ll_rw_blk.c 92.3 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
/*
 * Copyright (C) 1991, 1992 Linus Torvalds
 * Copyright (C) 1994,      Karl Keyte: Added support for disk statistics
 * Elevator latency, (C) 2000  Andrea Arcangeli <andrea@suse.de> SuSE
 * Queue request tables / lock, selectable elevator, Jens Axboe <axboe@suse.de>
 * kernel-doc documentation started by NeilBrown <neilb@cse.unsw.edu.au> -  July2000
 * bio rewrite, highmem i/o, etc, Jens Axboe <axboe@suse.de> - may 2001
 */

/*
 * This handles all read/write requests to block devices
 */
#include <linux/config.h>
#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/backing-dev.h>
#include <linux/bio.h>
#include <linux/blkdev.h>
#include <linux/highmem.h>
#include <linux/mm.h>
#include <linux/kernel_stat.h>
#include <linux/string.h>
#include <linux/init.h>
#include <linux/bootmem.h>	/* for max_pfn/max_low_pfn */
#include <linux/completion.h>
#include <linux/slab.h>
#include <linux/swap.h>
#include <linux/writeback.h>
29
#include <linux/blkdev.h>
L
Linus Torvalds 已提交
30 31 32 33 34 35 36 37

/*
 * for max sense size
 */
#include <scsi/scsi_cmnd.h>

static void blk_unplug_work(void *data);
static void blk_unplug_timeout(unsigned long data);
38
static void drive_stat_acct(struct request *rq, int nr_sectors, int new_io);
L
Linus Torvalds 已提交
39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235

/*
 * For the allocated request tables
 */
static kmem_cache_t *request_cachep;

/*
 * For queue allocation
 */
static kmem_cache_t *requestq_cachep;

/*
 * For io context allocations
 */
static kmem_cache_t *iocontext_cachep;

static wait_queue_head_t congestion_wqh[2] = {
		__WAIT_QUEUE_HEAD_INITIALIZER(congestion_wqh[0]),
		__WAIT_QUEUE_HEAD_INITIALIZER(congestion_wqh[1])
	};

/*
 * Controlling structure to kblockd
 */
static struct workqueue_struct *kblockd_workqueue; 

unsigned long blk_max_low_pfn, blk_max_pfn;

EXPORT_SYMBOL(blk_max_low_pfn);
EXPORT_SYMBOL(blk_max_pfn);

/* Amount of time in which a process may batch requests */
#define BLK_BATCH_TIME	(HZ/50UL)

/* Number of requests a "batching" process may submit */
#define BLK_BATCH_REQ	32

/*
 * Return the threshold (number of used requests) at which the queue is
 * considered to be congested.  It include a little hysteresis to keep the
 * context switch rate down.
 */
static inline int queue_congestion_on_threshold(struct request_queue *q)
{
	return q->nr_congestion_on;
}

/*
 * The threshold at which a queue is considered to be uncongested
 */
static inline int queue_congestion_off_threshold(struct request_queue *q)
{
	return q->nr_congestion_off;
}

static void blk_queue_congestion_threshold(struct request_queue *q)
{
	int nr;

	nr = q->nr_requests - (q->nr_requests / 8) + 1;
	if (nr > q->nr_requests)
		nr = q->nr_requests;
	q->nr_congestion_on = nr;

	nr = q->nr_requests - (q->nr_requests / 8) - (q->nr_requests / 16) - 1;
	if (nr < 1)
		nr = 1;
	q->nr_congestion_off = nr;
}

/*
 * 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.
 */
static void clear_queue_congested(request_queue_t *q, int rw)
{
	enum bdi_state bit;
	wait_queue_head_t *wqh = &congestion_wqh[rw];

	bit = (rw == WRITE) ? BDI_write_congested : BDI_read_congested;
	clear_bit(bit, &q->backing_dev_info.state);
	smp_mb__after_clear_bit();
	if (waitqueue_active(wqh))
		wake_up(wqh);
}

/*
 * A queue has just entered congestion.  Flag that in the queue's VM-visible
 * state flags and increment the global gounter of congested queues.
 */
static void set_queue_congested(request_queue_t *q, int rw)
{
	enum bdi_state bit;

	bit = (rw == WRITE) ? BDI_write_congested : BDI_read_congested;
	set_bit(bit, &q->backing_dev_info.state);
}

/**
 * blk_get_backing_dev_info - get the address of a queue's backing_dev_info
 * @bdev:	device
 *
 * Locates the passed device's request queue and returns the address of its
 * backing_dev_info
 *
 * Will return NULL if the request queue cannot be located.
 */
struct backing_dev_info *blk_get_backing_dev_info(struct block_device *bdev)
{
	struct backing_dev_info *ret = NULL;
	request_queue_t *q = bdev_get_queue(bdev);

	if (q)
		ret = &q->backing_dev_info;
	return ret;
}

EXPORT_SYMBOL(blk_get_backing_dev_info);

void blk_queue_activity_fn(request_queue_t *q, activity_fn *fn, void *data)
{
	q->activity_fn = fn;
	q->activity_data = data;
}

EXPORT_SYMBOL(blk_queue_activity_fn);

/**
 * blk_queue_prep_rq - set a prepare_request function for queue
 * @q:		queue
 * @pfn:	prepare_request function
 *
 * It's possible for a queue to register a prepare_request callback which
 * is invoked before the request is handed to the request_fn. The goal of
 * the function is to prepare a request for I/O, it can be used to build a
 * cdb from the request data for instance.
 *
 */
void blk_queue_prep_rq(request_queue_t *q, prep_rq_fn *pfn)
{
	q->prep_rq_fn = pfn;
}

EXPORT_SYMBOL(blk_queue_prep_rq);

/**
 * blk_queue_merge_bvec - set a merge_bvec function for queue
 * @q:		queue
 * @mbfn:	merge_bvec_fn
 *
 * Usually queues have static limitations on the max sectors or segments that
 * we can put in a request. Stacking drivers may have some settings that
 * are dynamic, and thus we have to query the queue whether it is ok to
 * add a new bio_vec to a bio at a given offset or not. If the block device
 * has such limitations, it needs to register a merge_bvec_fn to control
 * the size of bio's sent to it. Note that a block device *must* allow a
 * single page to be added to an empty bio. The block device driver may want
 * to use the bio_split() function to deal with these bio's. By default
 * no merge_bvec_fn is defined for a queue, and only the fixed limits are
 * honored.
 */
void blk_queue_merge_bvec(request_queue_t *q, merge_bvec_fn *mbfn)
{
	q->merge_bvec_fn = mbfn;
}

EXPORT_SYMBOL(blk_queue_merge_bvec);

/**
 * blk_queue_make_request - define an alternate make_request function for a device
 * @q:  the request queue for the device to be affected
 * @mfn: the alternate make_request function
 *
 * Description:
 *    The normal way for &struct bios to be passed to a device
 *    driver is for them to be collected into requests on a request
 *    queue, and then to allow the device driver to select requests
 *    off that queue when it is ready.  This works well for many block
 *    devices. However some block devices (typically virtual devices
 *    such as md or lvm) do not benefit from the processing on the
 *    request queue, and are served best by having the requests passed
 *    directly to them.  This can be achieved by providing a function
 *    to blk_queue_make_request().
 *
 * Caveat:
 *    The driver that does this *must* be able to deal appropriately
 *    with buffers in "highmemory". This can be accomplished by either calling
 *    __bio_kmap_atomic() to get a temporary kernel mapping, or by calling
 *    blk_queue_bounce() to create a buffer in normal memory.
 **/
void blk_queue_make_request(request_queue_t * q, make_request_fn * mfn)
{
	/*
	 * set defaults
	 */
	q->nr_requests = BLKDEV_MAX_RQ;
236 237
	blk_queue_max_phys_segments(q, MAX_PHYS_SEGMENTS);
	blk_queue_max_hw_segments(q, MAX_HW_SEGMENTS);
L
Linus Torvalds 已提交
238 239 240 241
	q->make_request_fn = mfn;
	q->backing_dev_info.ra_pages = (VM_MAX_READAHEAD * 1024) / PAGE_CACHE_SIZE;
	q->backing_dev_info.state = 0;
	q->backing_dev_info.capabilities = BDI_CAP_MAP_COPY;
242
	blk_queue_max_sectors(q, SAFE_MAX_SECTORS);
L
Linus Torvalds 已提交
243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274
	blk_queue_hardsect_size(q, 512);
	blk_queue_dma_alignment(q, 511);
	blk_queue_congestion_threshold(q);
	q->nr_batching = BLK_BATCH_REQ;

	q->unplug_thresh = 4;		/* hmm */
	q->unplug_delay = (3 * HZ) / 1000;	/* 3 milliseconds */
	if (q->unplug_delay == 0)
		q->unplug_delay = 1;

	INIT_WORK(&q->unplug_work, blk_unplug_work, q);

	q->unplug_timer.function = blk_unplug_timeout;
	q->unplug_timer.data = (unsigned long)q;

	/*
	 * by default assume old behaviour and bounce for any highmem page
	 */
	blk_queue_bounce_limit(q, BLK_BOUNCE_HIGH);

	blk_queue_activity_fn(q, NULL, NULL);
}

EXPORT_SYMBOL(blk_queue_make_request);

static inline void rq_init(request_queue_t *q, struct request *rq)
{
	INIT_LIST_HEAD(&rq->queuelist);

	rq->errors = 0;
	rq->rq_status = RQ_ACTIVE;
	rq->bio = rq->biotail = NULL;
275
	rq->ioprio = 0;
L
Linus Torvalds 已提交
276 277 278 279 280 281 282
	rq->buffer = NULL;
	rq->ref_count = 1;
	rq->q = q;
	rq->waiting = NULL;
	rq->special = NULL;
	rq->data_len = 0;
	rq->data = NULL;
M
Mike Christie 已提交
283
	rq->nr_phys_segments = 0;
L
Linus Torvalds 已提交
284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351
	rq->sense = NULL;
	rq->end_io = NULL;
	rq->end_io_data = NULL;
}

/**
 * blk_queue_ordered - does this queue support ordered writes
 * @q:     the request queue
 * @flag:  see below
 *
 * Description:
 *   For journalled file systems, doing ordered writes on a commit
 *   block instead of explicitly doing wait_on_buffer (which is bad
 *   for performance) can be a big win. Block drivers supporting this
 *   feature should call this function and indicate so.
 *
 **/
void blk_queue_ordered(request_queue_t *q, int flag)
{
	switch (flag) {
		case QUEUE_ORDERED_NONE:
			if (q->flush_rq)
				kmem_cache_free(request_cachep, q->flush_rq);
			q->flush_rq = NULL;
			q->ordered = flag;
			break;
		case QUEUE_ORDERED_TAG:
			q->ordered = flag;
			break;
		case QUEUE_ORDERED_FLUSH:
			q->ordered = flag;
			if (!q->flush_rq)
				q->flush_rq = kmem_cache_alloc(request_cachep,
								GFP_KERNEL);
			break;
		default:
			printk("blk_queue_ordered: bad value %d\n", flag);
			break;
	}
}

EXPORT_SYMBOL(blk_queue_ordered);

/**
 * blk_queue_issue_flush_fn - set function for issuing a flush
 * @q:     the request queue
 * @iff:   the function to be called issuing the flush
 *
 * Description:
 *   If a driver supports issuing a flush command, the support is notified
 *   to the block layer by defining it through this call.
 *
 **/
void blk_queue_issue_flush_fn(request_queue_t *q, issue_flush_fn *iff)
{
	q->issue_flush_fn = iff;
}

EXPORT_SYMBOL(blk_queue_issue_flush_fn);

/*
 * Cache flushing for ordered writes handling
 */
static void blk_pre_flush_end_io(struct request *flush_rq)
{
	struct request *rq = flush_rq->end_io_data;
	request_queue_t *q = rq->q;

352 353
	elv_completed_request(q, flush_rq);

L
Linus Torvalds 已提交
354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369
	rq->flags |= REQ_BAR_PREFLUSH;

	if (!flush_rq->errors)
		elv_requeue_request(q, rq);
	else {
		q->end_flush_fn(q, flush_rq);
		clear_bit(QUEUE_FLAG_FLUSH, &q->queue_flags);
		q->request_fn(q);
	}
}

static void blk_post_flush_end_io(struct request *flush_rq)
{
	struct request *rq = flush_rq->end_io_data;
	request_queue_t *q = rq->q;

370 371
	elv_completed_request(q, flush_rq);

L
Linus Torvalds 已提交
372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557
	rq->flags |= REQ_BAR_POSTFLUSH;

	q->end_flush_fn(q, flush_rq);
	clear_bit(QUEUE_FLAG_FLUSH, &q->queue_flags);
	q->request_fn(q);
}

struct request *blk_start_pre_flush(request_queue_t *q, struct request *rq)
{
	struct request *flush_rq = q->flush_rq;

	BUG_ON(!blk_barrier_rq(rq));

	if (test_and_set_bit(QUEUE_FLAG_FLUSH, &q->queue_flags))
		return NULL;

	rq_init(q, flush_rq);
	flush_rq->elevator_private = NULL;
	flush_rq->flags = REQ_BAR_FLUSH;
	flush_rq->rq_disk = rq->rq_disk;
	flush_rq->rl = NULL;

	/*
	 * prepare_flush returns 0 if no flush is needed, just mark both
	 * pre and post flush as done in that case
	 */
	if (!q->prepare_flush_fn(q, flush_rq)) {
		rq->flags |= REQ_BAR_PREFLUSH | REQ_BAR_POSTFLUSH;
		clear_bit(QUEUE_FLAG_FLUSH, &q->queue_flags);
		return rq;
	}

	/*
	 * some drivers dequeue requests right away, some only after io
	 * completion. make sure the request is dequeued.
	 */
	if (!list_empty(&rq->queuelist))
		blkdev_dequeue_request(rq);

	flush_rq->end_io_data = rq;
	flush_rq->end_io = blk_pre_flush_end_io;

	__elv_add_request(q, flush_rq, ELEVATOR_INSERT_FRONT, 0);
	return flush_rq;
}

static void blk_start_post_flush(request_queue_t *q, struct request *rq)
{
	struct request *flush_rq = q->flush_rq;

	BUG_ON(!blk_barrier_rq(rq));

	rq_init(q, flush_rq);
	flush_rq->elevator_private = NULL;
	flush_rq->flags = REQ_BAR_FLUSH;
	flush_rq->rq_disk = rq->rq_disk;
	flush_rq->rl = NULL;

	if (q->prepare_flush_fn(q, flush_rq)) {
		flush_rq->end_io_data = rq;
		flush_rq->end_io = blk_post_flush_end_io;

		__elv_add_request(q, flush_rq, ELEVATOR_INSERT_FRONT, 0);
		q->request_fn(q);
	}
}

static inline int blk_check_end_barrier(request_queue_t *q, struct request *rq,
					int sectors)
{
	if (sectors > rq->nr_sectors)
		sectors = rq->nr_sectors;

	rq->nr_sectors -= sectors;
	return rq->nr_sectors;
}

static int __blk_complete_barrier_rq(request_queue_t *q, struct request *rq,
				     int sectors, int queue_locked)
{
	if (q->ordered != QUEUE_ORDERED_FLUSH)
		return 0;
	if (!blk_fs_request(rq) || !blk_barrier_rq(rq))
		return 0;
	if (blk_barrier_postflush(rq))
		return 0;

	if (!blk_check_end_barrier(q, rq, sectors)) {
		unsigned long flags = 0;

		if (!queue_locked)
			spin_lock_irqsave(q->queue_lock, flags);

		blk_start_post_flush(q, rq);

		if (!queue_locked)
			spin_unlock_irqrestore(q->queue_lock, flags);
	}

	return 1;
}

/**
 * blk_complete_barrier_rq - complete possible barrier request
 * @q:  the request queue for the device
 * @rq:  the request
 * @sectors:  number of sectors to complete
 *
 * Description:
 *   Used in driver end_io handling to determine whether to postpone
 *   completion of a barrier request until a post flush has been done. This
 *   is the unlocked variant, used if the caller doesn't already hold the
 *   queue lock.
 **/
int blk_complete_barrier_rq(request_queue_t *q, struct request *rq, int sectors)
{
	return __blk_complete_barrier_rq(q, rq, sectors, 0);
}
EXPORT_SYMBOL(blk_complete_barrier_rq);

/**
 * blk_complete_barrier_rq_locked - complete possible barrier request
 * @q:  the request queue for the device
 * @rq:  the request
 * @sectors:  number of sectors to complete
 *
 * Description:
 *   See blk_complete_barrier_rq(). This variant must be used if the caller
 *   holds the queue lock.
 **/
int blk_complete_barrier_rq_locked(request_queue_t *q, struct request *rq,
				   int sectors)
{
	return __blk_complete_barrier_rq(q, rq, sectors, 1);
}
EXPORT_SYMBOL(blk_complete_barrier_rq_locked);

/**
 * blk_queue_bounce_limit - set bounce buffer limit for queue
 * @q:  the request queue for the device
 * @dma_addr:   bus address limit
 *
 * Description:
 *    Different hardware can have different requirements as to what pages
 *    it can do I/O directly to. A low level driver can call
 *    blk_queue_bounce_limit to have lower memory pages allocated as bounce
 *    buffers for doing I/O to pages residing above @page. By default
 *    the block layer sets this to the highest numbered "low" memory page.
 **/
void blk_queue_bounce_limit(request_queue_t *q, u64 dma_addr)
{
	unsigned long bounce_pfn = dma_addr >> PAGE_SHIFT;

	/*
	 * set appropriate bounce gfp mask -- unfortunately we don't have a
	 * full 4GB zone, so we have to resort to low memory for any bounces.
	 * ISA has its own < 16MB zone.
	 */
	if (bounce_pfn < blk_max_low_pfn) {
		BUG_ON(dma_addr < BLK_BOUNCE_ISA);
		init_emergency_isa_pool();
		q->bounce_gfp = GFP_NOIO | GFP_DMA;
	} else
		q->bounce_gfp = GFP_NOIO;

	q->bounce_pfn = bounce_pfn;
}

EXPORT_SYMBOL(blk_queue_bounce_limit);

/**
 * blk_queue_max_sectors - set max sectors for a request for this queue
 * @q:  the request queue for the device
 * @max_sectors:  max sectors in the usual 512b unit
 *
 * Description:
 *    Enables a low level driver to set an upper limit on the size of
 *    received requests.
 **/
void blk_queue_max_sectors(request_queue_t *q, unsigned short max_sectors)
{
	if ((max_sectors << 9) < PAGE_CACHE_SIZE) {
		max_sectors = 1 << (PAGE_CACHE_SHIFT - 9);
		printk("%s: set to minimum %d\n", __FUNCTION__, max_sectors);
	}

558 559 560 561 562 563
	if (BLK_DEF_MAX_SECTORS > max_sectors)
		q->max_hw_sectors = q->max_sectors = max_sectors;
 	else {
		q->max_sectors = BLK_DEF_MAX_SECTORS;
		q->max_hw_sectors = max_sectors;
	}
L
Linus Torvalds 已提交
564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664
}

EXPORT_SYMBOL(blk_queue_max_sectors);

/**
 * blk_queue_max_phys_segments - set max phys segments for a request for this queue
 * @q:  the request queue for the device
 * @max_segments:  max number of segments
 *
 * Description:
 *    Enables a low level driver to set an upper limit on the number of
 *    physical data segments in a request.  This would be the largest sized
 *    scatter list the driver could handle.
 **/
void blk_queue_max_phys_segments(request_queue_t *q, unsigned short max_segments)
{
	if (!max_segments) {
		max_segments = 1;
		printk("%s: set to minimum %d\n", __FUNCTION__, max_segments);
	}

	q->max_phys_segments = max_segments;
}

EXPORT_SYMBOL(blk_queue_max_phys_segments);

/**
 * blk_queue_max_hw_segments - set max hw segments for a request for this queue
 * @q:  the request queue for the device
 * @max_segments:  max number of segments
 *
 * Description:
 *    Enables a low level driver to set an upper limit on the number of
 *    hw data segments in a request.  This would be the largest number of
 *    address/length pairs the host adapter can actually give as once
 *    to the device.
 **/
void blk_queue_max_hw_segments(request_queue_t *q, unsigned short max_segments)
{
	if (!max_segments) {
		max_segments = 1;
		printk("%s: set to minimum %d\n", __FUNCTION__, max_segments);
	}

	q->max_hw_segments = max_segments;
}

EXPORT_SYMBOL(blk_queue_max_hw_segments);

/**
 * blk_queue_max_segment_size - set max segment size for blk_rq_map_sg
 * @q:  the request queue for the device
 * @max_size:  max size of segment in bytes
 *
 * Description:
 *    Enables a low level driver to set an upper limit on the size of a
 *    coalesced segment
 **/
void blk_queue_max_segment_size(request_queue_t *q, unsigned int max_size)
{
	if (max_size < PAGE_CACHE_SIZE) {
		max_size = PAGE_CACHE_SIZE;
		printk("%s: set to minimum %d\n", __FUNCTION__, max_size);
	}

	q->max_segment_size = max_size;
}

EXPORT_SYMBOL(blk_queue_max_segment_size);

/**
 * blk_queue_hardsect_size - set hardware sector size for the queue
 * @q:  the request queue for the device
 * @size:  the hardware sector size, in bytes
 *
 * Description:
 *   This should typically be set to the lowest possible sector size
 *   that the hardware can operate on (possible without reverting to
 *   even internal read-modify-write operations). Usually the default
 *   of 512 covers most hardware.
 **/
void blk_queue_hardsect_size(request_queue_t *q, unsigned short size)
{
	q->hardsect_size = size;
}

EXPORT_SYMBOL(blk_queue_hardsect_size);

/*
 * Returns the minimum that is _not_ zero, unless both are zero.
 */
#define min_not_zero(l, r) (l == 0) ? r : ((r == 0) ? l : min(l, r))

/**
 * blk_queue_stack_limits - inherit underlying queue limits for stacked drivers
 * @t:	the stacking driver (top)
 * @b:  the underlying device (bottom)
 **/
void blk_queue_stack_limits(request_queue_t *t, request_queue_t *b)
{
	/* zero is "infinity" */
665 666
	t->max_sectors = min_not_zero(t->max_sectors,b->max_sectors);
	t->max_hw_sectors = min_not_zero(t->max_hw_sectors,b->max_hw_sectors);
L
Linus Torvalds 已提交
667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724

	t->max_phys_segments = min(t->max_phys_segments,b->max_phys_segments);
	t->max_hw_segments = min(t->max_hw_segments,b->max_hw_segments);
	t->max_segment_size = min(t->max_segment_size,b->max_segment_size);
	t->hardsect_size = max(t->hardsect_size,b->hardsect_size);
}

EXPORT_SYMBOL(blk_queue_stack_limits);

/**
 * blk_queue_segment_boundary - set boundary rules for segment merging
 * @q:  the request queue for the device
 * @mask:  the memory boundary mask
 **/
void blk_queue_segment_boundary(request_queue_t *q, unsigned long mask)
{
	if (mask < PAGE_CACHE_SIZE - 1) {
		mask = PAGE_CACHE_SIZE - 1;
		printk("%s: set to minimum %lx\n", __FUNCTION__, mask);
	}

	q->seg_boundary_mask = mask;
}

EXPORT_SYMBOL(blk_queue_segment_boundary);

/**
 * blk_queue_dma_alignment - set dma length and memory alignment
 * @q:     the request queue for the device
 * @mask:  alignment mask
 *
 * description:
 *    set required memory and length aligment for direct dma transactions.
 *    this is used when buiding direct io requests for the queue.
 *
 **/
void blk_queue_dma_alignment(request_queue_t *q, int mask)
{
	q->dma_alignment = mask;
}

EXPORT_SYMBOL(blk_queue_dma_alignment);

/**
 * blk_queue_find_tag - find a request by its tag and queue
 * @q:	 The request queue for the device
 * @tag: The tag of the request
 *
 * Notes:
 *    Should be used when a device returns a tag and you want to match
 *    it with a request.
 *
 *    no locks need be held.
 **/
struct request *blk_queue_find_tag(request_queue_t *q, int tag)
{
	struct blk_queue_tag *bqt = q->queue_tags;

725
	if (unlikely(bqt == NULL || tag >= bqt->real_max_depth))
L
Linus Torvalds 已提交
726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784
		return NULL;

	return bqt->tag_index[tag];
}

EXPORT_SYMBOL(blk_queue_find_tag);

/**
 * __blk_queue_free_tags - release tag maintenance info
 * @q:  the request queue for the device
 *
 *  Notes:
 *    blk_cleanup_queue() will take care of calling this function, if tagging
 *    has been used. So there's no need to call this directly.
 **/
static void __blk_queue_free_tags(request_queue_t *q)
{
	struct blk_queue_tag *bqt = q->queue_tags;

	if (!bqt)
		return;

	if (atomic_dec_and_test(&bqt->refcnt)) {
		BUG_ON(bqt->busy);
		BUG_ON(!list_empty(&bqt->busy_list));

		kfree(bqt->tag_index);
		bqt->tag_index = NULL;

		kfree(bqt->tag_map);
		bqt->tag_map = NULL;

		kfree(bqt);
	}

	q->queue_tags = NULL;
	q->queue_flags &= ~(1 << QUEUE_FLAG_QUEUED);
}

/**
 * blk_queue_free_tags - release tag maintenance info
 * @q:  the request queue for the device
 *
 *  Notes:
 *	This is used to disabled tagged queuing to a device, yet leave
 *	queue in function.
 **/
void blk_queue_free_tags(request_queue_t *q)
{
	clear_bit(QUEUE_FLAG_QUEUED, &q->queue_flags);
}

EXPORT_SYMBOL(blk_queue_free_tags);

static int
init_tag_map(request_queue_t *q, struct blk_queue_tag *tags, int depth)
{
	struct request **tag_index;
	unsigned long *tag_map;
785
	int nr_ulongs;
L
Linus Torvalds 已提交
786 787 788 789 790 791 792 793 794 795 796

	if (depth > q->nr_requests * 2) {
		depth = q->nr_requests * 2;
		printk(KERN_ERR "%s: adjusted depth to %d\n",
				__FUNCTION__, depth);
	}

	tag_index = kmalloc(depth * sizeof(struct request *), GFP_ATOMIC);
	if (!tag_index)
		goto fail;

797
	nr_ulongs = ALIGN(depth, BITS_PER_LONG) / BITS_PER_LONG;
798
	tag_map = kmalloc(nr_ulongs * sizeof(unsigned long), GFP_ATOMIC);
L
Linus Torvalds 已提交
799 800 801 802
	if (!tag_map)
		goto fail;

	memset(tag_index, 0, depth * sizeof(struct request *));
803
	memset(tag_map, 0, nr_ulongs * sizeof(unsigned long));
804
	tags->real_max_depth = depth;
L
Linus Torvalds 已提交
805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872
	tags->max_depth = depth;
	tags->tag_index = tag_index;
	tags->tag_map = tag_map;

	return 0;
fail:
	kfree(tag_index);
	return -ENOMEM;
}

/**
 * blk_queue_init_tags - initialize the queue tag info
 * @q:  the request queue for the device
 * @depth:  the maximum queue depth supported
 * @tags: the tag to use
 **/
int blk_queue_init_tags(request_queue_t *q, int depth,
			struct blk_queue_tag *tags)
{
	int rc;

	BUG_ON(tags && q->queue_tags && tags != q->queue_tags);

	if (!tags && !q->queue_tags) {
		tags = kmalloc(sizeof(struct blk_queue_tag), GFP_ATOMIC);
		if (!tags)
			goto fail;

		if (init_tag_map(q, tags, depth))
			goto fail;

		INIT_LIST_HEAD(&tags->busy_list);
		tags->busy = 0;
		atomic_set(&tags->refcnt, 1);
	} else if (q->queue_tags) {
		if ((rc = blk_queue_resize_tags(q, depth)))
			return rc;
		set_bit(QUEUE_FLAG_QUEUED, &q->queue_flags);
		return 0;
	} else
		atomic_inc(&tags->refcnt);

	/*
	 * assign it, all done
	 */
	q->queue_tags = tags;
	q->queue_flags |= (1 << QUEUE_FLAG_QUEUED);
	return 0;
fail:
	kfree(tags);
	return -ENOMEM;
}

EXPORT_SYMBOL(blk_queue_init_tags);

/**
 * blk_queue_resize_tags - change the queueing depth
 * @q:  the request queue for the device
 * @new_depth: the new max command queueing depth
 *
 *  Notes:
 *    Must be called with the queue lock held.
 **/
int blk_queue_resize_tags(request_queue_t *q, int new_depth)
{
	struct blk_queue_tag *bqt = q->queue_tags;
	struct request **tag_index;
	unsigned long *tag_map;
873
	int max_depth, nr_ulongs;
L
Linus Torvalds 已提交
874 875 876 877

	if (!bqt)
		return -ENXIO;

878 879 880 881 882 883 884 885 886 887 888
	/*
	 * if we already have large enough real_max_depth.  just
	 * adjust max_depth.  *NOTE* as requests with tag value
	 * between new_depth and real_max_depth can be in-flight, tag
	 * map can not be shrunk blindly here.
	 */
	if (new_depth <= bqt->real_max_depth) {
		bqt->max_depth = new_depth;
		return 0;
	}

L
Linus Torvalds 已提交
889 890 891 892 893
	/*
	 * save the old state info, so we can copy it back
	 */
	tag_index = bqt->tag_index;
	tag_map = bqt->tag_map;
894
	max_depth = bqt->real_max_depth;
L
Linus Torvalds 已提交
895 896 897 898 899

	if (init_tag_map(q, bqt, new_depth))
		return -ENOMEM;

	memcpy(bqt->tag_index, tag_index, max_depth * sizeof(struct request *));
900
	nr_ulongs = ALIGN(max_depth, BITS_PER_LONG) / BITS_PER_LONG;
901
	memcpy(bqt->tag_map, tag_map, nr_ulongs * sizeof(unsigned long));
L
Linus Torvalds 已提交
902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930

	kfree(tag_index);
	kfree(tag_map);
	return 0;
}

EXPORT_SYMBOL(blk_queue_resize_tags);

/**
 * blk_queue_end_tag - end tag operations for a request
 * @q:  the request queue for the device
 * @rq: the request that has completed
 *
 *  Description:
 *    Typically called when end_that_request_first() returns 0, meaning
 *    all transfers have been done for a request. It's important to call
 *    this function before end_that_request_last(), as that will put the
 *    request back on the free list thus corrupting the internal tag list.
 *
 *  Notes:
 *   queue lock must be held.
 **/
void blk_queue_end_tag(request_queue_t *q, struct request *rq)
{
	struct blk_queue_tag *bqt = q->queue_tags;
	int tag = rq->tag;

	BUG_ON(tag == -1);

931
	if (unlikely(tag >= bqt->real_max_depth))
932 933 934 935
		/*
		 * This can happen after tag depth has been reduced.
		 * FIXME: how about a warning or info message here?
		 */
L
Linus Torvalds 已提交
936 937 938
		return;

	if (unlikely(!__test_and_clear_bit(tag, bqt->tag_map))) {
939 940
		printk(KERN_ERR "%s: attempt to clear non-busy tag (%d)\n",
		       __FUNCTION__, tag);
L
Linus Torvalds 已提交
941 942 943 944 945 946 947 948
		return;
	}

	list_del_init(&rq->queuelist);
	rq->flags &= ~REQ_QUEUED;
	rq->tag = -1;

	if (unlikely(bqt->tag_index[tag] == NULL))
949 950
		printk(KERN_ERR "%s: tag %d is missing\n",
		       __FUNCTION__, tag);
L
Linus Torvalds 已提交
951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978

	bqt->tag_index[tag] = NULL;
	bqt->busy--;
}

EXPORT_SYMBOL(blk_queue_end_tag);

/**
 * blk_queue_start_tag - find a free tag and assign it
 * @q:  the request queue for the device
 * @rq:  the block request that needs tagging
 *
 *  Description:
 *    This can either be used as a stand-alone helper, or possibly be
 *    assigned as the queue &prep_rq_fn (in which case &struct request
 *    automagically gets a tag assigned). Note that this function
 *    assumes that any type of request can be queued! if this is not
 *    true for your device, you must check the request type before
 *    calling this function.  The request will also be removed from
 *    the request queue, so it's the drivers responsibility to readd
 *    it if it should need to be restarted for some reason.
 *
 *  Notes:
 *   queue lock must be held.
 **/
int blk_queue_start_tag(request_queue_t *q, struct request *rq)
{
	struct blk_queue_tag *bqt = q->queue_tags;
979
	int tag;
L
Linus Torvalds 已提交
980 981 982

	if (unlikely((rq->flags & REQ_QUEUED))) {
		printk(KERN_ERR 
983 984 985
		       "%s: request %p for device [%s] already tagged %d",
		       __FUNCTION__, rq,
		       rq->rq_disk ? rq->rq_disk->disk_name : "?", rq->tag);
L
Linus Torvalds 已提交
986 987 988
		BUG();
	}

989 990 991
	tag = find_first_zero_bit(bqt->tag_map, bqt->max_depth);
	if (tag >= bqt->max_depth)
		return 1;
L
Linus Torvalds 已提交
992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027

	__set_bit(tag, bqt->tag_map);

	rq->flags |= REQ_QUEUED;
	rq->tag = tag;
	bqt->tag_index[tag] = rq;
	blkdev_dequeue_request(rq);
	list_add(&rq->queuelist, &bqt->busy_list);
	bqt->busy++;
	return 0;
}

EXPORT_SYMBOL(blk_queue_start_tag);

/**
 * blk_queue_invalidate_tags - invalidate all pending tags
 * @q:  the request queue for the device
 *
 *  Description:
 *   Hardware conditions may dictate a need to stop all pending requests.
 *   In this case, we will safely clear the block side of the tag queue and
 *   readd all requests to the request queue in the right order.
 *
 *  Notes:
 *   queue lock must be held.
 **/
void blk_queue_invalidate_tags(request_queue_t *q)
{
	struct blk_queue_tag *bqt = q->queue_tags;
	struct list_head *tmp, *n;
	struct request *rq;

	list_for_each_safe(tmp, n, &bqt->busy_list) {
		rq = list_entry_rq(tmp);

		if (rq->tag == -1) {
1028 1029
			printk(KERN_ERR
			       "%s: bad tag found on list\n", __FUNCTION__);
L
Linus Torvalds 已提交
1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044
			list_del_init(&rq->queuelist);
			rq->flags &= ~REQ_QUEUED;
		} else
			blk_queue_end_tag(q, rq);

		rq->flags &= ~REQ_STARTED;
		__elv_add_request(q, rq, ELEVATOR_INSERT_BACK, 0);
	}
}

EXPORT_SYMBOL(blk_queue_invalidate_tags);

static char *rq_flags[] = {
	"REQ_RW",
	"REQ_FAILFAST",
1045
	"REQ_SORTED",
L
Linus Torvalds 已提交
1046 1047 1048 1049 1050 1051 1052
	"REQ_SOFTBARRIER",
	"REQ_HARDBARRIER",
	"REQ_CMD",
	"REQ_NOMERGE",
	"REQ_STARTED",
	"REQ_DONTPREP",
	"REQ_QUEUED",
T
Tejun Heo 已提交
1053
	"REQ_ELVPRIV",
L
Linus Torvalds 已提交
1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158
	"REQ_PC",
	"REQ_BLOCK_PC",
	"REQ_SENSE",
	"REQ_FAILED",
	"REQ_QUIET",
	"REQ_SPECIAL",
	"REQ_DRIVE_CMD",
	"REQ_DRIVE_TASK",
	"REQ_DRIVE_TASKFILE",
	"REQ_PREEMPT",
	"REQ_PM_SUSPEND",
	"REQ_PM_RESUME",
	"REQ_PM_SHUTDOWN",
};

void blk_dump_rq_flags(struct request *rq, char *msg)
{
	int bit;

	printk("%s: dev %s: flags = ", msg,
		rq->rq_disk ? rq->rq_disk->disk_name : "?");
	bit = 0;
	do {
		if (rq->flags & (1 << bit))
			printk("%s ", rq_flags[bit]);
		bit++;
	} while (bit < __REQ_NR_BITS);

	printk("\nsector %llu, nr/cnr %lu/%u\n", (unsigned long long)rq->sector,
						       rq->nr_sectors,
						       rq->current_nr_sectors);
	printk("bio %p, biotail %p, buffer %p, data %p, len %u\n", rq->bio, rq->biotail, rq->buffer, rq->data, rq->data_len);

	if (rq->flags & (REQ_BLOCK_PC | REQ_PC)) {
		printk("cdb: ");
		for (bit = 0; bit < sizeof(rq->cmd); bit++)
			printk("%02x ", rq->cmd[bit]);
		printk("\n");
	}
}

EXPORT_SYMBOL(blk_dump_rq_flags);

void blk_recount_segments(request_queue_t *q, struct bio *bio)
{
	struct bio_vec *bv, *bvprv = NULL;
	int i, nr_phys_segs, nr_hw_segs, seg_size, hw_seg_size, cluster;
	int high, highprv = 1;

	if (unlikely(!bio->bi_io_vec))
		return;

	cluster = q->queue_flags & (1 << QUEUE_FLAG_CLUSTER);
	hw_seg_size = seg_size = nr_phys_segs = nr_hw_segs = 0;
	bio_for_each_segment(bv, bio, i) {
		/*
		 * the trick here is making sure that a high page is never
		 * considered part of another segment, since that might
		 * change with the bounce page.
		 */
		high = page_to_pfn(bv->bv_page) >= q->bounce_pfn;
		if (high || highprv)
			goto new_hw_segment;
		if (cluster) {
			if (seg_size + bv->bv_len > q->max_segment_size)
				goto new_segment;
			if (!BIOVEC_PHYS_MERGEABLE(bvprv, bv))
				goto new_segment;
			if (!BIOVEC_SEG_BOUNDARY(q, bvprv, bv))
				goto new_segment;
			if (BIOVEC_VIRT_OVERSIZE(hw_seg_size + bv->bv_len))
				goto new_hw_segment;

			seg_size += bv->bv_len;
			hw_seg_size += bv->bv_len;
			bvprv = bv;
			continue;
		}
new_segment:
		if (BIOVEC_VIRT_MERGEABLE(bvprv, bv) &&
		    !BIOVEC_VIRT_OVERSIZE(hw_seg_size + bv->bv_len)) {
			hw_seg_size += bv->bv_len;
		} else {
new_hw_segment:
			if (hw_seg_size > bio->bi_hw_front_size)
				bio->bi_hw_front_size = hw_seg_size;
			hw_seg_size = BIOVEC_VIRT_START_SIZE(bv) + bv->bv_len;
			nr_hw_segs++;
		}

		nr_phys_segs++;
		bvprv = bv;
		seg_size = bv->bv_len;
		highprv = high;
	}
	if (hw_seg_size > bio->bi_hw_back_size)
		bio->bi_hw_back_size = hw_seg_size;
	if (nr_hw_segs == 1 && hw_seg_size > bio->bi_hw_front_size)
		bio->bi_hw_front_size = hw_seg_size;
	bio->bi_phys_segments = nr_phys_segs;
	bio->bi_hw_segments = nr_hw_segs;
	bio->bi_flags |= (1 << BIO_SEG_VALID);
}


1159
static int blk_phys_contig_segment(request_queue_t *q, struct bio *bio,
L
Linus Torvalds 已提交
1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179
				   struct bio *nxt)
{
	if (!(q->queue_flags & (1 << QUEUE_FLAG_CLUSTER)))
		return 0;

	if (!BIOVEC_PHYS_MERGEABLE(__BVEC_END(bio), __BVEC_START(nxt)))
		return 0;
	if (bio->bi_size + nxt->bi_size > q->max_segment_size)
		return 0;

	/*
	 * bio and nxt are contigous in memory, check if the queue allows
	 * these two to be merged into one
	 */
	if (BIO_SEG_BOUNDARY(q, bio, nxt))
		return 1;

	return 0;
}

1180
static int blk_hw_contig_segment(request_queue_t *q, struct bio *bio,
L
Linus Torvalds 已提交
1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 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 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300
				 struct bio *nxt)
{
	if (unlikely(!bio_flagged(bio, BIO_SEG_VALID)))
		blk_recount_segments(q, bio);
	if (unlikely(!bio_flagged(nxt, BIO_SEG_VALID)))
		blk_recount_segments(q, nxt);
	if (!BIOVEC_VIRT_MERGEABLE(__BVEC_END(bio), __BVEC_START(nxt)) ||
	    BIOVEC_VIRT_OVERSIZE(bio->bi_hw_front_size + bio->bi_hw_back_size))
		return 0;
	if (bio->bi_size + nxt->bi_size > q->max_segment_size)
		return 0;

	return 1;
}

/*
 * map a request to scatterlist, return number of sg entries setup. Caller
 * must make sure sg can hold rq->nr_phys_segments entries
 */
int blk_rq_map_sg(request_queue_t *q, struct request *rq, struct scatterlist *sg)
{
	struct bio_vec *bvec, *bvprv;
	struct bio *bio;
	int nsegs, i, cluster;

	nsegs = 0;
	cluster = q->queue_flags & (1 << QUEUE_FLAG_CLUSTER);

	/*
	 * for each bio in rq
	 */
	bvprv = NULL;
	rq_for_each_bio(bio, rq) {
		/*
		 * for each segment in bio
		 */
		bio_for_each_segment(bvec, bio, i) {
			int nbytes = bvec->bv_len;

			if (bvprv && cluster) {
				if (sg[nsegs - 1].length + nbytes > q->max_segment_size)
					goto new_segment;

				if (!BIOVEC_PHYS_MERGEABLE(bvprv, bvec))
					goto new_segment;
				if (!BIOVEC_SEG_BOUNDARY(q, bvprv, bvec))
					goto new_segment;

				sg[nsegs - 1].length += nbytes;
			} else {
new_segment:
				memset(&sg[nsegs],0,sizeof(struct scatterlist));
				sg[nsegs].page = bvec->bv_page;
				sg[nsegs].length = nbytes;
				sg[nsegs].offset = bvec->bv_offset;

				nsegs++;
			}
			bvprv = bvec;
		} /* segments in bio */
	} /* bios in rq */

	return nsegs;
}

EXPORT_SYMBOL(blk_rq_map_sg);

/*
 * the standard queue merge functions, can be overridden with device
 * specific ones if so desired
 */

static inline int ll_new_mergeable(request_queue_t *q,
				   struct request *req,
				   struct bio *bio)
{
	int nr_phys_segs = bio_phys_segments(q, bio);

	if (req->nr_phys_segments + nr_phys_segs > q->max_phys_segments) {
		req->flags |= REQ_NOMERGE;
		if (req == q->last_merge)
			q->last_merge = NULL;
		return 0;
	}

	/*
	 * A hw segment is just getting larger, bump just the phys
	 * counter.
	 */
	req->nr_phys_segments += nr_phys_segs;
	return 1;
}

static inline int ll_new_hw_segment(request_queue_t *q,
				    struct request *req,
				    struct bio *bio)
{
	int nr_hw_segs = bio_hw_segments(q, bio);
	int nr_phys_segs = bio_phys_segments(q, bio);

	if (req->nr_hw_segments + nr_hw_segs > q->max_hw_segments
	    || req->nr_phys_segments + nr_phys_segs > q->max_phys_segments) {
		req->flags |= REQ_NOMERGE;
		if (req == q->last_merge)
			q->last_merge = NULL;
		return 0;
	}

	/*
	 * This will form the start of a new hw segment.  Bump both
	 * counters.
	 */
	req->nr_hw_segments += nr_hw_segs;
	req->nr_phys_segments += nr_phys_segs;
	return 1;
}

static int ll_back_merge_fn(request_queue_t *q, struct request *req, 
			    struct bio *bio)
{
1301
	unsigned short max_sectors;
L
Linus Torvalds 已提交
1302 1303
	int len;

1304 1305 1306 1307 1308 1309
	if (unlikely(blk_pc_request(req)))
		max_sectors = q->max_hw_sectors;
	else
		max_sectors = q->max_sectors;

	if (req->nr_sectors + bio_sectors(bio) > max_sectors) {
L
Linus Torvalds 已提交
1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338
		req->flags |= REQ_NOMERGE;
		if (req == q->last_merge)
			q->last_merge = NULL;
		return 0;
	}
	if (unlikely(!bio_flagged(req->biotail, BIO_SEG_VALID)))
		blk_recount_segments(q, req->biotail);
	if (unlikely(!bio_flagged(bio, BIO_SEG_VALID)))
		blk_recount_segments(q, bio);
	len = req->biotail->bi_hw_back_size + bio->bi_hw_front_size;
	if (BIOVEC_VIRT_MERGEABLE(__BVEC_END(req->biotail), __BVEC_START(bio)) &&
	    !BIOVEC_VIRT_OVERSIZE(len)) {
		int mergeable =  ll_new_mergeable(q, req, bio);

		if (mergeable) {
			if (req->nr_hw_segments == 1)
				req->bio->bi_hw_front_size = len;
			if (bio->bi_hw_segments == 1)
				bio->bi_hw_back_size = len;
		}
		return mergeable;
	}

	return ll_new_hw_segment(q, req, bio);
}

static int ll_front_merge_fn(request_queue_t *q, struct request *req, 
			     struct bio *bio)
{
1339
	unsigned short max_sectors;
L
Linus Torvalds 已提交
1340 1341
	int len;

1342 1343 1344 1345 1346 1347 1348
	if (unlikely(blk_pc_request(req)))
		max_sectors = q->max_hw_sectors;
	else
		max_sectors = q->max_sectors;


	if (req->nr_sectors + bio_sectors(bio) > max_sectors) {
L
Linus Torvalds 已提交
1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377
		req->flags |= REQ_NOMERGE;
		if (req == q->last_merge)
			q->last_merge = NULL;
		return 0;
	}
	len = bio->bi_hw_back_size + req->bio->bi_hw_front_size;
	if (unlikely(!bio_flagged(bio, BIO_SEG_VALID)))
		blk_recount_segments(q, bio);
	if (unlikely(!bio_flagged(req->bio, BIO_SEG_VALID)))
		blk_recount_segments(q, req->bio);
	if (BIOVEC_VIRT_MERGEABLE(__BVEC_END(bio), __BVEC_START(req->bio)) &&
	    !BIOVEC_VIRT_OVERSIZE(len)) {
		int mergeable =  ll_new_mergeable(q, req, bio);

		if (mergeable) {
			if (bio->bi_hw_segments == 1)
				bio->bi_hw_front_size = len;
			if (req->nr_hw_segments == 1)
				req->biotail->bi_hw_back_size = len;
		}
		return mergeable;
	}

	return ll_new_hw_segment(q, req, bio);
}

static int ll_merge_requests_fn(request_queue_t *q, struct request *req,
				struct request *next)
{
1378 1379
	int total_phys_segments;
	int total_hw_segments;
L
Linus Torvalds 已提交
1380 1381 1382 1383 1384 1385 1386 1387 1388

	/*
	 * First check if the either of the requests are re-queued
	 * requests.  Can't merge them if they are.
	 */
	if (req->special || next->special)
		return 0;

	/*
1389
	 * Will it become too large?
L
Linus Torvalds 已提交
1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469
	 */
	if ((req->nr_sectors + next->nr_sectors) > q->max_sectors)
		return 0;

	total_phys_segments = req->nr_phys_segments + next->nr_phys_segments;
	if (blk_phys_contig_segment(q, req->biotail, next->bio))
		total_phys_segments--;

	if (total_phys_segments > q->max_phys_segments)
		return 0;

	total_hw_segments = req->nr_hw_segments + next->nr_hw_segments;
	if (blk_hw_contig_segment(q, req->biotail, next->bio)) {
		int len = req->biotail->bi_hw_back_size + next->bio->bi_hw_front_size;
		/*
		 * propagate the combined length to the end of the requests
		 */
		if (req->nr_hw_segments == 1)
			req->bio->bi_hw_front_size = len;
		if (next->nr_hw_segments == 1)
			next->biotail->bi_hw_back_size = len;
		total_hw_segments--;
	}

	if (total_hw_segments > q->max_hw_segments)
		return 0;

	/* Merge is OK... */
	req->nr_phys_segments = total_phys_segments;
	req->nr_hw_segments = total_hw_segments;
	return 1;
}

/*
 * "plug" the device if there are no outstanding requests: this will
 * force the transfer to start only after we have put all the requests
 * on the list.
 *
 * This is called with interrupts off and no requests on the queue and
 * with the queue lock held.
 */
void blk_plug_device(request_queue_t *q)
{
	WARN_ON(!irqs_disabled());

	/*
	 * don't plug a stopped queue, it must be paired with blk_start_queue()
	 * which will restart the queueing
	 */
	if (test_bit(QUEUE_FLAG_STOPPED, &q->queue_flags))
		return;

	if (!test_and_set_bit(QUEUE_FLAG_PLUGGED, &q->queue_flags))
		mod_timer(&q->unplug_timer, jiffies + q->unplug_delay);
}

EXPORT_SYMBOL(blk_plug_device);

/*
 * remove the queue from the plugged list, if present. called with
 * queue lock held and interrupts disabled.
 */
int blk_remove_plug(request_queue_t *q)
{
	WARN_ON(!irqs_disabled());

	if (!test_and_clear_bit(QUEUE_FLAG_PLUGGED, &q->queue_flags))
		return 0;

	del_timer(&q->unplug_timer);
	return 1;
}

EXPORT_SYMBOL(blk_remove_plug);

/*
 * remove the plug and let it rip..
 */
void __generic_unplug_device(request_queue_t *q)
{
N
Nick Piggin 已提交
1470
	if (unlikely(test_bit(QUEUE_FLAG_STOPPED, &q->queue_flags)))
L
Linus Torvalds 已提交
1471 1472 1473 1474 1475
		return;

	if (!blk_remove_plug(q))
		return;

1476
	q->request_fn(q);
L
Linus Torvalds 已提交
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 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604
}
EXPORT_SYMBOL(__generic_unplug_device);

/**
 * generic_unplug_device - fire a request queue
 * @q:    The &request_queue_t in question
 *
 * Description:
 *   Linux uses plugging to build bigger requests queues before letting
 *   the device have at them. If a queue is plugged, the I/O scheduler
 *   is still adding and merging requests on the queue. Once the queue
 *   gets unplugged, the request_fn defined for the queue is invoked and
 *   transfers started.
 **/
void generic_unplug_device(request_queue_t *q)
{
	spin_lock_irq(q->queue_lock);
	__generic_unplug_device(q);
	spin_unlock_irq(q->queue_lock);
}
EXPORT_SYMBOL(generic_unplug_device);

static void blk_backing_dev_unplug(struct backing_dev_info *bdi,
				   struct page *page)
{
	request_queue_t *q = bdi->unplug_io_data;

	/*
	 * devices don't necessarily have an ->unplug_fn defined
	 */
	if (q->unplug_fn)
		q->unplug_fn(q);
}

static void blk_unplug_work(void *data)
{
	request_queue_t *q = data;

	q->unplug_fn(q);
}

static void blk_unplug_timeout(unsigned long data)
{
	request_queue_t *q = (request_queue_t *)data;

	kblockd_schedule_work(&q->unplug_work);
}

/**
 * blk_start_queue - restart a previously stopped queue
 * @q:    The &request_queue_t in question
 *
 * Description:
 *   blk_start_queue() will clear the stop flag on the queue, and call
 *   the request_fn for the queue if it was in a stopped state when
 *   entered. Also see blk_stop_queue(). Queue lock must be held.
 **/
void blk_start_queue(request_queue_t *q)
{
	clear_bit(QUEUE_FLAG_STOPPED, &q->queue_flags);

	/*
	 * one level of recursion is ok and is much faster than kicking
	 * the unplug handling
	 */
	if (!test_and_set_bit(QUEUE_FLAG_REENTER, &q->queue_flags)) {
		q->request_fn(q);
		clear_bit(QUEUE_FLAG_REENTER, &q->queue_flags);
	} else {
		blk_plug_device(q);
		kblockd_schedule_work(&q->unplug_work);
	}
}

EXPORT_SYMBOL(blk_start_queue);

/**
 * blk_stop_queue - stop a queue
 * @q:    The &request_queue_t in question
 *
 * Description:
 *   The Linux block layer assumes that a block driver will consume all
 *   entries on the request queue when the request_fn strategy is called.
 *   Often this will not happen, because of hardware limitations (queue
 *   depth settings). If a device driver gets a 'queue full' response,
 *   or if it simply chooses not to queue more I/O at one point, it can
 *   call this function to prevent the request_fn from being called until
 *   the driver has signalled it's ready to go again. This happens by calling
 *   blk_start_queue() to restart queue operations. Queue lock must be held.
 **/
void blk_stop_queue(request_queue_t *q)
{
	blk_remove_plug(q);
	set_bit(QUEUE_FLAG_STOPPED, &q->queue_flags);
}
EXPORT_SYMBOL(blk_stop_queue);

/**
 * blk_sync_queue - cancel any pending callbacks on a queue
 * @q: the queue
 *
 * Description:
 *     The block layer may perform asynchronous callback activity
 *     on a queue, such as calling the unplug function after a timeout.
 *     A block device may call blk_sync_queue to ensure that any
 *     such activity is cancelled, thus allowing it to release resources
 *     the the callbacks might use. The caller must already have made sure
 *     that its ->make_request_fn will not re-add plugging prior to calling
 *     this function.
 *
 */
void blk_sync_queue(struct request_queue *q)
{
	del_timer_sync(&q->unplug_timer);
	kblockd_flush();
}
EXPORT_SYMBOL(blk_sync_queue);

/**
 * blk_run_queue - run a single device queue
 * @q:	The queue to run
 */
void blk_run_queue(struct request_queue *q)
{
	unsigned long flags;

	spin_lock_irqsave(q->queue_lock, flags);
	blk_remove_plug(q);
1605 1606
	if (!elv_queue_empty(q))
		q->request_fn(q);
L
Linus Torvalds 已提交
1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656
	spin_unlock_irqrestore(q->queue_lock, flags);
}
EXPORT_SYMBOL(blk_run_queue);

/**
 * blk_cleanup_queue: - release a &request_queue_t when it is no longer needed
 * @q:    the request queue to be released
 *
 * Description:
 *     blk_cleanup_queue is the pair to blk_init_queue() or
 *     blk_queue_make_request().  It should be called when a request queue is
 *     being released; typically when a block device is being de-registered.
 *     Currently, its primary task it to free all the &struct request
 *     structures that were allocated to the queue and the queue itself.
 *
 * Caveat:
 *     Hopefully the low level driver will have finished any
 *     outstanding requests first...
 **/
void blk_cleanup_queue(request_queue_t * q)
{
	struct request_list *rl = &q->rq;

	if (!atomic_dec_and_test(&q->refcnt))
		return;

	if (q->elevator)
		elevator_exit(q->elevator);

	blk_sync_queue(q);

	if (rl->rq_pool)
		mempool_destroy(rl->rq_pool);

	if (q->queue_tags)
		__blk_queue_free_tags(q);

	blk_queue_ordered(q, QUEUE_ORDERED_NONE);

	kmem_cache_free(requestq_cachep, q);
}

EXPORT_SYMBOL(blk_cleanup_queue);

static int blk_init_free_list(request_queue_t *q)
{
	struct request_list *rl = &q->rq;

	rl->count[READ] = rl->count[WRITE] = 0;
	rl->starved[READ] = rl->starved[WRITE] = 0;
T
Tejun Heo 已提交
1657
	rl->elvpriv = 0;
L
Linus Torvalds 已提交
1658 1659 1660
	init_waitqueue_head(&rl->wait[READ]);
	init_waitqueue_head(&rl->wait[WRITE]);

1661 1662
	rl->rq_pool = mempool_create_node(BLKDEV_MIN_RQ, mempool_alloc_slab,
				mempool_free_slab, request_cachep, q->node);
L
Linus Torvalds 已提交
1663 1664 1665 1666 1667 1668 1669 1670 1671

	if (!rl->rq_pool)
		return -ENOMEM;

	return 0;
}

static int __make_request(request_queue_t *, struct bio *);

A
Al Viro 已提交
1672
request_queue_t *blk_alloc_queue(gfp_t gfp_mask)
L
Linus Torvalds 已提交
1673
{
1674 1675 1676
	return blk_alloc_queue_node(gfp_mask, -1);
}
EXPORT_SYMBOL(blk_alloc_queue);
L
Linus Torvalds 已提交
1677

A
Al Viro 已提交
1678
request_queue_t *blk_alloc_queue_node(gfp_t gfp_mask, int node_id)
1679 1680 1681 1682
{
	request_queue_t *q;

	q = kmem_cache_alloc_node(requestq_cachep, gfp_mask, node_id);
L
Linus Torvalds 已提交
1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694
	if (!q)
		return NULL;

	memset(q, 0, sizeof(*q));
	init_timer(&q->unplug_timer);
	atomic_set(&q->refcnt, 1);

	q->backing_dev_info.unplug_io_fn = blk_backing_dev_unplug;
	q->backing_dev_info.unplug_io_data = q;

	return q;
}
1695
EXPORT_SYMBOL(blk_alloc_queue_node);
L
Linus Torvalds 已提交
1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727

/**
 * blk_init_queue  - prepare a request queue for use with a block device
 * @rfn:  The function to be called to process requests that have been
 *        placed on the queue.
 * @lock: Request queue spin lock
 *
 * Description:
 *    If a block device wishes to use the standard request handling procedures,
 *    which sorts requests and coalesces adjacent requests, then it must
 *    call blk_init_queue().  The function @rfn will be called when there
 *    are requests on the queue that need to be processed.  If the device
 *    supports plugging, then @rfn may not be called immediately when requests
 *    are available on the queue, but may be called at some time later instead.
 *    Plugged queues are generally unplugged when a buffer belonging to one
 *    of the requests on the queue is needed, or due to memory pressure.
 *
 *    @rfn is not required, or even expected, to remove all requests off the
 *    queue, but only as many as it can handle at a time.  If it does leave
 *    requests on the queue, it is responsible for arranging that the requests
 *    get dealt with eventually.
 *
 *    The queue spin lock must be held while manipulating the requests on the
 *    request queue.
 *
 *    Function returns a pointer to the initialized request queue, or NULL if
 *    it didn't succeed.
 *
 * Note:
 *    blk_init_queue() must be paired with a blk_cleanup_queue() call
 *    when the block device is deactivated (such as at module unload).
 **/
1728

L
Linus Torvalds 已提交
1729 1730
request_queue_t *blk_init_queue(request_fn_proc *rfn, spinlock_t *lock)
{
1731 1732 1733 1734 1735 1736 1737 1738
	return blk_init_queue_node(rfn, lock, -1);
}
EXPORT_SYMBOL(blk_init_queue);

request_queue_t *
blk_init_queue_node(request_fn_proc *rfn, spinlock_t *lock, int node_id)
{
	request_queue_t *q = blk_alloc_queue_node(GFP_KERNEL, node_id);
L
Linus Torvalds 已提交
1739 1740 1741 1742

	if (!q)
		return NULL;

1743
	q->node = node_id;
L
Linus Torvalds 已提交
1744 1745 1746
	if (blk_init_free_list(q))
		goto out_init;

已提交
1747 1748 1749 1750 1751 1752 1753 1754 1755
	/*
	 * if caller didn't supply a lock, they get per-queue locking with
	 * our embedded lock
	 */
	if (!lock) {
		spin_lock_init(&q->__queue_lock);
		lock = &q->__queue_lock;
	}

L
Linus Torvalds 已提交
1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785
	q->request_fn		= rfn;
	q->back_merge_fn       	= ll_back_merge_fn;
	q->front_merge_fn      	= ll_front_merge_fn;
	q->merge_requests_fn	= ll_merge_requests_fn;
	q->prep_rq_fn		= NULL;
	q->unplug_fn		= generic_unplug_device;
	q->queue_flags		= (1 << QUEUE_FLAG_CLUSTER);
	q->queue_lock		= lock;

	blk_queue_segment_boundary(q, 0xffffffff);

	blk_queue_make_request(q, __make_request);
	blk_queue_max_segment_size(q, MAX_SEGMENT_SIZE);

	blk_queue_max_hw_segments(q, MAX_HW_SEGMENTS);
	blk_queue_max_phys_segments(q, MAX_PHYS_SEGMENTS);

	/*
	 * all done
	 */
	if (!elevator_init(q, NULL)) {
		blk_queue_congestion_threshold(q);
		return q;
	}

	blk_cleanup_queue(q);
out_init:
	kmem_cache_free(requestq_cachep, q);
	return NULL;
}
1786
EXPORT_SYMBOL(blk_init_queue_node);
L
Linus Torvalds 已提交
1787 1788 1789

int blk_get_queue(request_queue_t *q)
{
N
Nick Piggin 已提交
1790
	if (likely(!test_bit(QUEUE_FLAG_DEAD, &q->queue_flags))) {
L
Linus Torvalds 已提交
1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801
		atomic_inc(&q->refcnt);
		return 0;
	}

	return 1;
}

EXPORT_SYMBOL(blk_get_queue);

static inline void blk_free_request(request_queue_t *q, struct request *rq)
{
T
Tejun Heo 已提交
1802 1803
	if (rq->flags & REQ_ELVPRIV)
		elv_put_request(q, rq);
L
Linus Torvalds 已提交
1804 1805 1806
	mempool_free(rq, q->rq.rq_pool);
}

1807
static inline struct request *
T
Tejun Heo 已提交
1808
blk_alloc_request(request_queue_t *q, int rw, struct bio *bio,
1809
		  int priv, gfp_t gfp_mask)
L
Linus Torvalds 已提交
1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821
{
	struct request *rq = mempool_alloc(q->rq.rq_pool, gfp_mask);

	if (!rq)
		return NULL;

	/*
	 * first three bits are identical in rq->flags and bio->bi_rw,
	 * see bio.h and blkdev.h
	 */
	rq->flags = rw;

T
Tejun Heo 已提交
1822 1823 1824 1825 1826 1827 1828
	if (priv) {
		if (unlikely(elv_set_request(q, rq, bio, gfp_mask))) {
			mempool_free(rq, q->rq.rq_pool);
			return NULL;
		}
		rq->flags |= REQ_ELVPRIV;
	}
L
Linus Torvalds 已提交
1829

T
Tejun Heo 已提交
1830
	return rq;
L
Linus Torvalds 已提交
1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857
}

/*
 * ioc_batching returns true if the ioc is a valid batching request and
 * should be given priority access to a request.
 */
static inline int ioc_batching(request_queue_t *q, struct io_context *ioc)
{
	if (!ioc)
		return 0;

	/*
	 * Make sure the process is able to allocate at least 1 request
	 * even if the batch times out, otherwise we could theoretically
	 * lose wakeups.
	 */
	return ioc->nr_batch_requests == q->nr_batching ||
		(ioc->nr_batch_requests > 0
		&& time_before(jiffies, ioc->last_waited + BLK_BATCH_TIME));
}

/*
 * ioc_set_batching sets ioc to be a new "batcher" if it is not one. This
 * will cause the process to be a "batcher" on all queues in the system. This
 * is the behaviour we want though - once it gets a wakeup it should be given
 * a nice run.
 */
1858
static void ioc_set_batching(request_queue_t *q, struct io_context *ioc)
L
Linus Torvalds 已提交
1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885
{
	if (!ioc || ioc_batching(q, ioc))
		return;

	ioc->nr_batch_requests = q->nr_batching;
	ioc->last_waited = jiffies;
}

static void __freed_request(request_queue_t *q, int rw)
{
	struct request_list *rl = &q->rq;

	if (rl->count[rw] < queue_congestion_off_threshold(q))
		clear_queue_congested(q, rw);

	if (rl->count[rw] + 1 <= q->nr_requests) {
		if (waitqueue_active(&rl->wait[rw]))
			wake_up(&rl->wait[rw]);

		blk_clear_queue_full(q, rw);
	}
}

/*
 * A request has just been released.  Account for it, update the full and
 * congestion status, wake up any waiters.   Called under q->queue_lock.
 */
T
Tejun Heo 已提交
1886
static void freed_request(request_queue_t *q, int rw, int priv)
L
Linus Torvalds 已提交
1887 1888 1889 1890
{
	struct request_list *rl = &q->rq;

	rl->count[rw]--;
T
Tejun Heo 已提交
1891 1892
	if (priv)
		rl->elvpriv--;
L
Linus Torvalds 已提交
1893 1894 1895 1896 1897 1898 1899 1900 1901

	__freed_request(q, rw);

	if (unlikely(rl->starved[rw ^ 1]))
		__freed_request(q, rw ^ 1);
}

#define blkdev_free_rq(list) list_entry((list)->next, struct request, queuelist)
/*
N
Nick Piggin 已提交
1902 1903 1904
 * Get a free request, queue_lock must be held.
 * Returns NULL on failure, with queue_lock held.
 * Returns !NULL on success, with queue_lock *not held*.
L
Linus Torvalds 已提交
1905
 */
1906
static struct request *get_request(request_queue_t *q, int rw, struct bio *bio,
A
Al Viro 已提交
1907
				   gfp_t gfp_mask)
L
Linus Torvalds 已提交
1908 1909 1910
{
	struct request *rq = NULL;
	struct request_list *rl = &q->rq;
1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940
	struct io_context *ioc = NULL;
	int may_queue, priv;

	may_queue = elv_may_queue(q, rw, bio);
	if (may_queue == ELV_MQUEUE_NO)
		goto rq_starved;

	if (rl->count[rw]+1 >= queue_congestion_on_threshold(q)) {
		if (rl->count[rw]+1 >= q->nr_requests) {
			ioc = current_io_context(GFP_ATOMIC);
			/*
			 * The queue will fill after this allocation, so set
			 * it as full, and mark this process as "batching".
			 * This process will be allowed to complete a batch of
			 * requests, others will be blocked.
			 */
			if (!blk_queue_full(q, rw)) {
				ioc_set_batching(q, ioc);
				blk_set_queue_full(q, rw);
			} else {
				if (may_queue != ELV_MQUEUE_MUST
						&& !ioc_batching(q, ioc)) {
					/*
					 * The queue is full and the allocating
					 * process is not a "batcher", and not
					 * exempted by the IO scheduler
					 */
					goto out;
				}
			}
L
Linus Torvalds 已提交
1941
		}
1942
		set_queue_congested(q, rw);
L
Linus Torvalds 已提交
1943 1944
	}

1945 1946 1947 1948 1949
	/*
	 * Only allow batching queuers to allocate up to 50% over the defined
	 * limit of requests, otherwise we could have thousands of requests
	 * allocated with any setting of ->nr_requests
	 */
H
Hugh Dickins 已提交
1950
	if (rl->count[rw] >= (3 * q->nr_requests / 2))
1951
		goto out;
H
Hugh Dickins 已提交
1952

L
Linus Torvalds 已提交
1953 1954
	rl->count[rw]++;
	rl->starved[rw] = 0;
T
Tejun Heo 已提交
1955

J
Jens Axboe 已提交
1956
	priv = !test_bit(QUEUE_FLAG_ELVSWITCH, &q->queue_flags);
T
Tejun Heo 已提交
1957 1958 1959
	if (priv)
		rl->elvpriv++;

L
Linus Torvalds 已提交
1960 1961
	spin_unlock_irq(q->queue_lock);

T
Tejun Heo 已提交
1962
	rq = blk_alloc_request(q, rw, bio, priv, gfp_mask);
1963
	if (unlikely(!rq)) {
L
Linus Torvalds 已提交
1964 1965 1966 1967 1968 1969 1970 1971
		/*
		 * Allocation failed presumably due to memory. Undo anything
		 * we might have messed up.
		 *
		 * Allocating task should really be put onto the front of the
		 * wait queue, but this is pretty rare.
		 */
		spin_lock_irq(q->queue_lock);
T
Tejun Heo 已提交
1972
		freed_request(q, rw, priv);
L
Linus Torvalds 已提交
1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987

		/*
		 * in the very unlikely event that allocation failed and no
		 * requests for this direction was pending, mark us starved
		 * so that freeing of a request in the other direction will
		 * notice us. another possible fix would be to split the
		 * rq mempool into READ and WRITE
		 */
rq_starved:
		if (unlikely(rl->count[rw] == 0))
			rl->starved[rw] = 1;

		goto out;
	}

1988 1989 1990 1991 1992 1993
	/*
	 * ioc may be NULL here, and ioc_batching will be false. That's
	 * OK, if the queue is under the request limit then requests need
	 * not count toward the nr_batch_requests limit. There will always
	 * be some limit enforced by BLK_BATCH_TIME.
	 */
L
Linus Torvalds 已提交
1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
	if (ioc_batching(q, ioc))
		ioc->nr_batch_requests--;
	
	rq_init(q, rq);
	rq->rl = rl;
out:
	return rq;
}

/*
 * No available requests for this queue, unplug the device and wait for some
 * requests to become available.
N
Nick Piggin 已提交
2006 2007
 *
 * Called with q->queue_lock held, and returns with it unlocked.
L
Linus Torvalds 已提交
2008
 */
2009 2010
static struct request *get_request_wait(request_queue_t *q, int rw,
					struct bio *bio)
L
Linus Torvalds 已提交
2011 2012 2013
{
	struct request *rq;

2014 2015 2016
	rq = get_request(q, rw, bio, GFP_NOIO);
	while (!rq) {
		DEFINE_WAIT(wait);
L
Linus Torvalds 已提交
2017 2018 2019 2020 2021
		struct request_list *rl = &q->rq;

		prepare_to_wait_exclusive(&rl->wait[rw], &wait,
				TASK_UNINTERRUPTIBLE);

2022
		rq = get_request(q, rw, bio, GFP_NOIO);
L
Linus Torvalds 已提交
2023 2024 2025 2026

		if (!rq) {
			struct io_context *ioc;

N
Nick Piggin 已提交
2027 2028
			__generic_unplug_device(q);
			spin_unlock_irq(q->queue_lock);
L
Linus Torvalds 已提交
2029 2030 2031 2032 2033 2034 2035 2036
			io_schedule();

			/*
			 * After sleeping, we become a "batching" process and
			 * will be able to allocate at least one request, and
			 * up to a big batch of them for a small period time.
			 * See ioc_batching, ioc_set_batching
			 */
N
Nick Piggin 已提交
2037
			ioc = current_io_context(GFP_NOIO);
L
Linus Torvalds 已提交
2038
			ioc_set_batching(q, ioc);
N
Nick Piggin 已提交
2039 2040

			spin_lock_irq(q->queue_lock);
L
Linus Torvalds 已提交
2041 2042
		}
		finish_wait(&rl->wait[rw], &wait);
2043
	}
L
Linus Torvalds 已提交
2044 2045 2046 2047

	return rq;
}

A
Al Viro 已提交
2048
struct request *blk_get_request(request_queue_t *q, int rw, gfp_t gfp_mask)
L
Linus Torvalds 已提交
2049 2050 2051 2052 2053
{
	struct request *rq;

	BUG_ON(rw != READ && rw != WRITE);

N
Nick Piggin 已提交
2054 2055
	spin_lock_irq(q->queue_lock);
	if (gfp_mask & __GFP_WAIT) {
2056
		rq = get_request_wait(q, rw, NULL);
N
Nick Piggin 已提交
2057
	} else {
2058
		rq = get_request(q, rw, NULL, gfp_mask);
N
Nick Piggin 已提交
2059 2060 2061 2062
		if (!rq)
			spin_unlock_irq(q->queue_lock);
	}
	/* q->queue_lock is unlocked at this point */
L
Linus Torvalds 已提交
2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107

	return rq;
}
EXPORT_SYMBOL(blk_get_request);

/**
 * blk_requeue_request - put a request back on queue
 * @q:		request queue where request should be inserted
 * @rq:		request to be inserted
 *
 * Description:
 *    Drivers often keep queueing requests until the hardware cannot accept
 *    more, when that condition happens we need to put the request back
 *    on the queue. Must be called with queue lock held.
 */
void blk_requeue_request(request_queue_t *q, struct request *rq)
{
	if (blk_rq_tagged(rq))
		blk_queue_end_tag(q, rq);

	elv_requeue_request(q, rq);
}

EXPORT_SYMBOL(blk_requeue_request);

/**
 * blk_insert_request - insert a special request in to a request queue
 * @q:		request queue where request should be inserted
 * @rq:		request to be inserted
 * @at_head:	insert request at head or tail of queue
 * @data:	private data
 *
 * Description:
 *    Many block devices need to execute commands asynchronously, so they don't
 *    block the whole kernel from preemption during request execution.  This is
 *    accomplished normally by inserting aritficial requests tagged as
 *    REQ_SPECIAL in to the corresponding request queue, and letting them be
 *    scheduled for actual execution by the request queue.
 *
 *    We have the option of inserting the head or the tail of the queue.
 *    Typically we use the tail for new ioctls and so forth.  We use the head
 *    of the queue for things like a QUEUE_FULL message from a device, or a
 *    host that is unable to accept a particular command.
 */
void blk_insert_request(request_queue_t *q, struct request *rq,
2108
			int at_head, void *data)
L
Linus Torvalds 已提交
2109
{
2110
	int where = at_head ? ELEVATOR_INSERT_FRONT : ELEVATOR_INSERT_BACK;
L
Linus Torvalds 已提交
2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126
	unsigned long flags;

	/*
	 * tell I/O scheduler that this isn't a regular read/write (ie it
	 * must not attempt merges on this) and that it acts as a soft
	 * barrier
	 */
	rq->flags |= REQ_SPECIAL | REQ_SOFTBARRIER;

	rq->special = data;

	spin_lock_irqsave(q->queue_lock, flags);

	/*
	 * If command is tagged, release the tag
	 */
2127 2128
	if (blk_rq_tagged(rq))
		blk_queue_end_tag(q, rq);
L
Linus Torvalds 已提交
2129

2130 2131
	drive_stat_acct(rq, rq->nr_sectors, 1);
	__elv_add_request(q, rq, where, 0);
L
Linus Torvalds 已提交
2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144

	if (blk_queue_plugged(q))
		__generic_unplug_device(q);
	else
		q->request_fn(q);
	spin_unlock_irqrestore(q->queue_lock, flags);
}

EXPORT_SYMBOL(blk_insert_request);

/**
 * blk_rq_map_user - map user data to a request, for REQ_BLOCK_PC usage
 * @q:		request queue where request should be inserted
2145
 * @rq:		request structure to fill
L
Linus Torvalds 已提交
2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161
 * @ubuf:	the user buffer
 * @len:	length of user data
 *
 * Description:
 *    Data will be mapped directly for zero copy io, if possible. Otherwise
 *    a kernel bounce buffer is used.
 *
 *    A matching blk_rq_unmap_user() must be issued at the end of io, while
 *    still in process context.
 *
 *    Note: The mapped bio may need to be bounced through blk_queue_bounce()
 *    before being submitted to the device, as pages mapped may be out of
 *    reach. It's the callers responsibility to make sure this happens. The
 *    original bio must be passed back in to blk_rq_unmap_user() for proper
 *    unmapping.
 */
2162 2163
int blk_rq_map_user(request_queue_t *q, struct request *rq, void __user *ubuf,
		    unsigned int len)
L
Linus Torvalds 已提交
2164 2165 2166
{
	unsigned long uaddr;
	struct bio *bio;
2167
	int reading;
L
Linus Torvalds 已提交
2168

2169
	if (len > (q->max_hw_sectors << 9))
2170 2171 2172
		return -EINVAL;
	if (!len || !ubuf)
		return -EINVAL;
L
Linus Torvalds 已提交
2173

2174
	reading = rq_data_dir(rq) == READ;
L
Linus Torvalds 已提交
2175 2176 2177 2178 2179 2180 2181

	/*
	 * if alignment requirement is satisfied, map in user pages for
	 * direct dma. else, set up kernel bounce buffers
	 */
	uaddr = (unsigned long) ubuf;
	if (!(uaddr & queue_dma_alignment(q)) && !(len & queue_dma_alignment(q)))
2182
		bio = bio_map_user(q, NULL, uaddr, len, reading);
L
Linus Torvalds 已提交
2183
	else
2184
		bio = bio_copy_user(q, uaddr, len, reading);
L
Linus Torvalds 已提交
2185 2186 2187 2188 2189 2190 2191

	if (!IS_ERR(bio)) {
		rq->bio = rq->biotail = bio;
		blk_rq_bio_prep(q, rq, bio);

		rq->buffer = rq->data = NULL;
		rq->data_len = len;
2192
		return 0;
L
Linus Torvalds 已提交
2193 2194 2195 2196 2197
	}

	/*
	 * bio is the err-ptr
	 */
2198
	return PTR_ERR(bio);
L
Linus Torvalds 已提交
2199 2200 2201 2202
}

EXPORT_SYMBOL(blk_rq_map_user);

2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246
/**
 * blk_rq_map_user_iov - map user data to a request, for REQ_BLOCK_PC usage
 * @q:		request queue where request should be inserted
 * @rq:		request to map data to
 * @iov:	pointer to the iovec
 * @iov_count:	number of elements in the iovec
 *
 * Description:
 *    Data will be mapped directly for zero copy io, if possible. Otherwise
 *    a kernel bounce buffer is used.
 *
 *    A matching blk_rq_unmap_user() must be issued at the end of io, while
 *    still in process context.
 *
 *    Note: The mapped bio may need to be bounced through blk_queue_bounce()
 *    before being submitted to the device, as pages mapped may be out of
 *    reach. It's the callers responsibility to make sure this happens. The
 *    original bio must be passed back in to blk_rq_unmap_user() for proper
 *    unmapping.
 */
int blk_rq_map_user_iov(request_queue_t *q, struct request *rq,
			struct sg_iovec *iov, int iov_count)
{
	struct bio *bio;

	if (!iov || iov_count <= 0)
		return -EINVAL;

	/* we don't allow misaligned data like bio_map_user() does.  If the
	 * user is using sg, they're expected to know the alignment constraints
	 * and respect them accordingly */
	bio = bio_map_user_iov(q, NULL, iov, iov_count, rq_data_dir(rq)== READ);
	if (IS_ERR(bio))
		return PTR_ERR(bio);

	rq->bio = rq->biotail = bio;
	blk_rq_bio_prep(q, rq, bio);
	rq->buffer = rq->data = NULL;
	rq->data_len = bio->bi_size;
	return 0;
}

EXPORT_SYMBOL(blk_rq_map_user_iov);

L
Linus Torvalds 已提交
2247 2248
/**
 * blk_rq_unmap_user - unmap a request with user data
2249
 * @bio:	bio to be unmapped
L
Linus Torvalds 已提交
2250 2251 2252
 * @ulen:	length of user buffer
 *
 * Description:
2253
 *    Unmap a bio previously mapped by blk_rq_map_user().
L
Linus Torvalds 已提交
2254
 */
2255
int blk_rq_unmap_user(struct bio *bio, unsigned int ulen)
L
Linus Torvalds 已提交
2256 2257 2258 2259 2260 2261 2262 2263 2264 2265
{
	int ret = 0;

	if (bio) {
		if (bio_flagged(bio, BIO_USER_MAPPED))
			bio_unmap_user(bio);
		else
			ret = bio_uncopy_user(bio);
	}

2266
	return 0;
L
Linus Torvalds 已提交
2267 2268 2269 2270
}

EXPORT_SYMBOL(blk_rq_unmap_user);

M
Mike Christie 已提交
2271 2272 2273
/**
 * blk_rq_map_kern - map kernel data to a request, for REQ_BLOCK_PC usage
 * @q:		request queue where request should be inserted
2274
 * @rq:		request to fill
M
Mike Christie 已提交
2275 2276
 * @kbuf:	the kernel buffer
 * @len:	length of user data
2277
 * @gfp_mask:	memory allocation flags
M
Mike Christie 已提交
2278
 */
2279
int blk_rq_map_kern(request_queue_t *q, struct request *rq, void *kbuf,
A
Al Viro 已提交
2280
		    unsigned int len, gfp_t gfp_mask)
M
Mike Christie 已提交
2281 2282 2283
{
	struct bio *bio;

2284
	if (len > (q->max_hw_sectors << 9))
2285 2286 2287
		return -EINVAL;
	if (!len || !kbuf)
		return -EINVAL;
M
Mike Christie 已提交
2288 2289

	bio = bio_map_kern(q, kbuf, len, gfp_mask);
2290 2291
	if (IS_ERR(bio))
		return PTR_ERR(bio);
M
Mike Christie 已提交
2292

2293 2294
	if (rq_data_dir(rq) == WRITE)
		bio->bi_rw |= (1 << BIO_RW);
M
Mike Christie 已提交
2295

2296 2297
	rq->bio = rq->biotail = bio;
	blk_rq_bio_prep(q, rq, bio);
M
Mike Christie 已提交
2298

2299 2300 2301
	rq->buffer = rq->data = NULL;
	rq->data_len = len;
	return 0;
M
Mike Christie 已提交
2302 2303 2304 2305
}

EXPORT_SYMBOL(blk_rq_map_kern);

2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317
/**
 * blk_execute_rq_nowait - insert a request into queue for execution
 * @q:		queue to insert the request in
 * @bd_disk:	matching gendisk
 * @rq:		request to insert
 * @at_head:    insert request at head or tail of queue
 * @done:	I/O completion handler
 *
 * Description:
 *    Insert a fully prepared request at the back of the io scheduler queue
 *    for execution.  Don't wait for completion.
 */
2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330
void blk_execute_rq_nowait(request_queue_t *q, struct gendisk *bd_disk,
			   struct request *rq, int at_head,
			   void (*done)(struct request *))
{
	int where = at_head ? ELEVATOR_INSERT_FRONT : ELEVATOR_INSERT_BACK;

	rq->rq_disk = bd_disk;
	rq->flags |= REQ_NOMERGE;
	rq->end_io = done;
	elv_add_request(q, rq, where, 1);
	generic_unplug_device(q);
}

2331 2332
EXPORT_SYMBOL_GPL(blk_execute_rq_nowait);

L
Linus Torvalds 已提交
2333 2334 2335 2336 2337
/**
 * blk_execute_rq - insert a request into queue for execution
 * @q:		queue to insert the request in
 * @bd_disk:	matching gendisk
 * @rq:		request to insert
2338
 * @at_head:    insert request at head or tail of queue
L
Linus Torvalds 已提交
2339 2340 2341
 *
 * Description:
 *    Insert a fully prepared request at the back of the io scheduler queue
2342
 *    for execution and wait for completion.
L
Linus Torvalds 已提交
2343 2344
 */
int blk_execute_rq(request_queue_t *q, struct gendisk *bd_disk,
2345
		   struct request *rq, int at_head)
L
Linus Torvalds 已提交
2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363
{
	DECLARE_COMPLETION(wait);
	char sense[SCSI_SENSE_BUFFERSIZE];
	int err = 0;

	/*
	 * we need an extra reference to the request, so we can look at
	 * it after io completion
	 */
	rq->ref_count++;

	if (!rq->sense) {
		memset(sense, 0, sizeof(sense));
		rq->sense = sense;
		rq->sense_len = 0;
	}

	rq->waiting = &wait;
2364
	blk_execute_rq_nowait(q, bd_disk, rq, at_head, blk_end_sync_rq);
L
Linus Torvalds 已提交
2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403
	wait_for_completion(&wait);
	rq->waiting = NULL;

	if (rq->errors)
		err = -EIO;

	return err;
}

EXPORT_SYMBOL(blk_execute_rq);

/**
 * blkdev_issue_flush - queue a flush
 * @bdev:	blockdev to issue flush for
 * @error_sector:	error sector
 *
 * Description:
 *    Issue a flush for the block device in question. Caller can supply
 *    room for storing the error offset in case of a flush error, if they
 *    wish to.  Caller must run wait_for_completion() on its own.
 */
int blkdev_issue_flush(struct block_device *bdev, sector_t *error_sector)
{
	request_queue_t *q;

	if (bdev->bd_disk == NULL)
		return -ENXIO;

	q = bdev_get_queue(bdev);
	if (!q)
		return -ENXIO;
	if (!q->issue_flush_fn)
		return -EOPNOTSUPP;

	return q->issue_flush_fn(q, bdev->bd_disk, error_sector);
}

EXPORT_SYMBOL(blkdev_issue_flush);

2404
static void drive_stat_acct(struct request *rq, int nr_sectors, int new_io)
L
Linus Torvalds 已提交
2405 2406 2407 2408 2409 2410
{
	int rw = rq_data_dir(rq);

	if (!blk_fs_request(rq) || !rq->rq_disk)
		return;

2411
	if (!new_io) {
2412
		__disk_stat_inc(rq->rq_disk, merges[rw]);
2413
	} else {
L
Linus Torvalds 已提交
2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 2455 2456
		disk_round_stats(rq->rq_disk);
		rq->rq_disk->in_flight++;
	}
}

/*
 * add-request adds a request to the linked list.
 * queue lock is held and interrupts disabled, as we muck with the
 * request queue list.
 */
static inline void add_request(request_queue_t * q, struct request * req)
{
	drive_stat_acct(req, req->nr_sectors, 1);

	if (q->activity_fn)
		q->activity_fn(q->activity_data, rq_data_dir(req));

	/*
	 * elevator indicated where it wants this request to be
	 * inserted at elevator_merge time
	 */
	__elv_add_request(q, req, ELEVATOR_INSERT_SORT, 0);
}
 
/*
 * disk_round_stats()	- Round off the performance stats on a struct
 * disk_stats.
 *
 * The average IO queue length and utilisation statistics are maintained
 * by observing the current state of the queue length and the amount of
 * time it has been in this state for.
 *
 * Normally, that accounting is done on IO completion, but that can result
 * in more than a second's worth of IO being accounted for within any one
 * second, leading to >100% utilisation.  To deal with that, we call this
 * function to do a round-off before returning the results when reading
 * /proc/diskstats.  This accounts immediately for all queue usage up to
 * the current jiffies and restarts the counters again.
 */
void disk_round_stats(struct gendisk *disk)
{
	unsigned long now = jiffies;

2457 2458
	if (now == disk->stamp)
		return;
L
Linus Torvalds 已提交
2459

2460 2461 2462 2463 2464
	if (disk->in_flight) {
		__disk_stat_add(disk, time_in_queue,
				disk->in_flight * (now - disk->stamp));
		__disk_stat_add(disk, io_ticks, (now - disk->stamp));
	}
L
Linus Torvalds 已提交
2465 2466 2467 2468 2469 2470
	disk->stamp = now;
}

/*
 * queue lock must be held
 */
2471
void __blk_put_request(request_queue_t *q, struct request *req)
L
Linus Torvalds 已提交
2472 2473 2474 2475 2476 2477 2478 2479
{
	struct request_list *rl = req->rl;

	if (unlikely(!q))
		return;
	if (unlikely(--req->ref_count))
		return;

2480 2481
	elv_completed_request(q, req);

L
Linus Torvalds 已提交
2482 2483 2484 2485 2486 2487 2488 2489 2490
	req->rq_status = RQ_INACTIVE;
	req->rl = NULL;

	/*
	 * Request may not have originated from ll_rw_blk. if not,
	 * it didn't come out of our reserved rq pools
	 */
	if (rl) {
		int rw = rq_data_dir(req);
T
Tejun Heo 已提交
2491
		int priv = req->flags & REQ_ELVPRIV;
L
Linus Torvalds 已提交
2492 2493 2494 2495

		BUG_ON(!list_empty(&req->queuelist));

		blk_free_request(q, req);
T
Tejun Heo 已提交
2496
		freed_request(q, rw, priv);
L
Linus Torvalds 已提交
2497 2498 2499
	}
}

2500 2501
EXPORT_SYMBOL_GPL(__blk_put_request);

L
Linus Torvalds 已提交
2502 2503
void blk_put_request(struct request *req)
{
2504 2505 2506
	unsigned long flags;
	request_queue_t *q = req->q;

L
Linus Torvalds 已提交
2507
	/*
2508 2509
	 * Gee, IDE calls in w/ NULL q.  Fix IDE and remove the
	 * following if (q) test.
L
Linus Torvalds 已提交
2510
	 */
2511
	if (q) {
L
Linus Torvalds 已提交
2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611
		spin_lock_irqsave(q->queue_lock, flags);
		__blk_put_request(q, req);
		spin_unlock_irqrestore(q->queue_lock, flags);
	}
}

EXPORT_SYMBOL(blk_put_request);

/**
 * blk_end_sync_rq - executes a completion event on a request
 * @rq: request to complete
 */
void blk_end_sync_rq(struct request *rq)
{
	struct completion *waiting = rq->waiting;

	rq->waiting = NULL;
	__blk_put_request(rq->q, rq);

	/*
	 * complete last, if this is a stack request the process (and thus
	 * the rq pointer) could be invalid right after this complete()
	 */
	complete(waiting);
}
EXPORT_SYMBOL(blk_end_sync_rq);

/**
 * blk_congestion_wait - wait for a queue to become uncongested
 * @rw: READ or WRITE
 * @timeout: timeout in jiffies
 *
 * Waits for up to @timeout jiffies for a queue (any queue) to exit congestion.
 * If no queues are congested then just wait for the next request to be
 * returned.
 */
long blk_congestion_wait(int rw, long timeout)
{
	long ret;
	DEFINE_WAIT(wait);
	wait_queue_head_t *wqh = &congestion_wqh[rw];

	prepare_to_wait(wqh, &wait, TASK_UNINTERRUPTIBLE);
	ret = io_schedule_timeout(timeout);
	finish_wait(wqh, &wait);
	return ret;
}

EXPORT_SYMBOL(blk_congestion_wait);

/*
 * Has to be called with the request spinlock acquired
 */
static int attempt_merge(request_queue_t *q, struct request *req,
			  struct request *next)
{
	if (!rq_mergeable(req) || !rq_mergeable(next))
		return 0;

	/*
	 * not contigious
	 */
	if (req->sector + req->nr_sectors != next->sector)
		return 0;

	if (rq_data_dir(req) != rq_data_dir(next)
	    || req->rq_disk != next->rq_disk
	    || next->waiting || next->special)
		return 0;

	/*
	 * If we are allowed to merge, then append bio list
	 * from next to rq and release next. merge_requests_fn
	 * will have updated segment counts, update sector
	 * counts here.
	 */
	if (!q->merge_requests_fn(q, req, next))
		return 0;

	/*
	 * At this point we have either done a back merge
	 * or front merge. We need the smaller start_time of
	 * the merged requests to be the current request
	 * for accounting purposes.
	 */
	if (time_after(req->start_time, next->start_time))
		req->start_time = next->start_time;

	req->biotail->bi_next = next->bio;
	req->biotail = next->biotail;

	req->nr_sectors = req->hard_nr_sectors += next->hard_nr_sectors;

	elv_merge_requests(q, req, next);

	if (req->rq_disk) {
		disk_round_stats(req->rq_disk);
		req->rq_disk->in_flight--;
	}

2612 2613
	req->ioprio = ioprio_best(req->ioprio, next->ioprio);

L
Linus Torvalds 已提交
2614 2615 2616 2617 2618 2619 2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632 2633 2634 2635 2636 2637 2638 2639 2640 2641 2642 2643 2644 2645 2646 2647 2648 2649 2650 2651 2652 2653 2654 2655 2656 2657 2658 2659 2660 2661 2662 2663
	__blk_put_request(q, next);
	return 1;
}

static inline int attempt_back_merge(request_queue_t *q, struct request *rq)
{
	struct request *next = elv_latter_request(q, rq);

	if (next)
		return attempt_merge(q, rq, next);

	return 0;
}

static inline int attempt_front_merge(request_queue_t *q, struct request *rq)
{
	struct request *prev = elv_former_request(q, rq);

	if (prev)
		return attempt_merge(q, prev, rq);

	return 0;
}

/**
 * blk_attempt_remerge  - attempt to remerge active head with next request
 * @q:    The &request_queue_t belonging to the device
 * @rq:   The head request (usually)
 *
 * Description:
 *    For head-active devices, the queue can easily be unplugged so quickly
 *    that proper merging is not done on the front request. This may hurt
 *    performance greatly for some devices. The block layer cannot safely
 *    do merging on that first request for these queues, but the driver can
 *    call this function and make it happen any way. Only the driver knows
 *    when it is safe to do so.
 **/
void blk_attempt_remerge(request_queue_t *q, struct request *rq)
{
	unsigned long flags;

	spin_lock_irqsave(q->queue_lock, flags);
	attempt_back_merge(q, rq);
	spin_unlock_irqrestore(q->queue_lock, flags);
}

EXPORT_SYMBOL(blk_attempt_remerge);

static int __make_request(request_queue_t *q, struct bio *bio)
{
2664
	struct request *req;
2665
	int el_ret, rw, nr_sectors, cur_nr_sectors, barrier, err, sync;
2666
	unsigned short prio;
L
Linus Torvalds 已提交
2667 2668 2669 2670 2671
	sector_t sector;

	sector = bio->bi_sector;
	nr_sectors = bio_sectors(bio);
	cur_nr_sectors = bio_cur_sectors(bio);
2672
	prio = bio_prio(bio);
L
Linus Torvalds 已提交
2673 2674

	rw = bio_data_dir(bio);
2675
	sync = bio_sync(bio);
L
Linus Torvalds 已提交
2676 2677 2678 2679 2680 2681 2682 2683 2684 2685 2686

	/*
	 * low level driver can indicate that it wants pages above a
	 * certain limit bounced to low memory (ie for highmem, or even
	 * ISA dma in theory)
	 */
	blk_queue_bounce(q, &bio);

	spin_lock_prefetch(q->queue_lock);

	barrier = bio_barrier(bio);
N
Nick Piggin 已提交
2687
	if (unlikely(barrier) && (q->ordered == QUEUE_ORDERED_NONE)) {
L
Linus Torvalds 已提交
2688 2689 2690 2691 2692 2693
		err = -EOPNOTSUPP;
		goto end_io;
	}

	spin_lock_irq(q->queue_lock);

2694
	if (unlikely(barrier) || elv_queue_empty(q))
L
Linus Torvalds 已提交
2695 2696 2697 2698 2699 2700 2701 2702 2703 2704 2705 2706 2707
		goto get_rq;

	el_ret = elv_merge(q, &req, bio);
	switch (el_ret) {
		case ELEVATOR_BACK_MERGE:
			BUG_ON(!rq_mergeable(req));

			if (!q->back_merge_fn(q, req, bio))
				break;

			req->biotail->bi_next = bio;
			req->biotail = bio;
			req->nr_sectors = req->hard_nr_sectors += nr_sectors;
2708
			req->ioprio = ioprio_best(req->ioprio, prio);
L
Linus Torvalds 已提交
2709 2710 2711 2712 2713 2714 2715 2716 2717 2718 2719 2720 2721 2722 2723 2724 2725 2726 2727 2728 2729 2730 2731 2732
			drive_stat_acct(req, nr_sectors, 0);
			if (!attempt_back_merge(q, req))
				elv_merged_request(q, req);
			goto out;

		case ELEVATOR_FRONT_MERGE:
			BUG_ON(!rq_mergeable(req));

			if (!q->front_merge_fn(q, req, bio))
				break;

			bio->bi_next = req->bio;
			req->bio = bio;

			/*
			 * may not be valid. if the low level driver said
			 * it didn't need a bounce buffer then it better
			 * not touch req->buffer either...
			 */
			req->buffer = bio_data(bio);
			req->current_nr_sectors = cur_nr_sectors;
			req->hard_cur_sectors = cur_nr_sectors;
			req->sector = req->hard_sector = sector;
			req->nr_sectors = req->hard_nr_sectors += nr_sectors;
2733
			req->ioprio = ioprio_best(req->ioprio, prio);
L
Linus Torvalds 已提交
2734 2735 2736 2737 2738
			drive_stat_acct(req, nr_sectors, 0);
			if (!attempt_front_merge(q, req))
				elv_merged_request(q, req);
			goto out;

2739
		/* ELV_NO_MERGE: elevator says don't/can't merge. */
L
Linus Torvalds 已提交
2740
		default:
2741
			;
L
Linus Torvalds 已提交
2742 2743
	}

2744
get_rq:
L
Linus Torvalds 已提交
2745
	/*
2746
	 * Grab a free request. This is might sleep but can not fail.
N
Nick Piggin 已提交
2747
	 * Returns with the queue unlocked.
2748 2749
	 */
	req = get_request_wait(q, rw, bio);
N
Nick Piggin 已提交
2750

2751 2752 2753 2754 2755
	/*
	 * After dropping the lock and possibly sleeping here, our request
	 * may now be mergeable after it had proven unmergeable (above).
	 * We don't worry about that case for efficiency. It won't happen
	 * often, and the elevators are able to handle it.
L
Linus Torvalds 已提交
2756 2757 2758 2759 2760 2761 2762 2763 2764 2765 2766 2767 2768
	 */

	req->flags |= REQ_CMD;

	/*
	 * inherit FAILFAST from bio (for read-ahead, and explicit FAILFAST)
	 */
	if (bio_rw_ahead(bio) || bio_failfast(bio))
		req->flags |= REQ_FAILFAST;

	/*
	 * REQ_BARRIER implies no merging, but lets make it explicit
	 */
N
Nick Piggin 已提交
2769
	if (unlikely(barrier))
L
Linus Torvalds 已提交
2770 2771 2772 2773 2774 2775 2776 2777 2778 2779 2780
		req->flags |= (REQ_HARDBARRIER | REQ_NOMERGE);

	req->errors = 0;
	req->hard_sector = req->sector = sector;
	req->hard_nr_sectors = req->nr_sectors = nr_sectors;
	req->current_nr_sectors = req->hard_cur_sectors = cur_nr_sectors;
	req->nr_phys_segments = bio_phys_segments(q, bio);
	req->nr_hw_segments = bio_hw_segments(q, bio);
	req->buffer = bio_data(bio);	/* see ->buffer comment above */
	req->waiting = NULL;
	req->bio = req->biotail = bio;
2781
	req->ioprio = prio;
L
Linus Torvalds 已提交
2782 2783 2784
	req->rq_disk = bio->bi_bdev->bd_disk;
	req->start_time = jiffies;

2785 2786 2787
	spin_lock_irq(q->queue_lock);
	if (elv_queue_empty(q))
		blk_plug_device(q);
L
Linus Torvalds 已提交
2788 2789
	add_request(q, req);
out:
2790
	if (sync)
L
Linus Torvalds 已提交
2791 2792 2793 2794 2795 2796 2797 2798 2799 2800 2801 2802 2803 2804 2805 2806 2807 2808 2809
		__generic_unplug_device(q);

	spin_unlock_irq(q->queue_lock);
	return 0;

end_io:
	bio_endio(bio, nr_sectors << 9, err);
	return 0;
}

/*
 * If bio->bi_dev is a partition, remap the location
 */
static inline void blk_partition_remap(struct bio *bio)
{
	struct block_device *bdev = bio->bi_bdev;

	if (bdev != bdev->bd_contains) {
		struct hd_struct *p = bdev->bd_part;
2810 2811 2812 2813
		const int rw = bio_data_dir(bio);

		p->sectors[rw] += bio_sectors(bio);
		p->ios[rw]++;
L
Linus Torvalds 已提交
2814 2815 2816 2817 2818 2819 2820 2821 2822 2823 2824 2825 2826 2827 2828 2829 2830 2831 2832 2833 2834 2835 2836 2837 2838 2839 2840 2841 2842 2843 2844 2845 2846 2847 2848 2849 2850 2851 2852 2853 2854 2855 2856 2857 2858 2859 2860 2861 2862 2863 2864 2865 2866 2867 2868 2869 2870 2871 2872 2873 2874 2875 2876 2877 2878 2879 2880 2881 2882 2883 2884 2885 2886 2887 2888 2889 2890 2891 2892 2893 2894 2895 2896 2897 2898 2899 2900 2901 2902 2903 2904 2905 2906 2907 2908 2909 2910 2911

		bio->bi_sector += p->start_sect;
		bio->bi_bdev = bdev->bd_contains;
	}
}

static void handle_bad_sector(struct bio *bio)
{
	char b[BDEVNAME_SIZE];

	printk(KERN_INFO "attempt to access beyond end of device\n");
	printk(KERN_INFO "%s: rw=%ld, want=%Lu, limit=%Lu\n",
			bdevname(bio->bi_bdev, b),
			bio->bi_rw,
			(unsigned long long)bio->bi_sector + bio_sectors(bio),
			(long long)(bio->bi_bdev->bd_inode->i_size >> 9));

	set_bit(BIO_EOF, &bio->bi_flags);
}

/**
 * generic_make_request: hand a buffer to its device driver for I/O
 * @bio:  The bio describing the location in memory and on the device.
 *
 * generic_make_request() is used to make I/O requests of block
 * devices. It is passed a &struct bio, which describes the I/O that needs
 * to be done.
 *
 * generic_make_request() does not return any status.  The
 * success/failure status of the request, along with notification of
 * completion, is delivered asynchronously through the bio->bi_end_io
 * function described (one day) else where.
 *
 * The caller of generic_make_request must make sure that bi_io_vec
 * are set to describe the memory buffer, and that bi_dev and bi_sector are
 * set to describe the device address, and the
 * bi_end_io and optionally bi_private are set to describe how
 * completion notification should be signaled.
 *
 * generic_make_request and the drivers it calls may use bi_next if this
 * bio happens to be merged with someone else, and may change bi_dev and
 * bi_sector for remaps as it sees fit.  So the values of these fields
 * should NOT be depended on after the call to generic_make_request.
 */
void generic_make_request(struct bio *bio)
{
	request_queue_t *q;
	sector_t maxsector;
	int ret, nr_sectors = bio_sectors(bio);

	might_sleep();
	/* Test device or partition size, when known. */
	maxsector = bio->bi_bdev->bd_inode->i_size >> 9;
	if (maxsector) {
		sector_t sector = bio->bi_sector;

		if (maxsector < nr_sectors || maxsector - nr_sectors < sector) {
			/*
			 * This may well happen - the kernel calls bread()
			 * without checking the size of the device, e.g., when
			 * mounting a device.
			 */
			handle_bad_sector(bio);
			goto end_io;
		}
	}

	/*
	 * Resolve the mapping until finished. (drivers are
	 * still free to implement/resolve their own stacking
	 * by explicitly returning 0)
	 *
	 * NOTE: we don't repeat the blk_size check for each new device.
	 * Stacking drivers are expected to know what they are doing.
	 */
	do {
		char b[BDEVNAME_SIZE];

		q = bdev_get_queue(bio->bi_bdev);
		if (!q) {
			printk(KERN_ERR
			       "generic_make_request: Trying to access "
				"nonexistent block-device %s (%Lu)\n",
				bdevname(bio->bi_bdev, b),
				(long long) bio->bi_sector);
end_io:
			bio_endio(bio, bio->bi_size, -EIO);
			break;
		}

		if (unlikely(bio_sectors(bio) > q->max_hw_sectors)) {
			printk("bio too big device %s (%u > %u)\n", 
				bdevname(bio->bi_bdev, b),
				bio_sectors(bio),
				q->max_hw_sectors);
			goto end_io;
		}

N
Nick Piggin 已提交
2912
		if (unlikely(test_bit(QUEUE_FLAG_DEAD, &q->queue_flags)))
L
Linus Torvalds 已提交
2913 2914 2915 2916 2917 2918 2919 2920 2921 2922 2923 2924 2925 2926 2927 2928 2929 2930 2931 2932 2933 2934 2935 2936 2937 2938 2939 2940 2941 2942
			goto end_io;

		/*
		 * If this device has partitions, remap block n
		 * of partition p to block n+start(p) of the disk.
		 */
		blk_partition_remap(bio);

		ret = q->make_request_fn(q, bio);
	} while (ret);
}

EXPORT_SYMBOL(generic_make_request);

/**
 * submit_bio: submit a bio to the block device layer for I/O
 * @rw: whether to %READ or %WRITE, or maybe to %READA (read ahead)
 * @bio: The &struct bio which describes the I/O
 *
 * submit_bio() is very similar in purpose to generic_make_request(), and
 * uses that function to do most of the work. Both are fairly rough
 * interfaces, @bio must be presetup and ready for I/O.
 *
 */
void submit_bio(int rw, struct bio *bio)
{
	int count = bio_sectors(bio);

	BIO_BUG_ON(!bio->bi_size);
	BIO_BUG_ON(!bio->bi_io_vec);
2943
	bio->bi_rw |= rw;
L
Linus Torvalds 已提交
2944 2945 2946 2947 2948 2949 2950 2951 2952 2953 2954 2955 2956 2957 2958 2959 2960 2961 2962
	if (rw & WRITE)
		mod_page_state(pgpgout, count);
	else
		mod_page_state(pgpgin, count);

	if (unlikely(block_dump)) {
		char b[BDEVNAME_SIZE];
		printk(KERN_DEBUG "%s(%d): %s block %Lu on %s\n",
			current->comm, current->pid,
			(rw & WRITE) ? "WRITE" : "READ",
			(unsigned long long)bio->bi_sector,
			bdevname(bio->bi_bdev,b));
	}

	generic_make_request(bio);
}

EXPORT_SYMBOL(submit_bio);

2963
static void blk_recalc_rq_segments(struct request *rq)
L
Linus Torvalds 已提交
2964 2965 2966 2967 2968 2969 2970 2971 2972 2973 2974 2975 2976 2977 2978 2979 2980 2981 2982 2983 2984 2985 2986 2987 2988 2989 2990 2991 2992 2993 2994 2995 2996 2997 2998 2999 3000 3001 3002 3003 3004
{
	struct bio *bio, *prevbio = NULL;
	int nr_phys_segs, nr_hw_segs;
	unsigned int phys_size, hw_size;
	request_queue_t *q = rq->q;

	if (!rq->bio)
		return;

	phys_size = hw_size = nr_phys_segs = nr_hw_segs = 0;
	rq_for_each_bio(bio, rq) {
		/* Force bio hw/phys segs to be recalculated. */
		bio->bi_flags &= ~(1 << BIO_SEG_VALID);

		nr_phys_segs += bio_phys_segments(q, bio);
		nr_hw_segs += bio_hw_segments(q, bio);
		if (prevbio) {
			int pseg = phys_size + prevbio->bi_size + bio->bi_size;
			int hseg = hw_size + prevbio->bi_size + bio->bi_size;

			if (blk_phys_contig_segment(q, prevbio, bio) &&
			    pseg <= q->max_segment_size) {
				nr_phys_segs--;
				phys_size += prevbio->bi_size + bio->bi_size;
			} else
				phys_size = 0;

			if (blk_hw_contig_segment(q, prevbio, bio) &&
			    hseg <= q->max_segment_size) {
				nr_hw_segs--;
				hw_size += prevbio->bi_size + bio->bi_size;
			} else
				hw_size = 0;
		}
		prevbio = bio;
	}

	rq->nr_phys_segments = nr_phys_segs;
	rq->nr_hw_segments = nr_hw_segs;
}

3005
static void blk_recalc_rq_sectors(struct request *rq, int nsect)
L
Linus Torvalds 已提交
3006 3007 3008 3009 3010 3011 3012 3013 3014 3015 3016 3017 3018 3019 3020 3021 3022 3023 3024 3025 3026 3027 3028 3029 3030 3031 3032 3033 3034 3035 3036 3037 3038 3039 3040 3041 3042 3043 3044 3045 3046 3047 3048 3049 3050 3051 3052 3053 3054 3055 3056 3057 3058 3059 3060
{
	if (blk_fs_request(rq)) {
		rq->hard_sector += nsect;
		rq->hard_nr_sectors -= nsect;

		/*
		 * Move the I/O submission pointers ahead if required.
		 */
		if ((rq->nr_sectors >= rq->hard_nr_sectors) &&
		    (rq->sector <= rq->hard_sector)) {
			rq->sector = rq->hard_sector;
			rq->nr_sectors = rq->hard_nr_sectors;
			rq->hard_cur_sectors = bio_cur_sectors(rq->bio);
			rq->current_nr_sectors = rq->hard_cur_sectors;
			rq->buffer = bio_data(rq->bio);
		}

		/*
		 * if total number of sectors is less than the first segment
		 * size, something has gone terribly wrong
		 */
		if (rq->nr_sectors < rq->current_nr_sectors) {
			printk("blk: request botched\n");
			rq->nr_sectors = rq->current_nr_sectors;
		}
	}
}

static int __end_that_request_first(struct request *req, int uptodate,
				    int nr_bytes)
{
	int total_bytes, bio_nbytes, error, next_idx = 0;
	struct bio *bio;

	/*
	 * extend uptodate bool to allow < 0 value to be direct io error
	 */
	error = 0;
	if (end_io_error(uptodate))
		error = !uptodate ? -EIO : uptodate;

	/*
	 * for a REQ_BLOCK_PC request, we want to carry any eventual
	 * sense key with us all the way through
	 */
	if (!blk_pc_request(req))
		req->errors = 0;

	if (!uptodate) {
		if (blk_fs_request(req) && !(req->flags & REQ_QUIET))
			printk("end_request: I/O error, dev %s, sector %llu\n",
				req->rq_disk ? req->rq_disk->disk_name : "?",
				(unsigned long long)req->sector);
	}

3061
	if (blk_fs_request(req) && req->rq_disk) {
3062 3063 3064
		const int rw = rq_data_dir(req);

		__disk_stat_add(req->rq_disk, sectors[rw], nr_bytes >> 9);
3065 3066
	}

L
Linus Torvalds 已提交
3067 3068 3069 3070 3071 3072 3073 3074 3075 3076 3077 3078 3079 3080 3081 3082 3083 3084 3085 3086 3087 3088 3089 3090 3091 3092 3093 3094 3095 3096 3097 3098 3099 3100 3101 3102 3103 3104 3105 3106 3107 3108 3109 3110 3111 3112 3113 3114 3115 3116 3117 3118 3119 3120 3121 3122 3123 3124 3125 3126 3127 3128 3129 3130 3131 3132 3133 3134 3135 3136 3137 3138 3139 3140 3141 3142 3143 3144 3145 3146 3147 3148 3149 3150 3151 3152 3153 3154 3155 3156 3157 3158 3159 3160 3161 3162 3163 3164 3165 3166 3167 3168 3169 3170 3171 3172 3173 3174 3175 3176 3177 3178 3179 3180 3181 3182 3183 3184 3185 3186 3187 3188 3189 3190 3191 3192 3193 3194
	total_bytes = bio_nbytes = 0;
	while ((bio = req->bio) != NULL) {
		int nbytes;

		if (nr_bytes >= bio->bi_size) {
			req->bio = bio->bi_next;
			nbytes = bio->bi_size;
			bio_endio(bio, nbytes, error);
			next_idx = 0;
			bio_nbytes = 0;
		} else {
			int idx = bio->bi_idx + next_idx;

			if (unlikely(bio->bi_idx >= bio->bi_vcnt)) {
				blk_dump_rq_flags(req, "__end_that");
				printk("%s: bio idx %d >= vcnt %d\n",
						__FUNCTION__,
						bio->bi_idx, bio->bi_vcnt);
				break;
			}

			nbytes = bio_iovec_idx(bio, idx)->bv_len;
			BIO_BUG_ON(nbytes > bio->bi_size);

			/*
			 * not a complete bvec done
			 */
			if (unlikely(nbytes > nr_bytes)) {
				bio_nbytes += nr_bytes;
				total_bytes += nr_bytes;
				break;
			}

			/*
			 * advance to the next vector
			 */
			next_idx++;
			bio_nbytes += nbytes;
		}

		total_bytes += nbytes;
		nr_bytes -= nbytes;

		if ((bio = req->bio)) {
			/*
			 * end more in this run, or just return 'not-done'
			 */
			if (unlikely(nr_bytes <= 0))
				break;
		}
	}

	/*
	 * completely done
	 */
	if (!req->bio)
		return 0;

	/*
	 * if the request wasn't completed, update state
	 */
	if (bio_nbytes) {
		bio_endio(bio, bio_nbytes, error);
		bio->bi_idx += next_idx;
		bio_iovec(bio)->bv_offset += nr_bytes;
		bio_iovec(bio)->bv_len -= nr_bytes;
	}

	blk_recalc_rq_sectors(req, total_bytes >> 9);
	blk_recalc_rq_segments(req);
	return 1;
}

/**
 * end_that_request_first - end I/O on a request
 * @req:      the request being processed
 * @uptodate: 1 for success, 0 for I/O error, < 0 for specific error
 * @nr_sectors: number of sectors to end I/O on
 *
 * Description:
 *     Ends I/O on a number of sectors attached to @req, and sets it up
 *     for the next range of segments (if any) in the cluster.
 *
 * Return:
 *     0 - we are done with this request, call end_that_request_last()
 *     1 - still buffers pending for this request
 **/
int end_that_request_first(struct request *req, int uptodate, int nr_sectors)
{
	return __end_that_request_first(req, uptodate, nr_sectors << 9);
}

EXPORT_SYMBOL(end_that_request_first);

/**
 * end_that_request_chunk - end I/O on a request
 * @req:      the request being processed
 * @uptodate: 1 for success, 0 for I/O error, < 0 for specific error
 * @nr_bytes: number of bytes to complete
 *
 * Description:
 *     Ends I/O on a number of bytes attached to @req, and sets it up
 *     for the next range of segments (if any). Like end_that_request_first(),
 *     but deals with bytes instead of sectors.
 *
 * Return:
 *     0 - we are done with this request, call end_that_request_last()
 *     1 - still buffers pending for this request
 **/
int end_that_request_chunk(struct request *req, int uptodate, int nr_bytes)
{
	return __end_that_request_first(req, uptodate, nr_bytes);
}

EXPORT_SYMBOL(end_that_request_chunk);

/*
 * queue lock must be held
 */
void end_that_request_last(struct request *req)
{
	struct gendisk *disk = req->rq_disk;

	if (unlikely(laptop_mode) && blk_fs_request(req))
		laptop_io_completion();

	if (disk && blk_fs_request(req)) {
		unsigned long duration = jiffies - req->start_time;
3195 3196 3197 3198
		const int rw = rq_data_dir(req);

		__disk_stat_inc(disk, ios[rw]);
		__disk_stat_add(disk, ticks[rw], duration);
L
Linus Torvalds 已提交
3199 3200 3201 3202 3203 3204 3205 3206 3207 3208 3209 3210 3211 3212 3213 3214 3215 3216 3217 3218 3219 3220 3221 3222 3223 3224 3225 3226 3227 3228 3229 3230 3231 3232 3233 3234 3235 3236 3237 3238 3239 3240 3241 3242 3243 3244 3245 3246 3247 3248 3249 3250 3251 3252 3253 3254 3255 3256 3257 3258 3259 3260 3261 3262 3263 3264 3265 3266 3267 3268 3269 3270 3271 3272 3273 3274 3275 3276 3277 3278 3279 3280 3281 3282 3283 3284 3285 3286 3287 3288 3289 3290 3291 3292 3293 3294 3295 3296 3297 3298 3299
		disk_round_stats(disk);
		disk->in_flight--;
	}
	if (req->end_io)
		req->end_io(req);
	else
		__blk_put_request(req->q, req);
}

EXPORT_SYMBOL(end_that_request_last);

void end_request(struct request *req, int uptodate)
{
	if (!end_that_request_first(req, uptodate, req->hard_cur_sectors)) {
		add_disk_randomness(req->rq_disk);
		blkdev_dequeue_request(req);
		end_that_request_last(req);
	}
}

EXPORT_SYMBOL(end_request);

void blk_rq_bio_prep(request_queue_t *q, struct request *rq, struct bio *bio)
{
	/* first three bits are identical in rq->flags and bio->bi_rw */
	rq->flags |= (bio->bi_rw & 7);

	rq->nr_phys_segments = bio_phys_segments(q, bio);
	rq->nr_hw_segments = bio_hw_segments(q, bio);
	rq->current_nr_sectors = bio_cur_sectors(bio);
	rq->hard_cur_sectors = rq->current_nr_sectors;
	rq->hard_nr_sectors = rq->nr_sectors = bio_sectors(bio);
	rq->buffer = bio_data(bio);

	rq->bio = rq->biotail = bio;
}

EXPORT_SYMBOL(blk_rq_bio_prep);

int kblockd_schedule_work(struct work_struct *work)
{
	return queue_work(kblockd_workqueue, work);
}

EXPORT_SYMBOL(kblockd_schedule_work);

void kblockd_flush(void)
{
	flush_workqueue(kblockd_workqueue);
}
EXPORT_SYMBOL(kblockd_flush);

int __init blk_dev_init(void)
{
	kblockd_workqueue = create_workqueue("kblockd");
	if (!kblockd_workqueue)
		panic("Failed to create kblockd\n");

	request_cachep = kmem_cache_create("blkdev_requests",
			sizeof(struct request), 0, SLAB_PANIC, NULL, NULL);

	requestq_cachep = kmem_cache_create("blkdev_queue",
			sizeof(request_queue_t), 0, SLAB_PANIC, NULL, NULL);

	iocontext_cachep = kmem_cache_create("blkdev_ioc",
			sizeof(struct io_context), 0, SLAB_PANIC, NULL, NULL);

	blk_max_low_pfn = max_low_pfn;
	blk_max_pfn = max_pfn;

	return 0;
}

/*
 * IO Context helper functions
 */
void put_io_context(struct io_context *ioc)
{
	if (ioc == NULL)
		return;

	BUG_ON(atomic_read(&ioc->refcount) == 0);

	if (atomic_dec_and_test(&ioc->refcount)) {
		if (ioc->aic && ioc->aic->dtor)
			ioc->aic->dtor(ioc->aic);
		if (ioc->cic && ioc->cic->dtor)
			ioc->cic->dtor(ioc->cic);

		kmem_cache_free(iocontext_cachep, ioc);
	}
}
EXPORT_SYMBOL(put_io_context);

/* Called by the exitting task */
void exit_io_context(void)
{
	unsigned long flags;
	struct io_context *ioc;

	local_irq_save(flags);
3300
	task_lock(current);
L
Linus Torvalds 已提交
3301 3302
	ioc = current->io_context;
	current->io_context = NULL;
3303 3304
	ioc->task = NULL;
	task_unlock(current);
L
Linus Torvalds 已提交
3305 3306 3307 3308 3309 3310 3311 3312 3313 3314 3315 3316
	local_irq_restore(flags);

	if (ioc->aic && ioc->aic->exit)
		ioc->aic->exit(ioc->aic);
	if (ioc->cic && ioc->cic->exit)
		ioc->cic->exit(ioc->cic);

	put_io_context(ioc);
}

/*
 * If the current task has no IO context then create one and initialise it.
N
Nick Piggin 已提交
3317
 * Otherwise, return its existing IO context.
L
Linus Torvalds 已提交
3318
 *
N
Nick Piggin 已提交
3319 3320 3321
 * This returned IO context doesn't have a specifically elevated refcount,
 * but since the current task itself holds a reference, the context can be
 * used in general code, so long as it stays within `current` context.
L
Linus Torvalds 已提交
3322
 */
A
Al Viro 已提交
3323
struct io_context *current_io_context(gfp_t gfp_flags)
L
Linus Torvalds 已提交
3324 3325 3326 3327 3328
{
	struct task_struct *tsk = current;
	struct io_context *ret;

	ret = tsk->io_context;
N
Nick Piggin 已提交
3329 3330
	if (likely(ret))
		return ret;
L
Linus Torvalds 已提交
3331 3332 3333 3334

	ret = kmem_cache_alloc(iocontext_cachep, gfp_flags);
	if (ret) {
		atomic_set(&ret->refcount, 1);
3335 3336
		ret->task = current;
		ret->set_ioprio = NULL;
L
Linus Torvalds 已提交
3337 3338 3339 3340
		ret->last_waited = jiffies; /* doesn't matter... */
		ret->nr_batch_requests = 0; /* because this is 0 */
		ret->aic = NULL;
		ret->cic = NULL;
N
Nick Piggin 已提交
3341 3342
		tsk->io_context = ret;
	}
L
Linus Torvalds 已提交
3343

N
Nick Piggin 已提交
3344 3345 3346
	return ret;
}
EXPORT_SYMBOL(current_io_context);
L
Linus Torvalds 已提交
3347

N
Nick Piggin 已提交
3348 3349 3350 3351 3352 3353
/*
 * If the current task has no IO context then create one and initialise it.
 * If it does have a context, take a ref on it.
 *
 * This is always called in the context of the task which submitted the I/O.
 */
A
Al Viro 已提交
3354
struct io_context *get_io_context(gfp_t gfp_flags)
N
Nick Piggin 已提交
3355 3356 3357 3358
{
	struct io_context *ret;
	ret = current_io_context(gfp_flags);
	if (likely(ret))
L
Linus Torvalds 已提交
3359 3360 3361 3362 3363 3364 3365 3366 3367 3368 3369 3370 3371 3372 3373 3374 3375 3376 3377 3378 3379 3380 3381 3382 3383 3384 3385 3386 3387 3388 3389 3390 3391 3392 3393 3394 3395 3396 3397 3398 3399 3400 3401 3402 3403 3404 3405 3406 3407 3408 3409 3410 3411 3412 3413 3414 3415 3416 3417 3418 3419 3420 3421 3422 3423 3424 3425 3426 3427 3428 3429 3430 3431 3432 3433 3434 3435 3436 3437 3438 3439 3440 3441 3442 3443 3444 3445 3446 3447 3448 3449 3450 3451 3452 3453 3454 3455 3456 3457 3458 3459 3460 3461 3462 3463 3464 3465 3466 3467 3468 3469 3470 3471 3472 3473 3474 3475 3476 3477 3478 3479 3480 3481 3482 3483 3484 3485 3486 3487 3488 3489 3490 3491 3492 3493 3494 3495 3496 3497 3498 3499 3500 3501 3502 3503 3504 3505 3506 3507 3508 3509 3510 3511 3512 3513 3514 3515 3516 3517 3518 3519 3520 3521 3522 3523 3524 3525 3526 3527 3528 3529 3530 3531 3532 3533 3534 3535 3536 3537 3538 3539 3540 3541 3542 3543 3544 3545 3546 3547 3548 3549 3550 3551 3552 3553 3554 3555 3556 3557 3558 3559 3560 3561 3562 3563 3564 3565 3566 3567
		atomic_inc(&ret->refcount);
	return ret;
}
EXPORT_SYMBOL(get_io_context);

void copy_io_context(struct io_context **pdst, struct io_context **psrc)
{
	struct io_context *src = *psrc;
	struct io_context *dst = *pdst;

	if (src) {
		BUG_ON(atomic_read(&src->refcount) == 0);
		atomic_inc(&src->refcount);
		put_io_context(dst);
		*pdst = src;
	}
}
EXPORT_SYMBOL(copy_io_context);

void swap_io_context(struct io_context **ioc1, struct io_context **ioc2)
{
	struct io_context *temp;
	temp = *ioc1;
	*ioc1 = *ioc2;
	*ioc2 = temp;
}
EXPORT_SYMBOL(swap_io_context);

/*
 * sysfs parts below
 */
struct queue_sysfs_entry {
	struct attribute attr;
	ssize_t (*show)(struct request_queue *, char *);
	ssize_t (*store)(struct request_queue *, const char *, size_t);
};

static ssize_t
queue_var_show(unsigned int var, char *page)
{
	return sprintf(page, "%d\n", var);
}

static ssize_t
queue_var_store(unsigned long *var, const char *page, size_t count)
{
	char *p = (char *) page;

	*var = simple_strtoul(p, &p, 10);
	return count;
}

static ssize_t queue_requests_show(struct request_queue *q, char *page)
{
	return queue_var_show(q->nr_requests, (page));
}

static ssize_t
queue_requests_store(struct request_queue *q, const char *page, size_t count)
{
	struct request_list *rl = &q->rq;

	int ret = queue_var_store(&q->nr_requests, page, count);
	if (q->nr_requests < BLKDEV_MIN_RQ)
		q->nr_requests = BLKDEV_MIN_RQ;
	blk_queue_congestion_threshold(q);

	if (rl->count[READ] >= queue_congestion_on_threshold(q))
		set_queue_congested(q, READ);
	else if (rl->count[READ] < queue_congestion_off_threshold(q))
		clear_queue_congested(q, READ);

	if (rl->count[WRITE] >= queue_congestion_on_threshold(q))
		set_queue_congested(q, WRITE);
	else if (rl->count[WRITE] < queue_congestion_off_threshold(q))
		clear_queue_congested(q, WRITE);

	if (rl->count[READ] >= q->nr_requests) {
		blk_set_queue_full(q, READ);
	} else if (rl->count[READ]+1 <= q->nr_requests) {
		blk_clear_queue_full(q, READ);
		wake_up(&rl->wait[READ]);
	}

	if (rl->count[WRITE] >= q->nr_requests) {
		blk_set_queue_full(q, WRITE);
	} else if (rl->count[WRITE]+1 <= q->nr_requests) {
		blk_clear_queue_full(q, WRITE);
		wake_up(&rl->wait[WRITE]);
	}
	return ret;
}

static ssize_t queue_ra_show(struct request_queue *q, char *page)
{
	int ra_kb = q->backing_dev_info.ra_pages << (PAGE_CACHE_SHIFT - 10);

	return queue_var_show(ra_kb, (page));
}

static ssize_t
queue_ra_store(struct request_queue *q, const char *page, size_t count)
{
	unsigned long ra_kb;
	ssize_t ret = queue_var_store(&ra_kb, page, count);

	spin_lock_irq(q->queue_lock);
	if (ra_kb > (q->max_sectors >> 1))
		ra_kb = (q->max_sectors >> 1);

	q->backing_dev_info.ra_pages = ra_kb >> (PAGE_CACHE_SHIFT - 10);
	spin_unlock_irq(q->queue_lock);

	return ret;
}

static ssize_t queue_max_sectors_show(struct request_queue *q, char *page)
{
	int max_sectors_kb = q->max_sectors >> 1;

	return queue_var_show(max_sectors_kb, (page));
}

static ssize_t
queue_max_sectors_store(struct request_queue *q, const char *page, size_t count)
{
	unsigned long max_sectors_kb,
			max_hw_sectors_kb = q->max_hw_sectors >> 1,
			page_kb = 1 << (PAGE_CACHE_SHIFT - 10);
	ssize_t ret = queue_var_store(&max_sectors_kb, page, count);
	int ra_kb;

	if (max_sectors_kb > max_hw_sectors_kb || max_sectors_kb < page_kb)
		return -EINVAL;
	/*
	 * Take the queue lock to update the readahead and max_sectors
	 * values synchronously:
	 */
	spin_lock_irq(q->queue_lock);
	/*
	 * Trim readahead window as well, if necessary:
	 */
	ra_kb = q->backing_dev_info.ra_pages << (PAGE_CACHE_SHIFT - 10);
	if (ra_kb > max_sectors_kb)
		q->backing_dev_info.ra_pages =
				max_sectors_kb >> (PAGE_CACHE_SHIFT - 10);

	q->max_sectors = max_sectors_kb << 1;
	spin_unlock_irq(q->queue_lock);

	return ret;
}

static ssize_t queue_max_hw_sectors_show(struct request_queue *q, char *page)
{
	int max_hw_sectors_kb = q->max_hw_sectors >> 1;

	return queue_var_show(max_hw_sectors_kb, (page));
}


static struct queue_sysfs_entry queue_requests_entry = {
	.attr = {.name = "nr_requests", .mode = S_IRUGO | S_IWUSR },
	.show = queue_requests_show,
	.store = queue_requests_store,
};

static struct queue_sysfs_entry queue_ra_entry = {
	.attr = {.name = "read_ahead_kb", .mode = S_IRUGO | S_IWUSR },
	.show = queue_ra_show,
	.store = queue_ra_store,
};

static struct queue_sysfs_entry queue_max_sectors_entry = {
	.attr = {.name = "max_sectors_kb", .mode = S_IRUGO | S_IWUSR },
	.show = queue_max_sectors_show,
	.store = queue_max_sectors_store,
};

static struct queue_sysfs_entry queue_max_hw_sectors_entry = {
	.attr = {.name = "max_hw_sectors_kb", .mode = S_IRUGO },
	.show = queue_max_hw_sectors_show,
};

static struct queue_sysfs_entry queue_iosched_entry = {
	.attr = {.name = "scheduler", .mode = S_IRUGO | S_IWUSR },
	.show = elv_iosched_show,
	.store = elv_iosched_store,
};

static struct attribute *default_attrs[] = {
	&queue_requests_entry.attr,
	&queue_ra_entry.attr,
	&queue_max_hw_sectors_entry.attr,
	&queue_max_sectors_entry.attr,
	&queue_iosched_entry.attr,
	NULL,
};

#define to_queue(atr) container_of((atr), struct queue_sysfs_entry, attr)

static ssize_t
queue_attr_show(struct kobject *kobj, struct attribute *attr, char *page)
{
	struct queue_sysfs_entry *entry = to_queue(attr);
	struct request_queue *q;

	q = container_of(kobj, struct request_queue, kobj);
	if (!entry->show)
3568
		return -EIO;
L
Linus Torvalds 已提交
3569 3570 3571 3572 3573 3574 3575 3576 3577 3578 3579 3580 3581

	return entry->show(q, page);
}

static ssize_t
queue_attr_store(struct kobject *kobj, struct attribute *attr,
		    const char *page, size_t length)
{
	struct queue_sysfs_entry *entry = to_queue(attr);
	struct request_queue *q;

	q = container_of(kobj, struct request_queue, kobj);
	if (!entry->store)
3582
		return -EIO;
L
Linus Torvalds 已提交
3583 3584 3585 3586 3587 3588 3589 3590 3591

	return entry->store(q, page, length);
}

static struct sysfs_ops queue_sysfs_ops = {
	.show	= queue_attr_show,
	.store	= queue_attr_store,
};

3592
static struct kobj_type queue_ktype = {
L
Linus Torvalds 已提交
3593 3594 3595 3596 3597 3598 3599 3600 3601 3602 3603 3604 3605 3606 3607 3608 3609 3610 3611 3612 3613 3614 3615 3616 3617 3618 3619 3620 3621 3622 3623 3624 3625 3626 3627 3628 3629 3630 3631 3632 3633 3634 3635 3636
	.sysfs_ops	= &queue_sysfs_ops,
	.default_attrs	= default_attrs,
};

int blk_register_queue(struct gendisk *disk)
{
	int ret;

	request_queue_t *q = disk->queue;

	if (!q || !q->request_fn)
		return -ENXIO;

	q->kobj.parent = kobject_get(&disk->kobj);
	if (!q->kobj.parent)
		return -EBUSY;

	snprintf(q->kobj.name, KOBJ_NAME_LEN, "%s", "queue");
	q->kobj.ktype = &queue_ktype;

	ret = kobject_register(&q->kobj);
	if (ret < 0)
		return ret;

	ret = elv_register_queue(q);
	if (ret) {
		kobject_unregister(&q->kobj);
		return ret;
	}

	return 0;
}

void blk_unregister_queue(struct gendisk *disk)
{
	request_queue_t *q = disk->queue;

	if (q && q->request_fn) {
		elv_unregister_queue(q);

		kobject_unregister(&q->kobj);
		kobject_put(&disk->kobj);
	}
}