(1 << __REQ_STARTED)
#define REQ_DONTPREP	(1 << __REQ_DONTPREP)
#define REQ_QUEUED	(1 << __REQ_QUEUED)
#define REQ_ELVPRIV	(1 << __REQ_ELVPRIV)
#define REQ_FAILED	(1 << __REQ_FAILED)
#define REQ_QUIET	(1 << __REQ_QUIET)
#define REQ_PREEMPT	(1 << __REQ_PREEMPT)
#define REQ_ORDERED_COLOR	(1 << __REQ_ORDERED_COLOR)
#define REQ_RW_SYNC	(1 << __REQ_RW_SYNC)
140
#define REQ_ALLOCED	(1 << __REQ_ALLOCED)
141
#define REQ_RW_META	(1 << __REQ_RW_META)
142
#define REQ_COPY_USER	(1 << __REQ_COPY_USER)
143
#define REQ_INTEGRITY	(1 << __REQ_INTEGRITY)
144
#define REQ_NOIDLE	(1 << __REQ_NOIDLE)
145
#define REQ_IO_STAT	(1 << __REQ_IO_STAT)
146
#define REQ_MIXED_MERGE	(1 << __REQ_MIXED_MERGE)
147

148 149 150
#define REQ_FAILFAST_MASK	(REQ_FAILFAST_DEV | REQ_FAILFAST_TRANSPORT | \
				 REQ_FAILFAST_DRIVER)

L
Linus Torvalds 已提交
151 152 153
#define BLK_MAX_CDB	16

/*
J
Jens Axboe 已提交
154 155 156
 * try to put the fields that are referenced together in the same cacheline.
 * if you modify this structure, be sure to check block/blk-core.c:rq_init()
 * as well!
L
Linus Torvalds 已提交
157 158
 */
struct request {
159
	struct list_head queuelist;
160
	struct call_single_data csd;
161

162
	struct request_queue *q;
163

164 165
	unsigned int cmd_flags;
	enum rq_cmd_type_bits cmd_type;
J
Jens Axboe 已提交
166
	unsigned long atomic_flags;
L
Linus Torvalds 已提交
167

168 169
	int cpu;

170 171
	/* the following two fields are internal, NEVER access directly */
	unsigned int __data_len;	/* total data len */
172
	sector_t __sector;		/* sector cursor */
L
Linus Torvalds 已提交
173 174 175 176

	struct bio *bio;
	struct bio *biotail;

177
	struct hlist_node hash;	/* merge hash */
178 179 180 181 182 183 184 185 186
	/*
	 * The rb_node is only used inside the io scheduler, requests
	 * are pruned when moved to the dispatch queue. So let the
	 * completion_data share space with the rb_node.
	 */
	union {
		struct rb_node rb_node;	/* sort/lookup */
		void *completion_data;
	};
187

188
	/*
189
	 * Three pointers are available for the IO schedulers, if they need
190 191
	 * more they have to dynamically allocate it.
	 */
L
Linus Torvalds 已提交
192
	void *elevator_private;
193
	void *elevator_private2;
194
	void *elevator_private3;
195

196
	struct gendisk *rq_disk;
L
Linus Torvalds 已提交
197
	unsigned long start_time;
198 199 200 201
#ifdef CONFIG_BLK_CGROUP
	unsigned long long start_time_ns;
	unsigned long long io_start_time_ns;    /* when passed to hardware */
#endif
L
Linus Torvalds 已提交
202 203 204 205 206
	/* Number of scatter-gather DMA addr+len pairs after
	 * physical address coalescing is performed.
	 */
	unsigned short nr_phys_segments;

207 208
	unsigned short ioprio;

209 210
	int ref_count;

T
Tejun Heo 已提交
211 212
	void *special;		/* opaque pointer available for LLD use */
	char *buffer;		/* kaddr of the current segment if available */
L
Linus Torvalds 已提交
213

214 215 216
	int tag;
	int errors;

L
Linus Torvalds 已提交
217 218 219
	/*
	 * when request is used as a packet command carrier
	 */
220 221
	unsigned char __cmd[BLK_MAX_CDB];
	unsigned char *cmd;
222
	unsigned short cmd_len;
L
Linus Torvalds 已提交
223

224
	unsigned int extra_len;	/* length of alignment and padding */
L
Linus Torvalds 已提交
225
	unsigned int sense_len;
T
Tejun Heo 已提交
226
	unsigned int resid_len;	/* residual count */
L
Linus Torvalds 已提交
227 228
	void *sense;

J
Jens Axboe 已提交
229 230
	unsigned long deadline;
	struct list_head timeout_list;
L
Linus Torvalds 已提交
231
	unsigned int timeout;
232
	int retries;
L
Linus Torvalds 已提交
233 234

	/*
235
	 * completion callback.
L
Linus Torvalds 已提交
236 237 238
	 */
	rq_end_io_fn *end_io;
	void *end_io_data;
239 240 241

	/* for bidi */
	struct request *next_rq;
L
Linus Torvalds 已提交
242 243
};

244 245 246 247 248
static inline unsigned short req_get_ioprio(struct request *req)
{
	return req->ioprio;
}

L
Linus Torvalds 已提交
249
/*
250
 * State information carried for REQ_TYPE_PM_SUSPEND and REQ_TYPE_PM_RESUME
L
Linus Torvalds 已提交
251 252 253 254 255 256 257 258 259 260 261 262 263
 * 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>

264 265 266 267
typedef void (request_fn_proc) (struct request_queue *q);
typedef int (make_request_fn) (struct request_queue *q, struct bio *bio);
typedef int (prep_rq_fn) (struct request_queue *, struct request *);
typedef void (unplug_fn) (struct request_queue *);
L
Linus Torvalds 已提交
268 269

struct bio_vec;
270 271 272 273 274 275 276 277
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 *);
278
typedef void (prepare_flush_fn) (struct request_queue *, struct request *);
279
typedef void (softirq_done_fn)(struct request *);
280
typedef int (dma_drain_needed_fn)(struct request *);
281
typedef int (lld_busy_fn) (struct request_queue *q);
L
Linus Torvalds 已提交
282

J
Jens Axboe 已提交
283 284 285 286 287 288 289 290
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
291 292 293 294 295 296 297 298 299 300
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 */
301
	int real_max_depth;		/* what the array can hold */
L
Linus Torvalds 已提交
302 303 304
	atomic_t refcnt;		/* map can be shared */
};

305 306 307
#define BLK_SCSI_MAX_CMDS	(256)
#define BLK_SCSI_CMD_PER_LONG	(BLK_SCSI_MAX_CMDS / (sizeof(long) * 8))

308 309 310 311 312 313 314
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;
315 316 317 318
	unsigned int		physical_block_size;
	unsigned int		alignment_offset;
	unsigned int		io_min;
	unsigned int		io_opt;
319
	unsigned int		max_discard_sectors;
320 321
	unsigned int		discard_granularity;
	unsigned int		discard_alignment;
322 323

	unsigned short		logical_block_size;
324
	unsigned short		max_segments;
325

326
	unsigned char		misaligned;
327
	unsigned char		discard_misaligned;
328
	unsigned char		no_cluster;
329
	signed char		discard_zeroes_data;
330 331
};

L
Linus Torvalds 已提交
332 333 334 335 336 337 338
struct request_queue
{
	/*
	 * Together with queue_head for cacheline sharing
	 */
	struct list_head	queue_head;
	struct request		*last_merge;
J
Jens Axboe 已提交
339
	struct elevator_queue	*elevator;
L
Linus Torvalds 已提交
340 341 342 343 344 345 346 347 348 349 350 351

	/*
	 * the queue request freelist, one for reads and one for writes
	 */
	struct request_list	rq;

	request_fn_proc		*request_fn;
	make_request_fn		*make_request_fn;
	prep_rq_fn		*prep_rq_fn;
	unplug_fn		*unplug_fn;
	merge_bvec_fn		*merge_bvec_fn;
	prepare_flush_fn	*prepare_flush_fn;
352
	softirq_done_fn		*softirq_done_fn;
J
Jens Axboe 已提交
353
	rq_timed_out_fn		*rq_timed_out_fn;
354
	dma_drain_needed_fn	*dma_drain_needed;
355
	lld_busy_fn		*lld_busy_fn;
L
Linus Torvalds 已提交
356

357 358 359
	/*
	 * Dispatch queue sorting
	 */
J
Jens Axboe 已提交
360
	sector_t		end_sector;
361 362
	struct request		*boundary_rq;

L
Linus Torvalds 已提交
363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381
	/*
	 * Auto-unplugging state
	 */
	struct timer_list	unplug_timer;
	int			unplug_thresh;	/* After this many requests */
	unsigned long		unplug_delay;	/* After this many jiffies */
	struct work_struct	unplug_work;

	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;

	/*
	 * queue needs bounce pages for pages above this limit
	 */
A
Al Viro 已提交
382
	gfp_t			bounce_gfp;
L
Linus Torvalds 已提交
383 384 385 386 387 388 389

	/*
	 * various queue flags, see QUEUE_* below
	 */
	unsigned long		queue_flags;

	/*
已提交
390 391 392
	 * 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
393
	 */
已提交
394
	spinlock_t		__queue_lock;
L
Linus Torvalds 已提交
395 396 397 398 399 400 401 402 403 404 405 406 407 408 409
	spinlock_t		*queue_lock;

	/*
	 * queue kobject
	 */
	struct kobject kobj;

	/*
	 * 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
410 411
	void			*dma_drain_buffer;
	unsigned int		dma_drain_size;
412
	unsigned int		dma_pad_mask;
L
Linus Torvalds 已提交
413 414 415
	unsigned int		dma_alignment;

	struct blk_queue_tag	*queue_tags;
416
	struct list_head	tag_busy_list;
L
Linus Torvalds 已提交
417

418
	unsigned int		nr_sorted;
419
	unsigned int		in_flight[2];
L
Linus Torvalds 已提交
420

J
Jens Axboe 已提交
421 422 423 424
	unsigned int		rq_timeout;
	struct timer_list	timeout;
	struct list_head	timeout_list;

425 426
	struct queue_limits	limits;

L
Linus Torvalds 已提交
427 428 429 430 431
	/*
	 * sg stuff
	 */
	unsigned int		sg_timeout;
	unsigned int		sg_reserved_size;
432
	int			node;
433
#ifdef CONFIG_BLK_DEV_IO_TRACE
434
	struct blk_trace	*blk_trace;
435
#endif
L
Linus Torvalds 已提交
436 437 438
	/*
	 * reserved for flush operations
	 */
439 440 441 442
	unsigned int		ordered, next_ordered, ordseq;
	int			orderr, ordcolor;
	struct request		pre_flush_rq, bar_rq, post_flush_rq;
	struct request		*orig_bar_rq;
443 444

	struct mutex		sysfs_lock;
445 446 447 448

#if defined(CONFIG_BLK_DEV_BSG)
	struct bsg_class_device bsg_dev;
#endif
L
Linus Torvalds 已提交
449 450 451 452 453
};

#define QUEUE_FLAG_CLUSTER	0	/* cluster several segments into 1 */
#define QUEUE_FLAG_QUEUED	1	/* uses generic tag queueing */
#define QUEUE_FLAG_STOPPED	2	/* queue is stopped */
454 455
#define	QUEUE_FLAG_SYNCFULL	3	/* read queue has been filled */
#define QUEUE_FLAG_ASYNCFULL	4	/* write queue has been filled */
L
Linus Torvalds 已提交
456 457 458
#define QUEUE_FLAG_DEAD		5	/* queue being torn down */
#define QUEUE_FLAG_REENTER	6	/* Re-entrancy avoidance */
#define QUEUE_FLAG_PLUGGED	7	/* queue is plugged */
J
Jens Axboe 已提交
459
#define QUEUE_FLAG_ELVSWITCH	8	/* don't use elevator, just do FIFO */
460
#define QUEUE_FLAG_BIDI		9	/* queue supports bidi requests */
461
#define QUEUE_FLAG_NOMERGES    10	/* disable merge attempts */
462
#define QUEUE_FLAG_SAME_COMP   11	/* force complete on same CPU */
463
#define QUEUE_FLAG_FAIL_IO     12	/* fake timeout */
464
#define QUEUE_FLAG_STACKABLE   13	/* supports request stacking */
465
#define QUEUE_FLAG_NONROT      14	/* non-rotational device (SSD) */
466
#define QUEUE_FLAG_VIRT        QUEUE_FLAG_NONROT /* paravirt device */
467
#define QUEUE_FLAG_IO_STAT     15	/* do IO stats */
468
#define QUEUE_FLAG_DISCARD     16	/* supports DISCARD */
469
#define QUEUE_FLAG_NOXMERGES   17	/* No extended merges */
470 471 472

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

476 477
static inline int queue_is_locked(struct request_queue *q)
{
J
Jens Axboe 已提交
478
#ifdef CONFIG_SMP
479 480
	spinlock_t *lock = q->queue_lock;
	return lock && spin_is_locked(lock);
J
Jens Axboe 已提交
481 482 483
#else
	return 1;
#endif
484 485
}

N
Nick Piggin 已提交
486 487 488 489 490 491
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
492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517
static inline int queue_flag_test_and_clear(unsigned int flag,
					    struct request_queue *q)
{
	WARN_ON_ONCE(!queue_is_locked(q));

	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)
{
	WARN_ON_ONCE(!queue_is_locked(q));

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

	return 1;
}

N
Nick Piggin 已提交
518 519
static inline void queue_flag_set(unsigned int flag, struct request_queue *q)
{
520
	WARN_ON_ONCE(!queue_is_locked(q));
N
Nick Piggin 已提交
521 522 523 524 525 526 527 528 529
	__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);
}

530 531 532 533 534
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
535 536
static inline void queue_flag_clear(unsigned int flag, struct request_queue *q)
{
537
	WARN_ON_ONCE(!queue_is_locked(q));
N
Nick Piggin 已提交
538 539 540
	__clear_bit(flag, &q->queue_flags);
}

541 542 543 544 545 546 547 548 549 550 551 552
enum {
	/*
	 * Hardbarrier is supported with one of the following methods.
	 *
	 * NONE		: hardbarrier unsupported
	 * DRAIN	: ordering by draining is enough
	 * DRAIN_FLUSH	: ordering by draining w/ pre and post flushes
	 * DRAIN_FUA	: ordering by draining w/ pre flush and FUA write
	 * TAG		: ordering by tag is enough
	 * TAG_FLUSH	: ordering by tag w/ pre and post flushes
	 * TAG_FUA	: ordering by tag w/ pre flush and FUA write
	 */
553 554 555
	QUEUE_ORDERED_BY_DRAIN		= 0x01,
	QUEUE_ORDERED_BY_TAG		= 0x02,
	QUEUE_ORDERED_DO_PREFLUSH	= 0x10,
556
	QUEUE_ORDERED_DO_BAR		= 0x20,
557 558 559 560 561
	QUEUE_ORDERED_DO_POSTFLUSH	= 0x40,
	QUEUE_ORDERED_DO_FUA		= 0x80,

	QUEUE_ORDERED_NONE		= 0x00,

562 563
	QUEUE_ORDERED_DRAIN		= QUEUE_ORDERED_BY_DRAIN |
					  QUEUE_ORDERED_DO_BAR,
564 565 566 567 568 569 570
	QUEUE_ORDERED_DRAIN_FLUSH	= QUEUE_ORDERED_DRAIN |
					  QUEUE_ORDERED_DO_PREFLUSH |
					  QUEUE_ORDERED_DO_POSTFLUSH,
	QUEUE_ORDERED_DRAIN_FUA		= QUEUE_ORDERED_DRAIN |
					  QUEUE_ORDERED_DO_PREFLUSH |
					  QUEUE_ORDERED_DO_FUA,

571 572
	QUEUE_ORDERED_TAG		= QUEUE_ORDERED_BY_TAG |
					  QUEUE_ORDERED_DO_BAR,
573 574 575 576 577 578
	QUEUE_ORDERED_TAG_FLUSH		= QUEUE_ORDERED_TAG |
					  QUEUE_ORDERED_DO_PREFLUSH |
					  QUEUE_ORDERED_DO_POSTFLUSH,
	QUEUE_ORDERED_TAG_FUA		= QUEUE_ORDERED_TAG |
					  QUEUE_ORDERED_DO_PREFLUSH |
					  QUEUE_ORDERED_DO_FUA,
579 580 581 582 583 584 585 586 587 588 589

	/*
	 * Ordered operation sequence
	 */
	QUEUE_ORDSEQ_STARTED	= 0x01,	/* flushing in progress */
	QUEUE_ORDSEQ_DRAIN	= 0x02,	/* waiting for the queue to be drained */
	QUEUE_ORDSEQ_PREFLUSH	= 0x04,	/* pre-flushing in progress */
	QUEUE_ORDSEQ_BAR	= 0x08,	/* original barrier req in progress */
	QUEUE_ORDSEQ_POSTFLUSH	= 0x10,	/* post-flushing in progress */
	QUEUE_ORDSEQ_DONE	= 0x20,
};
L
Linus Torvalds 已提交
590 591 592 593

#define blk_queue_plugged(q)	test_bit(QUEUE_FLAG_PLUGGED, &(q)->queue_flags)
#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)
594
#define blk_queue_nomerges(q)	test_bit(QUEUE_FLAG_NOMERGES, &(q)->queue_flags)
595 596
#define blk_queue_noxmerges(q)	\
	test_bit(QUEUE_FLAG_NOXMERGES, &(q)->queue_flags)
597
#define blk_queue_nonrot(q)	test_bit(QUEUE_FLAG_NONROT, &(q)->queue_flags)
598
#define blk_queue_io_stat(q)	test_bit(QUEUE_FLAG_IO_STAT, &(q)->queue_flags)
599
#define blk_queue_flushing(q)	((q)->ordseq)
600 601
#define blk_queue_stackable(q)	\
	test_bit(QUEUE_FLAG_STACKABLE, &(q)->queue_flags)
602
#define blk_queue_discard(q)	test_bit(QUEUE_FLAG_DISCARD, &(q)->queue_flags)
L
Linus Torvalds 已提交
603

604 605 606 607 608
#define blk_fs_request(rq)	((rq)->cmd_type == REQ_TYPE_FS)
#define blk_pc_request(rq)	((rq)->cmd_type == REQ_TYPE_BLOCK_PC)
#define blk_special_request(rq)	((rq)->cmd_type == REQ_TYPE_SPECIAL)
#define blk_sense_request(rq)	((rq)->cmd_type == REQ_TYPE_SENSE)

609 610 611 612 613 614
#define blk_failfast_dev(rq)	((rq)->cmd_flags & REQ_FAILFAST_DEV)
#define blk_failfast_transport(rq) ((rq)->cmd_flags & REQ_FAILFAST_TRANSPORT)
#define blk_failfast_driver(rq)	((rq)->cmd_flags & REQ_FAILFAST_DRIVER)
#define blk_noretry_request(rq)	(blk_failfast_dev(rq) ||	\
				 blk_failfast_transport(rq) ||	\
				 blk_failfast_driver(rq))
615
#define blk_rq_started(rq)	((rq)->cmd_flags & REQ_STARTED)
616
#define blk_rq_io_stat(rq)	((rq)->cmd_flags & REQ_IO_STAT)
617
#define blk_rq_quiet(rq)	((rq)->cmd_flags & REQ_QUIET)
L
Linus Torvalds 已提交
618

619
#define blk_account_rq(rq)	(blk_rq_started(rq) && (blk_fs_request(rq) || blk_discard_rq(rq))) 
L
Linus Torvalds 已提交
620

621 622
#define blk_pm_suspend_request(rq)	((rq)->cmd_type == REQ_TYPE_PM_SUSPEND)
#define blk_pm_resume_request(rq)	((rq)->cmd_type == REQ_TYPE_PM_RESUME)
L
Linus Torvalds 已提交
623
#define blk_pm_request(rq)	\
624
	(blk_pm_suspend_request(rq) || blk_pm_resume_request(rq))
L
Linus Torvalds 已提交
625

626
#define blk_rq_cpu_valid(rq)	((rq)->cpu != -1)
627 628 629
#define blk_sorted_rq(rq)	((rq)->cmd_flags & REQ_SORTED)
#define blk_barrier_rq(rq)	((rq)->cmd_flags & REQ_HARDBARRIER)
#define blk_fua_rq(rq)		((rq)->cmd_flags & REQ_FUA)
D
David Woodhouse 已提交
630
#define blk_discard_rq(rq)	((rq)->cmd_flags & REQ_DISCARD)
631
#define blk_bidi_rq(rq)		((rq)->next_rq != NULL)
632 633
/* rq->queuelist of dequeued request must be list_empty() */
#define blk_queued_rq(rq)	(!list_empty(&(rq)->queuelist))
L
Linus Torvalds 已提交
634 635 636

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

637
#define rq_data_dir(rq)		((rq)->cmd_flags & 1)
L
Linus Torvalds 已提交
638

639
/*
640
 * We regard a request as sync, if either a read or a sync write
641
 */
642 643 644 645 646 647 648 649 650 651
static inline bool rw_is_sync(unsigned int rw_flags)
{
	return !(rw_flags & REQ_RW) || (rw_flags & REQ_RW_SYNC);
}

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

652
#define rq_is_meta(rq)		((rq)->cmd_flags & REQ_RW_META)
653
#define rq_noidle(rq)		((rq)->cmd_flags & REQ_NOIDLE)
654

655
static inline int blk_queue_full(struct request_queue *q, int sync)
L
Linus Torvalds 已提交
656
{
657 658 659
	if (sync)
		return test_bit(QUEUE_FLAG_SYNCFULL, &q->queue_flags);
	return test_bit(QUEUE_FLAG_ASYNCFULL, &q->queue_flags);
L
Linus Torvalds 已提交
660 661
}

662
static inline void blk_set_queue_full(struct request_queue *q, int sync)
L
Linus Torvalds 已提交
663
{
664 665
	if (sync)
		queue_flag_set(QUEUE_FLAG_SYNCFULL, q);
L
Linus Torvalds 已提交
666
	else
667
		queue_flag_set(QUEUE_FLAG_ASYNCFULL, q);
L
Linus Torvalds 已提交
668 669
}

670
static inline void blk_clear_queue_full(struct request_queue *q, int sync)
L
Linus Torvalds 已提交
671
{
672 673
	if (sync)
		queue_flag_clear(QUEUE_FLAG_SYNCFULL, q);
L
Linus Torvalds 已提交
674
	else
675
		queue_flag_clear(QUEUE_FLAG_ASYNCFULL, q);
L
Linus Torvalds 已提交
676 677 678 679 680 681 682 683 684 685
}


/*
 * mergeable request must not have _NOMERGE or _BARRIER bit set, nor may
 * it already be started by driver.
 */
#define RQ_NOMERGE_FLAGS	\
	(REQ_NOMERGE | REQ_STARTED | REQ_HARDBARRIER | REQ_SOFTBARRIER)
#define rq_mergeable(rq)	\
686 687
	(!((rq)->cmd_flags & RQ_NOMERGE_FLAGS) && \
	 (blk_discard_rq(rq) || blk_fs_request((rq))))
L
Linus Torvalds 已提交
688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704

/*
 * 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
 */
705 706

#if BITS_PER_LONG == 32
L
Linus Torvalds 已提交
707
#define BLK_BOUNCE_HIGH		((u64)blk_max_low_pfn << PAGE_SHIFT)
708 709 710 711
#else
#define BLK_BOUNCE_HIGH		-1ULL
#endif
#define BLK_BOUNCE_ANY		(-1ULL)
L
Linus Torvalds 已提交
712 713
#define BLK_BOUNCE_ISA		(ISA_DMA_THRESHOLD)

714 715 716 717
/*
 * default timeout for SG_IO if none specified
 */
#define BLK_DEFAULT_SG_TIMEOUT	(60 * HZ)
L
Linus Torvalds 已提交
718
#define BLK_MIN_SG_TIMEOUT	(7 * HZ)
719

720
#ifdef CONFIG_BOUNCE
L
Linus Torvalds 已提交
721
extern int init_emergency_isa_pool(void);
722
extern void blk_queue_bounce(struct request_queue *q, struct bio **bio);
L
Linus Torvalds 已提交
723 724 725 726 727
#else
static inline int init_emergency_isa_pool(void)
{
	return 0;
}
728
static inline void blk_queue_bounce(struct request_queue *q, struct bio **bio)
L
Linus Torvalds 已提交
729 730 731 732
{
}
#endif /* CONFIG_MMU */

733 734 735 736
struct rq_map_data {
	struct page **pages;
	int page_order;
	int nr_entries;
737
	unsigned long offset;
738
	int null_mapped;
739
	int from_user;
740 741
};

742 743 744 745 746 747
struct req_iterator {
	int i;
	struct bio *bio;
};

/* This should not be used directly - use rq_for_each_segment */
748 749
#define for_each_bio(_bio)		\
	for (; _bio; _bio = _bio->bi_next)
750
#define __rq_for_each_bio(_bio, rq)	\
L
Linus Torvalds 已提交
751 752 753
	if ((rq->bio))			\
		for (_bio = (rq)->bio; _bio; _bio = _bio->bi_next)

754 755 756 757 758 759 760
#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)

761 762 763 764 765 766 767 768 769 770 771
#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
772 773 774 775
extern int blk_register_queue(struct gendisk *disk);
extern void blk_unregister_queue(struct gendisk *disk);
extern void register_disk(struct gendisk *dev);
extern void generic_make_request(struct bio *bio);
776
extern void blk_rq_init(struct request_queue *q, struct request *rq);
L
Linus Torvalds 已提交
777
extern void blk_put_request(struct request *);
778 779
extern void __blk_put_request(struct request_queue *, struct request *);
extern struct request *blk_get_request(struct request_queue *, int, gfp_t);
780 781
extern struct request *blk_make_request(struct request_queue *, struct bio *,
					gfp_t);
782 783
extern void blk_insert_request(struct request_queue *, struct request *, int, void *);
extern void blk_requeue_request(struct request_queue *, struct request *);
784
extern int blk_rq_check_limits(struct request_queue *q, struct request *rq);
785
extern int blk_lld_busy(struct request_queue *q);
786 787 788 789 790
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);
791 792
extern int blk_insert_cloned_request(struct request_queue *q,
				     struct request *rq);
793
extern void blk_plug_device(struct request_queue *);
794
extern void blk_plug_device_unlocked(struct request_queue *);
795 796
extern int blk_remove_plug(struct request_queue *);
extern void blk_recount_segments(struct request_queue *, struct bio *);
797 798
extern int scsi_cmd_ioctl(struct request_queue *, struct gendisk *, fmode_t,
			  unsigned int, void __user *);
799 800
extern int sg_scsi_ioctl(struct request_queue *, struct gendisk *, fmode_t,
			 struct scsi_ioctl_command __user *);
801 802 803 804 805 806

/*
 * 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.
 */
807
static inline void blk_clear_queue_congested(struct request_queue *q, int sync)
808
{
809
	clear_bdi_congested(&q->backing_dev_info, sync);
810 811 812 813 814 815
}

/*
 * A queue has just entered congestion.  Flag that in the queue's VM-visible
 * state flags and increment the global gounter of congested queues.
 */
816
static inline void blk_set_queue_congested(struct request_queue *q, int sync)
817
{
818
	set_bdi_congested(&q->backing_dev_info, sync);
819 820
}

821 822
extern void blk_start_queue(struct request_queue *q);
extern void blk_stop_queue(struct request_queue *q);
L
Linus Torvalds 已提交
823
extern void blk_sync_queue(struct request_queue *q);
824
extern void __blk_stop_queue(struct request_queue *q);
N
Nick Piggin 已提交
825
extern void __blk_run_queue(struct request_queue *);
826
extern void blk_run_queue(struct request_queue *);
827
extern int blk_rq_map_user(struct request_queue *, struct request *,
828 829
			   struct rq_map_data *, void __user *, unsigned long,
			   gfp_t);
J
Jens Axboe 已提交
830
extern int blk_rq_unmap_user(struct bio *);
831 832
extern int blk_rq_map_kern(struct request_queue *, struct request *, void *, unsigned int, gfp_t);
extern int blk_rq_map_user_iov(struct request_queue *, struct request *,
833 834
			       struct rq_map_data *, struct sg_iovec *, int,
			       unsigned int, gfp_t);
835
extern int blk_execute_rq(struct request_queue *, struct gendisk *,
836
			  struct request *, int);
837
extern void blk_execute_rq_nowait(struct request_queue *, struct gendisk *,
838
				  struct request *, int, rq_end_io_fn *);
839
extern void blk_unplug(struct request_queue *q);
840

841
static inline struct request_queue *bdev_get_queue(struct block_device *bdev)
L
Linus Torvalds 已提交
842 843 844 845
{
	return bdev->bd_disk->queue;
}

846
/*
847 848 849 850 851 852
 * 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
853
 */
854 855
static inline sector_t blk_rq_pos(const struct request *rq)
{
856
	return rq->__sector;
857 858 859 860
}

static inline unsigned int blk_rq_bytes(const struct request *rq)
{
861
	return rq->__data_len;
862 863
}

864 865 866 867
static inline int blk_rq_cur_bytes(const struct request *rq)
{
	return rq->bio ? bio_cur_bytes(rq->bio) : 0;
}
868

869 870
extern unsigned int blk_rq_err_bytes(const struct request *rq);

871 872
static inline unsigned int blk_rq_sectors(const struct request *rq)
{
873
	return blk_rq_bytes(rq) >> 9;
874 875 876 877
}

static inline unsigned int blk_rq_cur_sectors(const struct request *rq)
{
878
	return blk_rq_cur_bytes(rq) >> 9;
879 880
}

881 882 883 884 885 886 887
/*
 * 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
888
/*
889 890 891 892 893
 * Request completion related functions.
 *
 * blk_update_request() completes given number of bytes and updates
 * the request without completing it.
 *
894 895
 * blk_end_request() and friends.  __blk_end_request() must be called
 * with the request queue spinlock acquired.
L
Linus Torvalds 已提交
896 897
 *
 * Several drivers define their own end_request and call
898 899
 * blk_end_request() for parts of the original function.
 * This prevents code duplication in drivers.
L
Linus Torvalds 已提交
900
 */
901 902
extern bool blk_update_request(struct request *rq, int error,
			       unsigned int nr_bytes);
903 904 905 906
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);
907
extern bool blk_end_request_err(struct request *rq, int error);
908 909 910 911
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);
912
extern bool __blk_end_request_err(struct request *rq, int error);
913

914
extern void blk_complete_request(struct request *);
J
Jens Axboe 已提交
915 916
extern void __blk_complete_request(struct request *);
extern void blk_abort_request(struct request *);
917
extern void blk_abort_queue(struct request_queue *);
918

L
Linus Torvalds 已提交
919 920 921
/*
 * Access functions for manipulating queue properties
 */
922
extern struct request_queue *blk_init_queue_node(request_fn_proc *rfn,
923
					spinlock_t *lock, int node_id);
924 925 926
extern struct request_queue *blk_init_allocated_queue_node(struct request_queue *,
							   request_fn_proc *,
							   spinlock_t *, int node_id);
927
extern struct request_queue *blk_init_queue(request_fn_proc *, spinlock_t *);
928 929
extern struct request_queue *blk_init_allocated_queue(struct request_queue *,
						      request_fn_proc *, spinlock_t *);
930 931 932
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);
933
extern void blk_queue_max_hw_sectors(struct request_queue *, unsigned int);
934
extern void blk_queue_max_segments(struct request_queue *, unsigned short);
935
extern void blk_queue_max_segment_size(struct request_queue *, unsigned int);
936 937
extern void blk_queue_max_discard_sectors(struct request_queue *q,
		unsigned int max_discard_sectors);
938
extern void blk_queue_logical_block_size(struct request_queue *, unsigned short);
939 940 941
extern void blk_queue_physical_block_size(struct request_queue *, unsigned short);
extern void blk_queue_alignment_offset(struct request_queue *q,
				       unsigned int alignment);
942
extern void blk_limits_io_min(struct queue_limits *limits, unsigned int min);
943
extern void blk_queue_io_min(struct request_queue *q, unsigned int min);
944
extern void blk_limits_io_opt(struct queue_limits *limits, unsigned int opt);
945
extern void blk_queue_io_opt(struct request_queue *q, unsigned int opt);
946
extern void blk_set_default_limits(struct queue_limits *lim);
947 948
extern int blk_stack_limits(struct queue_limits *t, struct queue_limits *b,
			    sector_t offset);
949 950
extern int bdev_stack_limits(struct queue_limits *t, struct block_device *bdev,
			    sector_t offset);
951 952
extern void disk_stack_limits(struct gendisk *disk, struct block_device *bdev,
			      sector_t offset);
953
extern void blk_queue_stack_limits(struct request_queue *t, struct request_queue *b);
954
extern void blk_queue_dma_pad(struct request_queue *, unsigned int);
955
extern void blk_queue_update_dma_pad(struct request_queue *, unsigned int);
956 957 958
extern int blk_queue_dma_drain(struct request_queue *q,
			       dma_drain_needed_fn *dma_drain_needed,
			       void *buf, unsigned int size);
959
extern void blk_queue_lld_busy(struct request_queue *q, lld_busy_fn *fn);
960 961 962 963
extern void blk_queue_segment_boundary(struct request_queue *, unsigned long);
extern void blk_queue_prep_rq(struct request_queue *, prep_rq_fn *pfn);
extern void blk_queue_merge_bvec(struct request_queue *, merge_bvec_fn *);
extern void blk_queue_dma_alignment(struct request_queue *, int);
964
extern void blk_queue_update_dma_alignment(struct request_queue *, int);
965
extern void blk_queue_softirq_done(struct request_queue *, softirq_done_fn *);
J
Jens Axboe 已提交
966 967
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);
L
Linus Torvalds 已提交
968
extern struct backing_dev_info *blk_get_backing_dev_info(struct block_device *bdev);
969
extern int blk_queue_ordered(struct request_queue *, unsigned, prepare_flush_fn *);
970
extern bool blk_do_ordered(struct request_queue *, struct request **);
971
extern unsigned blk_ordered_cur_seq(struct request_queue *);
972
extern unsigned blk_ordered_req_seq(struct request *);
973
extern bool blk_ordered_complete_seq(struct request_queue *, unsigned, int);
L
Linus Torvalds 已提交
974

975
extern int blk_rq_map_sg(struct request_queue *, struct request *, struct scatterlist *);
L
Linus Torvalds 已提交
976
extern void blk_dump_rq_flags(struct request *, char *);
977
extern void generic_unplug_device(struct request_queue *);
L
Linus Torvalds 已提交
978 979
extern long nr_blockdev_pages(void);

980 981 982 983
int blk_get_queue(struct request_queue *);
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
984 985 986 987

/*
 * tag stuff
 */
988
#define blk_rq_tagged(rq)		((rq)->cmd_flags & REQ_QUEUED)
989 990 991 992 993 994 995
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 *);
996 997
extern struct blk_queue_tag *blk_init_tags(int);
extern void blk_free_tags(struct blk_queue_tag *);
L
Linus Torvalds 已提交
998

999 1000 1001 1002 1003 1004 1005
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];
}
1006 1007 1008 1009 1010 1011 1012 1013 1014 1015
enum{
	BLKDEV_WAIT,	/* wait for completion */
	BLKDEV_BARRIER,	/*issue request with barrier */
};
#define BLKDEV_IFL_WAIT		(1 << BLKDEV_WAIT)
#define BLKDEV_IFL_BARRIER	(1 << BLKDEV_BARRIER)
extern int blkdev_issue_flush(struct block_device *, gfp_t, sector_t *,
			unsigned long);
extern int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
		sector_t nr_sects, gfp_t gfp_mask, unsigned long flags);
1016 1017
extern int blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
			sector_t nr_sects, gfp_t gfp_mask, unsigned long flags);
D
David Woodhouse 已提交
1018
static inline int sb_issue_discard(struct super_block *sb,
1019
				   sector_t block, sector_t nr_blocks)
D
David Woodhouse 已提交
1020 1021 1022
{
	block <<= (sb->s_blocksize_bits - 9);
	nr_blocks <<= (sb->s_blocksize_bits - 9);
1023
	return blkdev_issue_discard(sb->s_bdev, block, nr_blocks, GFP_KERNEL,
1024
				   BLKDEV_IFL_WAIT | BLKDEV_IFL_BARRIER);
D
David Woodhouse 已提交
1025
}
L
Linus Torvalds 已提交
1026

1027
extern int blk_verify_command(unsigned char *cmd, fmode_t has_write_perm);
1028

1029 1030 1031 1032 1033 1034 1035
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,
};
1036

L
Linus Torvalds 已提交
1037 1038
#define blkdev_entry_to_request(entry) list_entry((entry), struct request, queuelist)

1039 1040
static inline unsigned long queue_bounce_pfn(struct request_queue *q)
{
1041
	return q->limits.bounce_pfn;
1042 1043 1044 1045
}

static inline unsigned long queue_segment_boundary(struct request_queue *q)
{
1046
	return q->limits.seg_boundary_mask;
1047 1048 1049 1050
}

static inline unsigned int queue_max_sectors(struct request_queue *q)
{
1051
	return q->limits.max_sectors;
1052 1053 1054 1055
}

static inline unsigned int queue_max_hw_sectors(struct request_queue *q)
{
1056
	return q->limits.max_hw_sectors;
1057 1058
}

1059
static inline unsigned short queue_max_segments(struct request_queue *q)
1060
{
1061
	return q->limits.max_segments;
1062 1063 1064 1065
}

static inline unsigned int queue_max_segment_size(struct request_queue *q)
{
1066
	return q->limits.max_segment_size;
1067 1068
}

1069
static inline unsigned short queue_logical_block_size(struct request_queue *q)
L
Linus Torvalds 已提交
1070 1071 1072
{
	int retval = 512;

1073 1074
	if (q && q->limits.logical_block_size)
		retval = q->limits.logical_block_size;
L
Linus Torvalds 已提交
1075 1076 1077 1078

	return retval;
}

1079
static inline unsigned short bdev_logical_block_size(struct block_device *bdev)
L
Linus Torvalds 已提交
1080
{
1081
	return queue_logical_block_size(bdev_get_queue(bdev));
L
Linus Torvalds 已提交
1082 1083
}

1084 1085 1086 1087 1088
static inline unsigned int queue_physical_block_size(struct request_queue *q)
{
	return q->limits.physical_block_size;
}

M
Martin K. Petersen 已提交
1089 1090 1091 1092 1093
static inline int bdev_physical_block_size(struct block_device *bdev)
{
	return queue_physical_block_size(bdev_get_queue(bdev));
}

1094 1095 1096 1097 1098
static inline unsigned int queue_io_min(struct request_queue *q)
{
	return q->limits.io_min;
}

M
Martin K. Petersen 已提交
1099 1100 1101 1102 1103
static inline int bdev_io_min(struct block_device *bdev)
{
	return queue_io_min(bdev_get_queue(bdev));
}

1104 1105 1106 1107 1108
static inline unsigned int queue_io_opt(struct request_queue *q)
{
	return q->limits.io_opt;
}

M
Martin K. Petersen 已提交
1109 1110 1111 1112 1113
static inline int bdev_io_opt(struct block_device *bdev)
{
	return queue_io_opt(bdev_get_queue(bdev));
}

1114 1115
static inline int queue_alignment_offset(struct request_queue *q)
{
M
Martin K. Petersen 已提交
1116
	if (q->limits.misaligned)
1117 1118
		return -1;

M
Martin K. Petersen 已提交
1119
	return q->limits.alignment_offset;
1120 1121
}

1122
static inline int queue_limit_alignment_offset(struct queue_limits *lim, sector_t sector)
1123 1124
{
	unsigned int granularity = max(lim->physical_block_size, lim->io_min);
1125
	unsigned int alignment = (sector << 9) & (granularity - 1);
1126

1127 1128
	return (granularity + lim->alignment_offset - alignment)
		& (granularity - 1);
1129 1130
}

M
Martin K. Petersen 已提交
1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143
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;
}

1144 1145 1146 1147 1148 1149 1150 1151
static inline int queue_discard_alignment(struct request_queue *q)
{
	if (q->limits.discard_misaligned)
		return -1;

	return q->limits.discard_alignment;
}

1152
static inline int queue_limit_discard_alignment(struct queue_limits *lim, sector_t sector)
1153
{
1154 1155 1156 1157
	unsigned int alignment = (sector << 9) & (lim->discard_granularity - 1);

	return (lim->discard_granularity + lim->discard_alignment - alignment)
		& (lim->discard_granularity - 1);
1158 1159
}

1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172
static inline unsigned int queue_discard_zeroes_data(struct request_queue *q)
{
	if (q->limits.discard_zeroes_data == 1)
		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));
}

1173
static inline int queue_dma_alignment(struct request_queue *q)
L
Linus Torvalds 已提交
1174
{
1175
	return q ? q->dma_alignment : 511;
L
Linus Torvalds 已提交
1176 1177
}

1178 1179 1180 1181 1182 1183 1184
static inline int blk_rq_aligned(struct request_queue *q, void *addr,
				 unsigned int len)
{
	unsigned int alignment = queue_dma_alignment(q) | q->dma_pad_mask;
	return !((unsigned long)addr & alignment) && !(len & alignment);
}

L
Linus Torvalds 已提交
1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195
/* 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;
}

1196
static inline unsigned int block_size(struct block_device *bdev)
L
Linus Torvalds 已提交
1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210
{
	return bdev->bd_block_size;
}

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;
1211
int kblockd_schedule_work(struct request_queue *q, struct work_struct *work);
L
Linus Torvalds 已提交
1212

1213 1214 1215 1216 1217 1218 1219 1220 1221 1222
#ifdef CONFIG_BLK_CGROUP
static inline void set_start_time_ns(struct request *req)
{
	req->start_time_ns = sched_clock();
}

static inline void set_io_start_time_ns(struct request *req)
{
	req->io_start_time_ns = sched_clock();
}
1223 1224 1225 1226 1227 1228 1229 1230 1231 1232

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;
}
1233 1234 1235
#else
static inline void set_start_time_ns(struct request *req) {}
static inline void set_io_start_time_ns(struct request *req) {}
1236 1237 1238 1239 1240 1241 1242 1243
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;
}
1244 1245
#endif

L
Linus Torvalds 已提交
1246 1247 1248 1249 1250
#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) "-*")

1251 1252
#if defined(CONFIG_BLK_DEV_INTEGRITY)

1253 1254
#define INTEGRITY_FLAG_READ	2	/* verify data integrity on read */
#define INTEGRITY_FLAG_WRITE	4	/* generate data integrity on write */
1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287

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;
};

extern int blk_integrity_register(struct gendisk *, struct blk_integrity *);
extern void blk_integrity_unregister(struct gendisk *);
1288
extern int blk_integrity_compare(struct gendisk *, struct gendisk *);
1289 1290 1291
extern int blk_rq_map_integrity_sg(struct request *, struct scatterlist *);
extern int blk_rq_count_integrity_sg(struct request *);

1292 1293 1294 1295 1296 1297
static inline
struct blk_integrity *bdev_get_integrity(struct block_device *bdev)
{
	return bdev->bd_disk->integrity;
}

1298 1299 1300 1301 1302
static inline struct blk_integrity *blk_get_integrity(struct gendisk *disk)
{
	return disk->integrity;
}

1303 1304
static inline int blk_integrity_rq(struct request *rq)
{
1305 1306 1307
	if (rq->bio == NULL)
		return 0;

1308 1309 1310 1311 1312 1313 1314 1315
	return bio_integrity(rq->bio);
}

#else /* CONFIG_BLK_DEV_INTEGRITY */

#define blk_integrity_rq(rq)			(0)
#define blk_rq_count_integrity_sg(a)		(0)
#define blk_rq_map_integrity_sg(a, b)		(0)
1316
#define bdev_get_integrity(a)			(0)
1317
#define blk_get_integrity(a)			(0)
1318 1319 1320 1321 1322 1323
#define blk_integrity_compare(a, b)		(0)
#define blk_integrity_register(a, b)		(0)
#define blk_integrity_unregister(a)		do { } while (0);

#endif /* CONFIG_BLK_DEV_INTEGRITY */

1324
struct block_device_operations {
A
Al Viro 已提交
1325 1326 1327 1328 1329
	int (*open) (struct block_device *, fmode_t);
	int (*release) (struct gendisk *, fmode_t);
	int (*locked_ioctl) (struct block_device *, fmode_t, unsigned, unsigned long);
	int (*ioctl) (struct block_device *, fmode_t, unsigned, unsigned long);
	int (*compat_ioctl) (struct block_device *, fmode_t, unsigned, unsigned long);
1330 1331 1332
	int (*direct_access) (struct block_device *, sector_t,
						void **, unsigned long *);
	int (*media_changed) (struct gendisk *);
1333
	void (*unlock_native_capacity) (struct gendisk *);
1334 1335
	int (*revalidate_disk) (struct gendisk *);
	int (*getgeo)(struct block_device *, struct hd_geometry *);
1336 1337
	/* this callback is with swap_lock and sometimes page table lock held */
	void (*swap_slot_free_notify) (struct block_device *, unsigned long);
1338 1339 1340
	struct module *owner;
};

1341 1342
extern int __blkdev_driver_ioctl(struct block_device *, fmode_t, unsigned int,
				 unsigned long);
1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355
#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;
}

#endif /* CONFIG_BLOCK */

L
Linus Torvalds 已提交
1356
#endif