blk-settings.c 28.7 KB
Newer Older
1 2 3 4 5 6 7 8 9
/*
 * Functions related to setting various queue properties from drivers
 */
#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/init.h>
#include <linux/bio.h>
#include <linux/blkdev.h>
#include <linux/bootmem.h>	/* for max_pfn/max_low_pfn */
10
#include <linux/gcd.h>
11
#include <linux/lcm.h>
Randy Dunlap's avatar
Randy Dunlap committed
12
#include <linux/jiffies.h>
13
#include <linux/gfp.h>
14 15

#include "blk.h"
16
#include "blk-wbt.h"
17

18
unsigned long blk_max_low_pfn;
19
EXPORT_SYMBOL(blk_max_low_pfn);
20 21

unsigned long blk_max_pfn;
22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39

/**
 * 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(struct request_queue *q, prep_rq_fn *pfn)
{
	q->prep_rq_fn = pfn;
}
EXPORT_SYMBOL(blk_queue_prep_rq);

40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56
/**
 * blk_queue_unprep_rq - set an unprepare_request function for queue
 * @q:		queue
 * @ufn:	unprepare_request function
 *
 * It's possible for a queue to register an unprepare_request callback
 * which is invoked before the request is finally completed. The goal
 * of the function is to deallocate any data that was allocated in the
 * prepare_request callback.
 *
 */
void blk_queue_unprep_rq(struct request_queue *q, unprep_rq_fn *ufn)
{
	q->unprep_rq_fn = ufn;
}
EXPORT_SYMBOL(blk_queue_unprep_rq);

57 58 59 60 61 62
void blk_queue_softirq_done(struct request_queue *q, softirq_done_fn *fn)
{
	q->softirq_done_fn = fn;
}
EXPORT_SYMBOL(blk_queue_softirq_done);

63 64 65 66 67 68 69 70
void blk_queue_rq_timeout(struct request_queue *q, unsigned int timeout)
{
	q->rq_timeout = timeout;
}
EXPORT_SYMBOL_GPL(blk_queue_rq_timeout);

void blk_queue_rq_timed_out(struct request_queue *q, rq_timed_out_fn *fn)
{
71
	WARN_ON_ONCE(q->mq_ops);
72 73 74 75
	q->rq_timed_out_fn = fn;
}
EXPORT_SYMBOL_GPL(blk_queue_rq_timed_out);

76 77 78 79 80 81
void blk_queue_lld_busy(struct request_queue *q, lld_busy_fn *fn)
{
	q->lld_busy_fn = fn;
}
EXPORT_SYMBOL_GPL(blk_queue_lld_busy);

82 83
/**
 * blk_set_default_limits - reset limits to default values
84
 * @lim:  the queue_limits structure to reset
85 86
 *
 * Description:
87
 *   Returns a queue_limit struct to its default state.
88 89 90
 */
void blk_set_default_limits(struct queue_limits *lim)
{
91
	lim->max_segments = BLK_MAX_SEGMENTS;
92
	lim->max_discard_segments = 1;
93
	lim->max_integrity_segments = 0;
94
	lim->seg_boundary_mask = BLK_SEG_BOUNDARY_MASK;
95
	lim->virt_boundary_mask = 0;
96
	lim->max_segment_size = BLK_MAX_SEGMENT_SIZE;
97 98
	lim->max_sectors = lim->max_hw_sectors = BLK_SAFE_MAX_SECTORS;
	lim->max_dev_sectors = 0;
99
	lim->chunk_sectors = 0;
100
	lim->max_write_same_sectors = 0;
101
	lim->max_write_zeroes_sectors = 0;
102
	lim->max_discard_sectors = 0;
103
	lim->max_hw_discard_sectors = 0;
104 105 106
	lim->discard_granularity = 0;
	lim->discard_alignment = 0;
	lim->discard_misaligned = 0;
107
	lim->logical_block_size = lim->physical_block_size = lim->io_min = 512;
108
	lim->bounce_pfn = (unsigned long)(BLK_BOUNCE_ANY >> PAGE_SHIFT);
109 110 111
	lim->alignment_offset = 0;
	lim->io_opt = 0;
	lim->misaligned = 0;
112
	lim->cluster = 1;
113
	lim->zoned = BLK_ZONED_NONE;
114 115 116
}
EXPORT_SYMBOL(blk_set_default_limits);

117 118 119 120 121 122 123 124 125 126 127 128 129 130
/**
 * blk_set_stacking_limits - set default limits for stacking devices
 * @lim:  the queue_limits structure to reset
 *
 * Description:
 *   Returns a queue_limit struct to its default state. Should be used
 *   by stacking drivers like DM that have no internal limits.
 */
void blk_set_stacking_limits(struct queue_limits *lim)
{
	blk_set_default_limits(lim);

	/* Inherit limits from component devices */
	lim->max_segments = USHRT_MAX;
131
	lim->max_discard_segments = 1;
132
	lim->max_hw_sectors = UINT_MAX;
133
	lim->max_segment_size = UINT_MAX;
134
	lim->max_sectors = UINT_MAX;
135
	lim->max_dev_sectors = UINT_MAX;
136
	lim->max_write_same_sectors = UINT_MAX;
137
	lim->max_write_zeroes_sectors = UINT_MAX;
138 139 140
}
EXPORT_SYMBOL(blk_set_stacking_limits);

141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162
/**
 * 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.
 **/
163
void blk_queue_make_request(struct request_queue *q, make_request_fn *mfn)
164 165 166 167 168
{
	/*
	 * set defaults
	 */
	q->nr_requests = BLKDEV_MAX_RQ;
169

170 171 172 173 174
	q->make_request_fn = mfn;
	blk_queue_dma_alignment(q, 511);
	blk_queue_congestion_threshold(q);
	q->nr_batching = BLK_BATCH_REQ;

175
	blk_set_default_limits(&q->limits);
176 177 178 179 180
}
EXPORT_SYMBOL(blk_queue_make_request);

/**
 * blk_queue_bounce_limit - set bounce buffer limit for queue
181
 * @q: the request queue for the device
182
 * @max_addr: the maximum address the device can handle
183 184 185 186 187
 *
 * 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
188
 *    buffers for doing I/O to pages residing above @max_addr.
189
 **/
190
void blk_queue_bounce_limit(struct request_queue *q, u64 max_addr)
191
{
192
	unsigned long b_pfn = max_addr >> PAGE_SHIFT;
193 194 195 196
	int dma = 0;

	q->bounce_gfp = GFP_NOIO;
#if BITS_PER_LONG == 64
197 198 199 200 201 202
	/*
	 * Assume anything <= 4GB can be handled by IOMMU.  Actually
	 * some IOMMUs can handle everything, but I don't know of a
	 * way to test this here.
	 */
	if (b_pfn < (min_t(u64, 0xffffffffUL, BLK_BOUNCE_HIGH) >> PAGE_SHIFT))
203
		dma = 1;
204
	q->limits.bounce_pfn = max(max_low_pfn, b_pfn);
205
#else
206
	if (b_pfn < blk_max_low_pfn)
207
		dma = 1;
208
	q->limits.bounce_pfn = b_pfn;
209
#endif
210 211 212
	if (dma) {
		init_emergency_isa_pool();
		q->bounce_gfp = GFP_NOIO | GFP_DMA;
213
		q->limits.bounce_pfn = b_pfn;
214 215 216 217 218
	}
}
EXPORT_SYMBOL(blk_queue_bounce_limit);

/**
219 220
 * blk_queue_max_hw_sectors - set max sectors for a request for this queue
 * @q:  the request queue for the device
221
 * @max_hw_sectors:  max hardware sectors in the usual 512b unit
222 223
 *
 * Description:
224 225
 *    Enables a low level driver to set a hard upper limit,
 *    max_hw_sectors, on the size of requests.  max_hw_sectors is set by
226 227
 *    the device driver based upon the capabilities of the I/O
 *    controller.
228
 *
229 230 231
 *    max_dev_sectors is a hard limit imposed by the storage device for
 *    READ/WRITE requests. It is set by the disk driver.
 *
232 233 234 235
 *    max_sectors is a soft limit imposed by the block layer for
 *    filesystem type requests.  This value can be overridden on a
 *    per-device basis in /sys/block/<device>/queue/max_sectors_kb.
 *    The soft limit can not exceed max_hw_sectors.
236
 **/
237
void blk_queue_max_hw_sectors(struct request_queue *q, unsigned int max_hw_sectors)
238
{
239 240 241
	struct queue_limits *limits = &q->limits;
	unsigned int max_sectors;

242 243
	if ((max_hw_sectors << 9) < PAGE_SIZE) {
		max_hw_sectors = 1 << (PAGE_SHIFT - 9);
244
		printk(KERN_INFO "%s: set to minimum %d\n",
245
		       __func__, max_hw_sectors);
246 247
	}

248
	limits->max_hw_sectors = max_hw_sectors;
249 250 251
	max_sectors = min_not_zero(max_hw_sectors, limits->max_dev_sectors);
	max_sectors = min_t(unsigned int, max_sectors, BLK_DEF_MAX_SECTORS);
	limits->max_sectors = max_sectors;
252
	q->backing_dev_info->io_pages = max_sectors >> (PAGE_SHIFT - 9);
253
}
254
EXPORT_SYMBOL(blk_queue_max_hw_sectors);
255

256 257 258 259 260 261 262 263
/**
 * blk_queue_chunk_sectors - set size of the chunk for this queue
 * @q:  the request queue for the device
 * @chunk_sectors:  chunk sectors in the usual 512b unit
 *
 * Description:
 *    If a driver doesn't want IOs to cross a given chunk size, it can set
 *    this limit and prevent merging across chunks. Note that the chunk size
264 265 266 267
 *    must currently be a power-of-2 in sectors. Also note that the block
 *    layer must accept a page worth of data at any offset. So if the
 *    crossing of chunks is a hard limitation in the driver, it must still be
 *    prepared to split single page bios.
268 269 270 271 272 273 274 275
 **/
void blk_queue_chunk_sectors(struct request_queue *q, unsigned int chunk_sectors)
{
	BUG_ON(!is_power_of_2(chunk_sectors));
	q->limits.chunk_sectors = chunk_sectors;
}
EXPORT_SYMBOL(blk_queue_chunk_sectors);

276 277 278
/**
 * blk_queue_max_discard_sectors - set max sectors for a single discard
 * @q:  the request queue for the device
279
 * @max_discard_sectors: maximum number of sectors to discard
280 281 282 283
 **/
void blk_queue_max_discard_sectors(struct request_queue *q,
		unsigned int max_discard_sectors)
{
284
	q->limits.max_hw_discard_sectors = max_discard_sectors;
285 286 287 288
	q->limits.max_discard_sectors = max_discard_sectors;
}
EXPORT_SYMBOL(blk_queue_max_discard_sectors);

289 290 291 292 293 294 295 296 297 298 299 300
/**
 * blk_queue_max_write_same_sectors - set max sectors for a single write same
 * @q:  the request queue for the device
 * @max_write_same_sectors: maximum number of sectors to write per command
 **/
void blk_queue_max_write_same_sectors(struct request_queue *q,
				      unsigned int max_write_same_sectors)
{
	q->limits.max_write_same_sectors = max_write_same_sectors;
}
EXPORT_SYMBOL(blk_queue_max_write_same_sectors);

301 302 303 304 305 306 307 308 309 310 311 312 313
/**
 * blk_queue_max_write_zeroes_sectors - set max sectors for a single
 *                                      write zeroes
 * @q:  the request queue for the device
 * @max_write_zeroes_sectors: maximum number of sectors to write per command
 **/
void blk_queue_max_write_zeroes_sectors(struct request_queue *q,
		unsigned int max_write_zeroes_sectors)
{
	q->limits.max_write_zeroes_sectors = max_write_zeroes_sectors;
}
EXPORT_SYMBOL(blk_queue_max_write_zeroes_sectors);

314
/**
315
 * blk_queue_max_segments - set max hw segments for a request for this queue
316 317 318 319 320
 * @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
321
 *    hw data segments in a request.
322
 **/
323
void blk_queue_max_segments(struct request_queue *q, unsigned short max_segments)
324 325 326
{
	if (!max_segments) {
		max_segments = 1;
327 328
		printk(KERN_INFO "%s: set to minimum %d\n",
		       __func__, max_segments);
329 330
	}

331
	q->limits.max_segments = max_segments;
332
}
333
EXPORT_SYMBOL(blk_queue_max_segments);
334

335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350
/**
 * blk_queue_max_discard_segments - set max segments for discard requests
 * @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
 *    segments in a discard request.
 **/
void blk_queue_max_discard_segments(struct request_queue *q,
		unsigned short max_segments)
{
	q->limits.max_discard_segments = max_segments;
}
EXPORT_SYMBOL_GPL(blk_queue_max_discard_segments);

351 352 353 354 355 356 357 358 359 360 361
/**
 * 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(struct request_queue *q, unsigned int max_size)
{
362 363
	if (max_size < PAGE_SIZE) {
		max_size = PAGE_SIZE;
364 365
		printk(KERN_INFO "%s: set to minimum %d\n",
		       __func__, max_size);
366 367
	}

368
	q->limits.max_segment_size = max_size;
369 370 371 372
}
EXPORT_SYMBOL(blk_queue_max_segment_size);

/**
373
 * blk_queue_logical_block_size - set logical block size for the queue
374
 * @q:  the request queue for the device
375
 * @size:  the logical block size, in bytes
376 377
 *
 * Description:
378 379 380
 *   This should be set to the lowest possible block size that the
 *   storage device can address.  The default of 512 covers most
 *   hardware.
381
 **/
382
void blk_queue_logical_block_size(struct request_queue *q, unsigned short size)
383
{
384
	q->limits.logical_block_size = size;
385 386 387 388 389 390

	if (q->limits.physical_block_size < size)
		q->limits.physical_block_size = size;

	if (q->limits.io_min < q->limits.physical_block_size)
		q->limits.io_min = q->limits.physical_block_size;
391
}
392
EXPORT_SYMBOL(blk_queue_logical_block_size);
393

394 395 396 397 398 399 400 401 402 403
/**
 * blk_queue_physical_block_size - set physical block size for the queue
 * @q:  the request queue for the device
 * @size:  the physical block size, in bytes
 *
 * Description:
 *   This should be set to the lowest possible sector size that the
 *   hardware can operate on without reverting to read-modify-write
 *   operations.
 */
404
void blk_queue_physical_block_size(struct request_queue *q, unsigned int size)
405 406 407 408 409 410 411 412 413 414 415 416 417 418
{
	q->limits.physical_block_size = size;

	if (q->limits.physical_block_size < q->limits.logical_block_size)
		q->limits.physical_block_size = q->limits.logical_block_size;

	if (q->limits.io_min < q->limits.physical_block_size)
		q->limits.io_min = q->limits.physical_block_size;
}
EXPORT_SYMBOL(blk_queue_physical_block_size);

/**
 * blk_queue_alignment_offset - set physical block alignment offset
 * @q:	the request queue for the device
419
 * @offset: alignment offset in bytes
420 421 422 423 424 425 426 427 428 429 430 431 432 433 434
 *
 * Description:
 *   Some devices are naturally misaligned to compensate for things like
 *   the legacy DOS partition table 63-sector offset.  Low-level drivers
 *   should call this function for devices whose first sector is not
 *   naturally aligned.
 */
void blk_queue_alignment_offset(struct request_queue *q, unsigned int offset)
{
	q->limits.alignment_offset =
		offset & (q->limits.physical_block_size - 1);
	q->limits.misaligned = 0;
}
EXPORT_SYMBOL(blk_queue_alignment_offset);

435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457
/**
 * blk_limits_io_min - set minimum request size for a device
 * @limits: the queue limits
 * @min:  smallest I/O size in bytes
 *
 * Description:
 *   Some devices have an internal block size bigger than the reported
 *   hardware sector size.  This function can be used to signal the
 *   smallest I/O the device can perform without incurring a performance
 *   penalty.
 */
void blk_limits_io_min(struct queue_limits *limits, unsigned int min)
{
	limits->io_min = min;

	if (limits->io_min < limits->logical_block_size)
		limits->io_min = limits->logical_block_size;

	if (limits->io_min < limits->physical_block_size)
		limits->io_min = limits->physical_block_size;
}
EXPORT_SYMBOL(blk_limits_io_min);

458 459 460
/**
 * blk_queue_io_min - set minimum request size for the queue
 * @q:	the request queue for the device
461
 * @min:  smallest I/O size in bytes
462 463
 *
 * Description:
464 465 466 467 468 469 470
 *   Storage devices may report a granularity or preferred minimum I/O
 *   size which is the smallest request the device can perform without
 *   incurring a performance penalty.  For disk drives this is often the
 *   physical block size.  For RAID arrays it is often the stripe chunk
 *   size.  A properly aligned multiple of minimum_io_size is the
 *   preferred request size for workloads where a high number of I/O
 *   operations is desired.
471 472 473
 */
void blk_queue_io_min(struct request_queue *q, unsigned int min)
{
474
	blk_limits_io_min(&q->limits, min);
475 476 477
}
EXPORT_SYMBOL(blk_queue_io_min);

478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496
/**
 * blk_limits_io_opt - set optimal request size for a device
 * @limits: the queue limits
 * @opt:  smallest I/O size in bytes
 *
 * Description:
 *   Storage devices may report an optimal I/O size, which is the
 *   device's preferred unit for sustained I/O.  This is rarely reported
 *   for disk drives.  For RAID arrays it is usually the stripe width or
 *   the internal track size.  A properly aligned multiple of
 *   optimal_io_size is the preferred request size for workloads where
 *   sustained throughput is desired.
 */
void blk_limits_io_opt(struct queue_limits *limits, unsigned int opt)
{
	limits->io_opt = opt;
}
EXPORT_SYMBOL(blk_limits_io_opt);

497 498 499
/**
 * blk_queue_io_opt - set optimal request size for the queue
 * @q:	the request queue for the device
500
 * @opt:  optimal request size in bytes
501 502
 *
 * Description:
503 504 505 506 507 508
 *   Storage devices may report an optimal I/O size, which is the
 *   device's preferred unit for sustained I/O.  This is rarely reported
 *   for disk drives.  For RAID arrays it is usually the stripe width or
 *   the internal track size.  A properly aligned multiple of
 *   optimal_io_size is the preferred request size for workloads where
 *   sustained throughput is desired.
509 510 511
 */
void blk_queue_io_opt(struct request_queue *q, unsigned int opt)
{
512
	blk_limits_io_opt(&q->limits, opt);
513 514 515
}
EXPORT_SYMBOL(blk_queue_io_opt);

516 517 518 519 520 521 522
/**
 * 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(struct request_queue *t, struct request_queue *b)
{
523
	blk_stack_limits(&t->limits, &b->limits, 0);
524 525 526
}
EXPORT_SYMBOL(blk_queue_stack_limits);

527 528
/**
 * blk_stack_limits - adjust queue_limits for stacked devices
529 530
 * @t:	the stacking driver limits (top device)
 * @b:  the underlying queue limits (bottom, component device)
531
 * @start:  first data sector within component device
532 533
 *
 * Description:
534 535 536 537 538 539 540 541 542 543 544 545 546
 *    This function is used by stacking drivers like MD and DM to ensure
 *    that all component devices have compatible block sizes and
 *    alignments.  The stacking driver must provide a queue_limits
 *    struct (top) and then iteratively call the stacking function for
 *    all component (bottom) devices.  The stacking function will
 *    attempt to combine the values and ensure proper alignment.
 *
 *    Returns 0 if the top and bottom queue_limits are compatible.  The
 *    top device's block sizes and alignment offsets may be adjusted to
 *    ensure alignment with the bottom device. If no compatible sizes
 *    and alignments exist, -1 is returned and the resulting top
 *    queue_limits will have the misaligned flag set to indicate that
 *    the alignment_offset is undefined.
547 548
 */
int blk_stack_limits(struct queue_limits *t, struct queue_limits *b,
549
		     sector_t start)
550
{
551
	unsigned int top, bottom, alignment, ret = 0;
552

553 554
	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);
555
	t->max_dev_sectors = min_not_zero(t->max_dev_sectors, b->max_dev_sectors);
556 557
	t->max_write_same_sectors = min(t->max_write_same_sectors,
					b->max_write_same_sectors);
558 559
	t->max_write_zeroes_sectors = min(t->max_write_zeroes_sectors,
					b->max_write_zeroes_sectors);
560
	t->bounce_pfn = min_not_zero(t->bounce_pfn, b->bounce_pfn);
561 562 563

	t->seg_boundary_mask = min_not_zero(t->seg_boundary_mask,
					    b->seg_boundary_mask);
564 565
	t->virt_boundary_mask = min_not_zero(t->virt_boundary_mask,
					    b->virt_boundary_mask);
566

567
	t->max_segments = min_not_zero(t->max_segments, b->max_segments);
568 569
	t->max_discard_segments = min_not_zero(t->max_discard_segments,
					       b->max_discard_segments);
570 571
	t->max_integrity_segments = min_not_zero(t->max_integrity_segments,
						 b->max_integrity_segments);
572 573 574 575

	t->max_segment_size = min_not_zero(t->max_segment_size,
					   b->max_segment_size);

576 577
	t->misaligned |= b->misaligned;

578
	alignment = queue_limit_alignment_offset(b, start);
579

580 581 582
	/* Bottom device has different alignment.  Check that it is
	 * compatible with the current top alignment.
	 */
583 584 585 586
	if (t->alignment_offset != alignment) {

		top = max(t->physical_block_size, t->io_min)
			+ t->alignment_offset;
587
		bottom = max(b->physical_block_size, b->io_min) + alignment;
588

589
		/* Verify that top and bottom intervals line up */
590
		if (max(top, bottom) % min(top, bottom)) {
591
			t->misaligned = 1;
592 593
			ret = -1;
		}
594 595
	}

596 597 598 599 600 601 602
	t->logical_block_size = max(t->logical_block_size,
				    b->logical_block_size);

	t->physical_block_size = max(t->physical_block_size,
				     b->physical_block_size);

	t->io_min = max(t->io_min, b->io_min);
603
	t->io_opt = lcm_not_zero(t->io_opt, b->io_opt);
604

605
	t->cluster &= b->cluster;
606

607
	/* Physical block size a multiple of the logical block size? */
608 609
	if (t->physical_block_size & (t->logical_block_size - 1)) {
		t->physical_block_size = t->logical_block_size;
610
		t->misaligned = 1;
611
		ret = -1;
612 613
	}

614
	/* Minimum I/O a multiple of the physical block size? */
615 616 617
	if (t->io_min & (t->physical_block_size - 1)) {
		t->io_min = t->physical_block_size;
		t->misaligned = 1;
618
		ret = -1;
619 620
	}

621
	/* Optimal I/O a multiple of the physical block size? */
622 623 624
	if (t->io_opt & (t->physical_block_size - 1)) {
		t->io_opt = 0;
		t->misaligned = 1;
625
		ret = -1;
626
	}
627

628 629 630 631
	t->raid_partial_stripes_expensive =
		max(t->raid_partial_stripes_expensive,
		    b->raid_partial_stripes_expensive);

632
	/* Find lowest common alignment_offset */
633
	t->alignment_offset = lcm_not_zero(t->alignment_offset, alignment)
634
		% max(t->physical_block_size, t->io_min);
635

636
	/* Verify that new alignment_offset is on a logical block boundary */
637
	if (t->alignment_offset & (t->logical_block_size - 1)) {
638
		t->misaligned = 1;
639 640
		ret = -1;
	}
641

642 643
	/* Discard alignment and granularity */
	if (b->discard_granularity) {
644
		alignment = queue_limit_discard_alignment(b, start);
645 646 647 648 649

		if (t->discard_granularity != 0 &&
		    t->discard_alignment != alignment) {
			top = t->discard_granularity + t->discard_alignment;
			bottom = b->discard_granularity + alignment;
650

651
			/* Verify that top and bottom intervals line up */
652
			if ((max(top, bottom) % min(top, bottom)) != 0)
653 654 655
				t->discard_misaligned = 1;
		}

656 657
		t->max_discard_sectors = min_not_zero(t->max_discard_sectors,
						      b->max_discard_sectors);
658 659
		t->max_hw_discard_sectors = min_not_zero(t->max_hw_discard_sectors,
							 b->max_hw_discard_sectors);
660 661
		t->discard_granularity = max(t->discard_granularity,
					     b->discard_granularity);
662
		t->discard_alignment = lcm_not_zero(t->discard_alignment, alignment) %
663
			t->discard_granularity;
664
	}
665

666 667 668 669
	if (b->chunk_sectors)
		t->chunk_sectors = min_not_zero(t->chunk_sectors,
						b->chunk_sectors);

670
	return ret;
671
}
672
EXPORT_SYMBOL(blk_stack_limits);
673

674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691
/**
 * bdev_stack_limits - adjust queue limits for stacked drivers
 * @t:	the stacking driver limits (top device)
 * @bdev:  the component block_device (bottom)
 * @start:  first data sector within component device
 *
 * Description:
 *    Merges queue limits for a top device and a block_device.  Returns
 *    0 if alignment didn't change.  Returns -1 if adding the bottom
 *    device caused misalignment.
 */
int bdev_stack_limits(struct queue_limits *t, struct block_device *bdev,
		      sector_t start)
{
	struct request_queue *bq = bdev_get_queue(bdev);

	start += get_start_sect(bdev);

692
	return blk_stack_limits(t, &bq->limits, start);
693 694 695
}
EXPORT_SYMBOL(bdev_stack_limits);

696 697
/**
 * disk_stack_limits - adjust queue limits for stacked drivers
698
 * @disk:  MD/DM gendisk (top)
699 700 701 702
 * @bdev:  the underlying block device (bottom)
 * @offset:  offset to beginning of data within component device
 *
 * Description:
703 704
 *    Merges the limits for a top level gendisk and a bottom level
 *    block_device.
705 706 707 708 709 710
 */
void disk_stack_limits(struct gendisk *disk, struct block_device *bdev,
		       sector_t offset)
{
	struct request_queue *t = disk->queue;

711
	if (bdev_stack_limits(&t->limits, bdev, offset >> 9) < 0) {
712 713 714 715 716 717 718 719 720 721 722
		char top[BDEVNAME_SIZE], bottom[BDEVNAME_SIZE];

		disk_name(disk, 0, top);
		bdevname(bdev, bottom);

		printk(KERN_NOTICE "%s: Warning: Device %s is misaligned\n",
		       top, bottom);
	}
}
EXPORT_SYMBOL(disk_stack_limits);

723 724 725 726 727
/**
 * blk_queue_dma_pad - set pad mask
 * @q:     the request queue for the device
 * @mask:  pad mask
 *
728
 * Set dma pad mask.
729
 *
730 731
 * Appending pad buffer to a request modifies the last entry of a
 * scatter list such that it includes the pad buffer.
732 733 734 735 736 737 738
 **/
void blk_queue_dma_pad(struct request_queue *q, unsigned int mask)
{
	q->dma_pad_mask = mask;
}
EXPORT_SYMBOL(blk_queue_dma_pad);

739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755
/**
 * blk_queue_update_dma_pad - update pad mask
 * @q:     the request queue for the device
 * @mask:  pad mask
 *
 * Update dma pad mask.
 *
 * Appending pad buffer to a request modifies the last entry of a
 * scatter list such that it includes the pad buffer.
 **/
void blk_queue_update_dma_pad(struct request_queue *q, unsigned int mask)
{
	if (mask > q->dma_pad_mask)
		q->dma_pad_mask = mask;
}
EXPORT_SYMBOL(blk_queue_update_dma_pad);

756 757 758
/**
 * blk_queue_dma_drain - Set up a drain buffer for excess dma.
 * @q:  the request queue for the device
759
 * @dma_drain_needed: fn which returns non-zero if drain is necessary
760 761 762 763 764 765 766 767 768 769 770 771
 * @buf:	physically contiguous buffer
 * @size:	size of the buffer in bytes
 *
 * Some devices have excess DMA problems and can't simply discard (or
 * zero fill) the unwanted piece of the transfer.  They have to have a
 * real area of memory to transfer it into.  The use case for this is
 * ATAPI devices in DMA mode.  If the packet command causes a transfer
 * bigger than the transfer size some HBAs will lock up if there
 * aren't DMA elements to contain the excess transfer.  What this API
 * does is adjust the queue so that the buf is always appended
 * silently to the scatterlist.
 *
772 773 774 775
 * Note: This routine adjusts max_hw_segments to make room for appending
 * the drain buffer.  If you call blk_queue_max_segments() after calling
 * this routine, you must set the limit to one fewer than your device
 * can support otherwise there won't be room for the drain buffer.
776
 */
777
int blk_queue_dma_drain(struct request_queue *q,
778 779
			       dma_drain_needed_fn *dma_drain_needed,
			       void *buf, unsigned int size)
780
{
781
	if (queue_max_segments(q) < 2)
782 783
		return -EINVAL;
	/* make room for appending the drain */
784
	blk_queue_max_segments(q, queue_max_segments(q) - 1);
785
	q->dma_drain_needed = dma_drain_needed;
786 787 788 789 790 791 792 793 794 795 796 797 798 799
	q->dma_drain_buffer = buf;
	q->dma_drain_size = size;

	return 0;
}
EXPORT_SYMBOL_GPL(blk_queue_dma_drain);

/**
 * 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(struct request_queue *q, unsigned long mask)
{
800 801
	if (mask < PAGE_SIZE - 1) {
		mask = PAGE_SIZE - 1;
802 803
		printk(KERN_INFO "%s: set to minimum %lx\n",
		       __func__, mask);
804 805
	}

806
	q->limits.seg_boundary_mask = mask;
807 808 809
}
EXPORT_SYMBOL(blk_queue_segment_boundary);

810 811 812 813 814 815 816 817 818 819 820
/**
 * blk_queue_virt_boundary - set boundary rules for bio merging
 * @q:  the request queue for the device
 * @mask:  the memory boundary mask
 **/
void blk_queue_virt_boundary(struct request_queue *q, unsigned long mask)
{
	q->limits.virt_boundary_mask = mask;
}
EXPORT_SYMBOL(blk_queue_virt_boundary);

821 822 823 824 825 826
/**
 * blk_queue_dma_alignment - set dma length and memory alignment
 * @q:     the request queue for the device
 * @mask:  alignment mask
 *
 * description:
827
 *    set required memory and length alignment for direct dma transactions.
Alan Cox's avatar
Alan Cox committed
828
 *    this is used when building direct io requests for the queue.
829 830 831 832 833 834 835 836 837 838 839 840 841 842
 *
 **/
void blk_queue_dma_alignment(struct request_queue *q, int mask)
{
	q->dma_alignment = mask;
}
EXPORT_SYMBOL(blk_queue_dma_alignment);

/**
 * blk_queue_update_dma_alignment - update dma length and memory alignment
 * @q:     the request queue for the device
 * @mask:  alignment mask
 *
 * description:
843
 *    update required memory and length alignment for direct dma transactions.
844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859
 *    If the requested alignment is larger than the current alignment, then
 *    the current queue alignment is updated to the new value, otherwise it
 *    is left alone.  The design of this is to allow multiple objects
 *    (driver, device, transport etc) to set their respective
 *    alignments without having them interfere.
 *
 **/
void blk_queue_update_dma_alignment(struct request_queue *q, int mask)
{
	BUG_ON(mask > PAGE_SIZE);

	if (mask > q->dma_alignment)
		q->dma_alignment = mask;
}
EXPORT_SYMBOL(blk_queue_update_dma_alignment);

860 861
void blk_queue_flush_queueable(struct request_queue *q, bool queueable)
{
Jens Axboe's avatar
Jens Axboe committed
862 863 864 865 866 867
	spin_lock_irq(q->queue_lock);
	if (queueable)
		clear_bit(QUEUE_FLAG_FLUSH_NQ, &q->queue_flags);
	else
		set_bit(QUEUE_FLAG_FLUSH_NQ, &q->queue_flags);
	spin_unlock_irq(q->queue_lock);
868 869 870
}
EXPORT_SYMBOL_GPL(blk_queue_flush_queueable);

871 872 873 874 875 876 877 878 879
/**
 * blk_set_queue_depth - tell the block layer about the device queue depth
 * @q:		the request queue for the device
 * @depth:		queue depth
 *
 */
void blk_set_queue_depth(struct request_queue *q, unsigned int depth)
{
	q->queue_depth = depth;
880
	wbt_set_queue_depth(q->rq_wb, depth);
881 882 883
}
EXPORT_SYMBOL(blk_set_queue_depth);

884 885 886 887 888 889 890 891 892 893 894
/**
 * blk_queue_write_cache - configure queue's write cache
 * @q:		the request queue for the device
 * @wc:		write back cache on or off
 * @fua:	device supports FUA writes, if true
 *
 * Tell the block layer about the write cache of @q.
 */
void blk_queue_write_cache(struct request_queue *q, bool wc, bool fua)
{
	spin_lock_irq(q->queue_lock);
Jens Axboe's avatar
Jens Axboe committed
895
	if (wc)
896
		queue_flag_set(QUEUE_FLAG_WC, q);
Jens Axboe's avatar
Jens Axboe committed
897
	else
898
		queue_flag_clear(QUEUE_FLAG_WC, q);
Jens Axboe's avatar
Jens Axboe committed
899
	if (fua)
900
		queue_flag_set(QUEUE_FLAG_FUA, q);
Jens Axboe's avatar
Jens Axboe committed
901
	else
902 903
		queue_flag_clear(QUEUE_FLAG_FUA, q);
	spin_unlock_irq(q->queue_lock);
904 905

	wbt_set_write_cache(q->rq_wb, test_bit(QUEUE_FLAG_WC, &q->queue_flags));
906 907 908
}
EXPORT_SYMBOL_GPL(blk_queue_write_cache);

909
static int __init blk_settings_init(void)
910 911 912 913 914 915
{
	blk_max_low_pfn = max_low_pfn - 1;
	blk_max_pfn = max_pfn - 1;
	return 0;
}
subsys_initcall(blk_settings_init);