blk-settings.c 23.5 KB
Newer Older
J
Jens Axboe 已提交
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>
R
Randy Dunlap 已提交
11
#include <linux/jiffies.h>
J
Jens Axboe 已提交
12 13 14

#include "blk.h"

15
unsigned long blk_max_low_pfn;
J
Jens Axboe 已提交
16
EXPORT_SYMBOL(blk_max_low_pfn);
17 18

unsigned long blk_max_pfn;
J
Jens Axboe 已提交
19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 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

/**
 * 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);

/**
 * 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(struct request_queue *q, merge_bvec_fn *mbfn)
{
	q->merge_bvec_fn = mbfn;
}
EXPORT_SYMBOL(blk_queue_merge_bvec);

void blk_queue_softirq_done(struct request_queue *q, softirq_done_fn *fn)
{
	q->softirq_done_fn = fn;
}
EXPORT_SYMBOL(blk_queue_softirq_done);

J
Jens Axboe 已提交
65 66 67 68 69 70 71 72 73 74 75 76
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)
{
	q->rq_timed_out_fn = fn;
}
EXPORT_SYMBOL_GPL(blk_queue_rq_timed_out);

77 78 79 80 81 82
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);

83 84
/**
 * blk_set_default_limits - reset limits to default values
85
 * @lim:  the queue_limits structure to reset
86 87 88 89 90 91 92 93 94 95 96 97
 *
 * Description:
 *   Returns a queue_limit struct to its default state.  Can be used by
 *   stacking drivers like DM that stage table swaps and reuse an
 *   existing device queue.
 */
void blk_set_default_limits(struct queue_limits *lim)
{
	lim->max_phys_segments = MAX_PHYS_SEGMENTS;
	lim->max_hw_segments = MAX_HW_SEGMENTS;
	lim->seg_boundary_mask = BLK_SEG_BOUNDARY_MASK;
	lim->max_segment_size = MAX_SEGMENT_SIZE;
98 99
	lim->max_sectors = BLK_DEF_MAX_SECTORS;
	lim->max_hw_sectors = INT_MAX;
100 101 102 103
	lim->max_discard_sectors = 0;
	lim->discard_granularity = 0;
	lim->discard_alignment = 0;
	lim->discard_misaligned = 0;
104
	lim->discard_zeroes_data = -1;
105
	lim->logical_block_size = lim->physical_block_size = lim->io_min = 512;
106
	lim->bounce_pfn = (unsigned long)(BLK_BOUNCE_ANY >> PAGE_SHIFT);
107 108 109 110 111 112 113
	lim->alignment_offset = 0;
	lim->io_opt = 0;
	lim->misaligned = 0;
	lim->no_cluster = 0;
}
EXPORT_SYMBOL(blk_set_default_limits);

J
Jens Axboe 已提交
114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135
/**
 * 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.
 **/
136
void blk_queue_make_request(struct request_queue *q, make_request_fn *mfn)
J
Jens Axboe 已提交
137 138 139 140 141
{
	/*
	 * set defaults
	 */
	q->nr_requests = BLKDEV_MAX_RQ;
142

J
Jens Axboe 已提交
143 144 145 146 147 148
	q->make_request_fn = mfn;
	blk_queue_dma_alignment(q, 511);
	blk_queue_congestion_threshold(q);
	q->nr_batching = BLK_BATCH_REQ;

	q->unplug_thresh = 4;		/* hmm */
R
Randy Dunlap 已提交
149
	q->unplug_delay = msecs_to_jiffies(3);	/* 3 milliseconds */
J
Jens Axboe 已提交
150 151 152 153 154 155
	if (q->unplug_delay == 0)
		q->unplug_delay = 1;

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

156
	blk_set_default_limits(&q->limits);
157
	blk_queue_max_sectors(q, SAFE_MAX_SECTORS);
158

159 160 161 162 163 164 165
	/*
	 * If the caller didn't supply a lock, fall back to our embedded
	 * per-queue locks
	 */
	if (!q->queue_lock)
		q->queue_lock = &q->__queue_lock;

J
Jens Axboe 已提交
166 167 168 169 170 171 172 173 174
	/*
	 * by default assume old behaviour and bounce for any highmem page
	 */
	blk_queue_bounce_limit(q, BLK_BOUNCE_HIGH);
}
EXPORT_SYMBOL(blk_queue_make_request);

/**
 * blk_queue_bounce_limit - set bounce buffer limit for queue
175 176
 * @q: the request queue for the device
 * @dma_mask: the maximum address the device can handle
J
Jens Axboe 已提交
177 178 179 180 181
 *
 * 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
182
 *    buffers for doing I/O to pages residing above @dma_mask.
J
Jens Axboe 已提交
183
 **/
184
void blk_queue_bounce_limit(struct request_queue *q, u64 dma_mask)
J
Jens Axboe 已提交
185
{
186
	unsigned long b_pfn = dma_mask >> PAGE_SHIFT;
J
Jens Axboe 已提交
187 188 189 190
	int dma = 0;

	q->bounce_gfp = GFP_NOIO;
#if BITS_PER_LONG == 64
191 192 193 194 195 196
	/*
	 * 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))
J
Jens Axboe 已提交
197
		dma = 1;
198
	q->limits.bounce_pfn = max_low_pfn;
J
Jens Axboe 已提交
199
#else
200
	if (b_pfn < blk_max_low_pfn)
J
Jens Axboe 已提交
201
		dma = 1;
202
	q->limits.bounce_pfn = b_pfn;
J
Jens Axboe 已提交
203 204 205 206
#endif
	if (dma) {
		init_emergency_isa_pool();
		q->bounce_gfp = GFP_NOIO | GFP_DMA;
207
		q->limits.bounce_pfn = b_pfn;
J
Jens Axboe 已提交
208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224
	}
}
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(struct request_queue *q, unsigned int max_sectors)
{
	if ((max_sectors << 9) < PAGE_CACHE_SIZE) {
		max_sectors = 1 << (PAGE_CACHE_SHIFT - 9);
225 226
		printk(KERN_INFO "%s: set to minimum %d\n",
		       __func__, max_sectors);
J
Jens Axboe 已提交
227 228 229
	}

	if (BLK_DEF_MAX_SECTORS > max_sectors)
230
		q->limits.max_hw_sectors = q->limits.max_sectors = max_sectors;
J
Jens Axboe 已提交
231
	else {
232 233
		q->limits.max_sectors = BLK_DEF_MAX_SECTORS;
		q->limits.max_hw_sectors = max_sectors;
J
Jens Axboe 已提交
234 235 236 237
	}
}
EXPORT_SYMBOL(blk_queue_max_sectors);

238 239 240
void blk_queue_max_hw_sectors(struct request_queue *q, unsigned int max_sectors)
{
	if (BLK_DEF_MAX_SECTORS > max_sectors)
241
		q->limits.max_hw_sectors = BLK_DEF_MAX_SECTORS;
242
	else
243
		q->limits.max_hw_sectors = max_sectors;
244 245 246
}
EXPORT_SYMBOL(blk_queue_max_hw_sectors);

247 248 249
/**
 * blk_queue_max_discard_sectors - set max sectors for a single discard
 * @q:  the request queue for the device
250
 * @max_discard_sectors: maximum number of sectors to discard
251 252 253 254 255 256 257 258
 **/
void blk_queue_max_discard_sectors(struct request_queue *q,
		unsigned int max_discard_sectors)
{
	q->limits.max_discard_sectors = max_discard_sectors;
}
EXPORT_SYMBOL(blk_queue_max_discard_sectors);

J
Jens Axboe 已提交
259 260 261 262 263 264 265 266 267 268 269 270 271 272 273
/**
 * 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(struct request_queue *q,
				 unsigned short max_segments)
{
	if (!max_segments) {
		max_segments = 1;
274 275
		printk(KERN_INFO "%s: set to minimum %d\n",
		       __func__, max_segments);
J
Jens Axboe 已提交
276 277
	}

278
	q->limits.max_phys_segments = max_segments;
J
Jens Axboe 已提交
279 280 281 282 283 284 285 286 287 288 289
}
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
290
 *    address/length pairs the host adapter can actually give at once
J
Jens Axboe 已提交
291 292 293 294 295 296 297
 *    to the device.
 **/
void blk_queue_max_hw_segments(struct request_queue *q,
			       unsigned short max_segments)
{
	if (!max_segments) {
		max_segments = 1;
298 299
		printk(KERN_INFO "%s: set to minimum %d\n",
		       __func__, max_segments);
J
Jens Axboe 已提交
300 301
	}

302
	q->limits.max_hw_segments = max_segments;
J
Jens Axboe 已提交
303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318
}
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(struct request_queue *q, unsigned int max_size)
{
	if (max_size < PAGE_CACHE_SIZE) {
		max_size = PAGE_CACHE_SIZE;
319 320
		printk(KERN_INFO "%s: set to minimum %d\n",
		       __func__, max_size);
J
Jens Axboe 已提交
321 322
	}

323
	q->limits.max_segment_size = max_size;
J
Jens Axboe 已提交
324 325 326 327
}
EXPORT_SYMBOL(blk_queue_max_segment_size);

/**
328
 * blk_queue_logical_block_size - set logical block size for the queue
J
Jens Axboe 已提交
329
 * @q:  the request queue for the device
330
 * @size:  the logical block size, in bytes
J
Jens Axboe 已提交
331 332
 *
 * Description:
333 334 335
 *   This should be set to the lowest possible block size that the
 *   storage device can address.  The default of 512 covers most
 *   hardware.
J
Jens Axboe 已提交
336
 **/
337
void blk_queue_logical_block_size(struct request_queue *q, unsigned short size)
J
Jens Axboe 已提交
338
{
339
	q->limits.logical_block_size = size;
340 341 342 343 344 345

	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;
J
Jens Axboe 已提交
346
}
347
EXPORT_SYMBOL(blk_queue_logical_block_size);
J
Jens Axboe 已提交
348

349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373
/**
 * 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.
 */
void blk_queue_physical_block_size(struct request_queue *q, unsigned short size)
{
	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
374
 * @offset: alignment offset in bytes
375 376 377 378 379 380 381 382 383 384 385 386 387 388 389
 *
 * 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);

390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412
/**
 * 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);

413 414 415
/**
 * blk_queue_io_min - set minimum request size for the queue
 * @q:	the request queue for the device
416
 * @min:  smallest I/O size in bytes
417 418
 *
 * Description:
419 420 421 422 423 424 425
 *   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.
426 427 428
 */
void blk_queue_io_min(struct request_queue *q, unsigned int min)
{
429
	blk_limits_io_min(&q->limits, min);
430 431 432
}
EXPORT_SYMBOL(blk_queue_io_min);

433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451
/**
 * 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);

452 453 454
/**
 * blk_queue_io_opt - set optimal request size for the queue
 * @q:	the request queue for the device
455
 * @opt:  optimal request size in bytes
456 457
 *
 * Description:
458 459 460 461 462 463
 *   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.
464 465 466
 */
void blk_queue_io_opt(struct request_queue *q, unsigned int opt)
{
467
	blk_limits_io_opt(&q->limits, opt);
468 469 470
}
EXPORT_SYMBOL(blk_queue_io_opt);

J
Jens Axboe 已提交
471 472 473 474 475 476 477 478 479 480 481 482
/*
 * 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(struct request_queue *t, struct request_queue *b)
{
483
	blk_stack_limits(&t->limits, &b->limits, 0);
484

485 486 487 488 489
	if (!t->queue_lock)
		WARN_ON_ONCE(1);
	else if (!test_bit(QUEUE_FLAG_CLUSTER, &b->queue_flags)) {
		unsigned long flags;
		spin_lock_irqsave(t->queue_lock, flags);
N
Nick Piggin 已提交
490
		queue_flag_clear(QUEUE_FLAG_CLUSTER, t);
491 492
		spin_unlock_irqrestore(t->queue_lock, flags);
	}
J
Jens Axboe 已提交
493 494 495
}
EXPORT_SYMBOL(blk_queue_stack_limits);

496 497 498 499 500 501 502 503 504 505
static unsigned int lcm(unsigned int a, unsigned int b)
{
	if (a && b)
		return (a * b) / gcd(a, b);
	else if (b)
		return b;

	return a;
}

506 507 508
/**
 * blk_stack_limits - adjust queue_limits for stacked devices
 * @t:	the stacking driver limits (top)
509
 * @b:  the underlying queue limits (bottom)
510 511 512 513 514 515 516 517 518 519
 * @offset:  offset to beginning of data within component device
 *
 * Description:
 *    Merges two queue_limit structs.  Returns 0 if alignment didn't
 *    change.  Returns -1 if adding the bottom device caused
 *    misalignment.
 */
int blk_stack_limits(struct queue_limits *t, struct queue_limits *b,
		     sector_t offset)
{
520 521 522 523
	int ret;

	ret = 0;

524 525
	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);
526
	t->bounce_pfn = min_not_zero(t->bounce_pfn, b->bounce_pfn);
527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547

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

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

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

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

	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);
	t->no_cluster |= b->no_cluster;
548
	t->discard_zeroes_data &= b->discard_zeroes_data;
549 550 551 552 553

	/* Bottom device offset aligned? */
	if (offset &&
	    (offset & (b->physical_block_size - 1)) != b->alignment_offset) {
		t->misaligned = 1;
554 555 556
		ret = -1;
	}

557 558 559 560 561 562
	/*
	 * Temporarily disable discard granularity. It's currently buggy
	 * since we default to 0 for discard_granularity, hence this
	 * "failure" will always trigger for non-zero offsets.
	 */
#if 0
563 564 565 566
	if (offset &&
	    (offset & (b->discard_granularity - 1)) != b->discard_alignment) {
		t->discard_misaligned = 1;
		ret = -1;
567
	}
568
#endif
569 570 571 572 573 574

	/* If top has no alignment offset, inherit from bottom */
	if (!t->alignment_offset)
		t->alignment_offset =
			b->alignment_offset & (b->physical_block_size - 1);

575 576 577 578
	if (!t->discard_alignment)
		t->discard_alignment =
			b->discard_alignment & (b->discard_granularity - 1);

579 580 581
	/* Top device aligned on logical block boundary? */
	if (t->alignment_offset & (t->logical_block_size - 1)) {
		t->misaligned = 1;
582
		ret = -1;
583 584
	}

585 586 587 588
	/* Find lcm() of optimal I/O size and granularity */
	t->io_opt = lcm(t->io_opt, b->io_opt);
	t->discard_granularity = lcm(t->discard_granularity,
				     b->discard_granularity);
589 590 591

	/* Verify that optimal I/O size is a multiple of io_min */
	if (t->io_min && t->io_opt % t->io_min)
592
		ret = -1;
593

594
	return ret;
595
}
M
Mike Snitzer 已提交
596
EXPORT_SYMBOL(blk_stack_limits);
597 598 599

/**
 * disk_stack_limits - adjust queue limits for stacked drivers
600
 * @disk:  MD/DM gendisk (top)
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
 * @bdev:  the underlying block device (bottom)
 * @offset:  offset to beginning of data within component device
 *
 * Description:
 *    Merges the limits for two queues.  Returns 0 if alignment
 *    didn't change.  Returns -1 if adding the bottom device caused
 *    misalignment.
 */
void disk_stack_limits(struct gendisk *disk, struct block_device *bdev,
		       sector_t offset)
{
	struct request_queue *t = disk->queue;
	struct request_queue *b = bdev_get_queue(bdev);

	offset += get_start_sect(bdev) << 9;

	if (blk_stack_limits(&t->limits, &b->limits, offset) < 0) {
		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);
	}

	if (!t->queue_lock)
		WARN_ON_ONCE(1);
	else if (!test_bit(QUEUE_FLAG_CLUSTER, &b->queue_flags)) {
		unsigned long flags;

		spin_lock_irqsave(t->queue_lock, flags);
		if (!test_bit(QUEUE_FLAG_CLUSTER, &b->queue_flags))
			queue_flag_clear(QUEUE_FLAG_CLUSTER, t);
		spin_unlock_irqrestore(t->queue_lock, flags);
	}
}
EXPORT_SYMBOL(disk_stack_limits);

640 641 642 643 644
/**
 * blk_queue_dma_pad - set pad mask
 * @q:     the request queue for the device
 * @mask:  pad mask
 *
645
 * Set dma pad mask.
646
 *
647 648
 * Appending pad buffer to a request modifies the last entry of a
 * scatter list such that it includes the pad buffer.
649 650 651 652 653 654 655
 **/
void blk_queue_dma_pad(struct request_queue *q, unsigned int mask)
{
	q->dma_pad_mask = mask;
}
EXPORT_SYMBOL(blk_queue_dma_pad);

656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672
/**
 * 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);

J
Jens Axboe 已提交
673 674 675
/**
 * blk_queue_dma_drain - Set up a drain buffer for excess dma.
 * @q:  the request queue for the device
676
 * @dma_drain_needed: fn which returns non-zero if drain is necessary
J
Jens Axboe 已提交
677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695
 * @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.
 *
 * Note: This routine adjusts max_hw_segments to make room for
 * appending the drain buffer.  If you call
 * blk_queue_max_hw_segments() or blk_queue_max_phys_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.
 */
696
int blk_queue_dma_drain(struct request_queue *q,
697 698
			       dma_drain_needed_fn *dma_drain_needed,
			       void *buf, unsigned int size)
J
Jens Axboe 已提交
699
{
700
	if (queue_max_hw_segments(q) < 2 || queue_max_phys_segments(q) < 2)
J
Jens Axboe 已提交
701 702
		return -EINVAL;
	/* make room for appending the drain */
703 704
	blk_queue_max_hw_segments(q, queue_max_hw_segments(q) - 1);
	blk_queue_max_phys_segments(q, queue_max_phys_segments(q) - 1);
705
	q->dma_drain_needed = dma_drain_needed;
J
Jens Axboe 已提交
706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721
	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)
{
	if (mask < PAGE_CACHE_SIZE - 1) {
		mask = PAGE_CACHE_SIZE - 1;
722 723
		printk(KERN_INFO "%s: set to minimum %lx\n",
		       __func__, mask);
J
Jens Axboe 已提交
724 725
	}

726
	q->limits.seg_boundary_mask = mask;
J
Jens Axboe 已提交
727 728 729 730 731 732 733 734 735
}
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:
736
 *    set required memory and length alignment for direct dma transactions.
A
Alan Cox 已提交
737
 *    this is used when building direct io requests for the queue.
J
Jens Axboe 已提交
738 739 740 741 742 743 744 745 746 747 748 749 750 751
 *
 **/
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:
752
 *    update required memory and length alignment for direct dma transactions.
J
Jens Axboe 已提交
753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768
 *    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);

769
static int __init blk_settings_init(void)
J
Jens Axboe 已提交
770 771 772 773 774 775
{
	blk_max_low_pfn = max_low_pfn - 1;
	blk_max_pfn = max_pfn - 1;
	return 0;
}
subsys_initcall(blk_settings_init);