Blame view
block/blk-settings.c
28.6 KB
86db1e297
|
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 */ |
70dd5bf3b
|
10 |
#include <linux/gcd.h> |
2cda2728a
|
11 |
#include <linux/lcm.h> |
ad5ebd2fa
|
12 |
#include <linux/jiffies.h> |
5a0e3ad6a
|
13 |
#include <linux/gfp.h> |
86db1e297
|
14 15 |
#include "blk.h" |
87760e5ee
|
16 |
#include "blk-wbt.h" |
86db1e297
|
17 |
|
6728cb0e6
|
18 |
unsigned long blk_max_low_pfn; |
86db1e297
|
19 |
EXPORT_SYMBOL(blk_max_low_pfn); |
6728cb0e6
|
20 21 |
unsigned long blk_max_pfn; |
86db1e297
|
22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 |
/** * 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; } |
86db1e297
|
38 39 40 |
EXPORT_SYMBOL(blk_queue_prep_rq); /** |
28018c242
|
41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 |
* 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); |
86db1e297
|
56 57 58 59 |
void blk_queue_softirq_done(struct request_queue *q, softirq_done_fn *fn) { q->softirq_done_fn = fn; } |
86db1e297
|
60 |
EXPORT_SYMBOL(blk_queue_softirq_done); |
242f9dcb8
|
61 62 63 64 65 66 67 68 |
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) { |
130d733a6
|
69 |
WARN_ON_ONCE(q->mq_ops); |
242f9dcb8
|
70 71 72 |
q->rq_timed_out_fn = fn; } EXPORT_SYMBOL_GPL(blk_queue_rq_timed_out); |
ef9e3facd
|
73 74 75 76 77 |
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); |
86db1e297
|
78 |
/** |
e475bba2f
|
79 |
* blk_set_default_limits - reset limits to default values |
f740f5ca0
|
80 |
* @lim: the queue_limits structure to reset |
e475bba2f
|
81 82 |
* * Description: |
b1bd055d3
|
83 |
* Returns a queue_limit struct to its default state. |
e475bba2f
|
84 85 86 |
*/ void blk_set_default_limits(struct queue_limits *lim) { |
8a78362c4
|
87 |
lim->max_segments = BLK_MAX_SEGMENTS; |
1e739730c
|
88 |
lim->max_discard_segments = 1; |
13f05c8d8
|
89 |
lim->max_integrity_segments = 0; |
e475bba2f
|
90 |
lim->seg_boundary_mask = BLK_SEG_BOUNDARY_MASK; |
03100aada
|
91 |
lim->virt_boundary_mask = 0; |
eb28d31bc
|
92 |
lim->max_segment_size = BLK_MAX_SEGMENT_SIZE; |
5f009d3f8
|
93 94 |
lim->max_sectors = lim->max_hw_sectors = BLK_SAFE_MAX_SECTORS; lim->max_dev_sectors = 0; |
762380ad9
|
95 |
lim->chunk_sectors = 0; |
4363ac7c1
|
96 |
lim->max_write_same_sectors = 0; |
a6f0788ec
|
97 |
lim->max_write_zeroes_sectors = 0; |
86b372814
|
98 |
lim->max_discard_sectors = 0; |
0034af036
|
99 |
lim->max_hw_discard_sectors = 0; |
86b372814
|
100 101 102 |
lim->discard_granularity = 0; lim->discard_alignment = 0; lim->discard_misaligned = 0; |
e475bba2f
|
103 |
lim->logical_block_size = lim->physical_block_size = lim->io_min = 512; |
3a02c8e81
|
104 |
lim->bounce_pfn = (unsigned long)(BLK_BOUNCE_ANY >> PAGE_SHIFT); |
e475bba2f
|
105 106 107 |
lim->alignment_offset = 0; lim->io_opt = 0; lim->misaligned = 0; |
e692cb668
|
108 |
lim->cluster = 1; |
797476b88
|
109 |
lim->zoned = BLK_ZONED_NONE; |
e475bba2f
|
110 111 112 113 |
} EXPORT_SYMBOL(blk_set_default_limits); /** |
b1bd055d3
|
114 115 116 117 118 119 120 121 122 123 124 125 |
* 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 */ |
b1bd055d3
|
126 |
lim->max_segments = USHRT_MAX; |
42c9cdfe1
|
127 |
lim->max_discard_segments = USHRT_MAX; |
b1bd055d3
|
128 |
lim->max_hw_sectors = UINT_MAX; |
d82ae52e6
|
129 |
lim->max_segment_size = UINT_MAX; |
fe86cdcef
|
130 |
lim->max_sectors = UINT_MAX; |
ca369d51b
|
131 |
lim->max_dev_sectors = UINT_MAX; |
4363ac7c1
|
132 |
lim->max_write_same_sectors = UINT_MAX; |
a6f0788ec
|
133 |
lim->max_write_zeroes_sectors = UINT_MAX; |
b1bd055d3
|
134 135 136 137 |
} EXPORT_SYMBOL(blk_set_stacking_limits); /** |
86db1e297
|
138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 |
* 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 |
d004a5e7d
|
156 |
* kmap_atomic() to get a temporary kernel mapping, or by calling |
86db1e297
|
157 158 |
* blk_queue_bounce() to create a buffer in normal memory. **/ |
6728cb0e6
|
159 |
void blk_queue_make_request(struct request_queue *q, make_request_fn *mfn) |
86db1e297
|
160 161 162 163 164 |
{ /* * set defaults */ q->nr_requests = BLKDEV_MAX_RQ; |
0e435ac26
|
165 |
|
86db1e297
|
166 |
q->make_request_fn = mfn; |
86db1e297
|
167 168 169 |
blk_queue_dma_alignment(q, 511); blk_queue_congestion_threshold(q); q->nr_batching = BLK_BATCH_REQ; |
e475bba2f
|
170 |
blk_set_default_limits(&q->limits); |
86db1e297
|
171 |
} |
86db1e297
|
172 173 174 175 |
EXPORT_SYMBOL(blk_queue_make_request); /** * blk_queue_bounce_limit - set bounce buffer limit for queue |
cd0aca2d5
|
176 |
* @q: the request queue for the device |
9f7e45d83
|
177 |
* @max_addr: the maximum address the device can handle |
86db1e297
|
178 179 180 181 182 |
* * 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 |
9f7e45d83
|
183 |
* buffers for doing I/O to pages residing above @max_addr. |
86db1e297
|
184 |
**/ |
9f7e45d83
|
185 |
void blk_queue_bounce_limit(struct request_queue *q, u64 max_addr) |
86db1e297
|
186 |
{ |
9f7e45d83
|
187 |
unsigned long b_pfn = max_addr >> PAGE_SHIFT; |
86db1e297
|
188 189 190 191 |
int dma = 0; q->bounce_gfp = GFP_NOIO; #if BITS_PER_LONG == 64 |
cd0aca2d5
|
192 193 194 195 196 197 |
/* * 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)) |
86db1e297
|
198 |
dma = 1; |
efb012b36
|
199 |
q->limits.bounce_pfn = max(max_low_pfn, b_pfn); |
86db1e297
|
200 |
#else |
6728cb0e6
|
201 |
if (b_pfn < blk_max_low_pfn) |
86db1e297
|
202 |
dma = 1; |
c49825fac
|
203 |
q->limits.bounce_pfn = b_pfn; |
260a67a9e
|
204 |
#endif |
86db1e297
|
205 206 207 |
if (dma) { init_emergency_isa_pool(); q->bounce_gfp = GFP_NOIO | GFP_DMA; |
260a67a9e
|
208 |
q->limits.bounce_pfn = b_pfn; |
86db1e297
|
209 210 |
} } |
86db1e297
|
211 212 213 |
EXPORT_SYMBOL(blk_queue_bounce_limit); /** |
ca369d51b
|
214 215 |
* blk_queue_max_hw_sectors - set max sectors for a request for this queue * @q: the request queue for the device |
2800aac11
|
216 |
* @max_hw_sectors: max hardware sectors in the usual 512b unit |
86db1e297
|
217 218 |
* * Description: |
2800aac11
|
219 220 |
* 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 |
4f258a463
|
221 222 |
* the device driver based upon the capabilities of the I/O * controller. |
2800aac11
|
223 |
* |
ca369d51b
|
224 225 226 |
* max_dev_sectors is a hard limit imposed by the storage device for * READ/WRITE requests. It is set by the disk driver. * |
2800aac11
|
227 228 229 230 |
* 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. |
86db1e297
|
231 |
**/ |
ca369d51b
|
232 |
void blk_queue_max_hw_sectors(struct request_queue *q, unsigned int max_hw_sectors) |
86db1e297
|
233 |
{ |
ca369d51b
|
234 235 |
struct queue_limits *limits = &q->limits; unsigned int max_sectors; |
09cbfeaf1
|
236 237 |
if ((max_hw_sectors << 9) < PAGE_SIZE) { max_hw_sectors = 1 << (PAGE_SHIFT - 9); |
24c03d47d
|
238 239 |
printk(KERN_INFO "%s: set to minimum %d ", |
2800aac11
|
240 |
__func__, max_hw_sectors); |
86db1e297
|
241 |
} |
30e2bc08b
|
242 |
limits->max_hw_sectors = max_hw_sectors; |
ca369d51b
|
243 244 245 |
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; |
dc3b17cc8
|
246 |
q->backing_dev_info->io_pages = max_sectors >> (PAGE_SHIFT - 9); |
86db1e297
|
247 |
} |
086fa5ff0
|
248 |
EXPORT_SYMBOL(blk_queue_max_hw_sectors); |
86db1e297
|
249 250 |
/** |
762380ad9
|
251 252 253 254 255 256 257 |
* 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 |
58a4915ad
|
258 259 260 261 |
* 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. |
762380ad9
|
262 263 264 265 266 267 268 269 270 |
**/ 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); /** |
67efc9258
|
271 272 |
* blk_queue_max_discard_sectors - set max sectors for a single discard * @q: the request queue for the device |
c7ebf0657
|
273 |
* @max_discard_sectors: maximum number of sectors to discard |
67efc9258
|
274 275 276 277 |
**/ void blk_queue_max_discard_sectors(struct request_queue *q, unsigned int max_discard_sectors) { |
0034af036
|
278 |
q->limits.max_hw_discard_sectors = max_discard_sectors; |
67efc9258
|
279 280 281 282 283 |
q->limits.max_discard_sectors = max_discard_sectors; } EXPORT_SYMBOL(blk_queue_max_discard_sectors); /** |
4363ac7c1
|
284 285 286 287 288 289 290 291 292 293 294 295 |
* 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); /** |
a6f0788ec
|
296 297 298 299 300 301 302 303 304 305 306 307 308 |
* 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); /** |
8a78362c4
|
309 |
* blk_queue_max_segments - set max hw segments for a request for this queue |
86db1e297
|
310 311 312 313 314 |
* @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 |
8a78362c4
|
315 |
* hw data segments in a request. |
86db1e297
|
316 |
**/ |
8a78362c4
|
317 |
void blk_queue_max_segments(struct request_queue *q, unsigned short max_segments) |
86db1e297
|
318 319 320 |
{ if (!max_segments) { max_segments = 1; |
24c03d47d
|
321 322 323 |
printk(KERN_INFO "%s: set to minimum %d ", __func__, max_segments); |
86db1e297
|
324 |
} |
8a78362c4
|
325 |
q->limits.max_segments = max_segments; |
86db1e297
|
326 |
} |
8a78362c4
|
327 |
EXPORT_SYMBOL(blk_queue_max_segments); |
86db1e297
|
328 329 |
/** |
1e739730c
|
330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 |
* 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); /** |
86db1e297
|
346 347 348 349 350 351 352 353 354 355 |
* 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) { |
09cbfeaf1
|
356 357 |
if (max_size < PAGE_SIZE) { max_size = PAGE_SIZE; |
24c03d47d
|
358 359 360 |
printk(KERN_INFO "%s: set to minimum %d ", __func__, max_size); |
86db1e297
|
361 |
} |
025146e13
|
362 |
q->limits.max_segment_size = max_size; |
86db1e297
|
363 |
} |
86db1e297
|
364 365 366 |
EXPORT_SYMBOL(blk_queue_max_segment_size); /** |
e1defc4ff
|
367 |
* blk_queue_logical_block_size - set logical block size for the queue |
86db1e297
|
368 |
* @q: the request queue for the device |
e1defc4ff
|
369 |
* @size: the logical block size, in bytes |
86db1e297
|
370 371 |
* * Description: |
e1defc4ff
|
372 373 374 |
* This should be set to the lowest possible block size that the * storage device can address. The default of 512 covers most * hardware. |
86db1e297
|
375 |
**/ |
e1defc4ff
|
376 |
void blk_queue_logical_block_size(struct request_queue *q, unsigned short size) |
86db1e297
|
377 |
{ |
025146e13
|
378 |
q->limits.logical_block_size = size; |
c72758f33
|
379 380 381 382 383 384 |
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; |
86db1e297
|
385 |
} |
e1defc4ff
|
386 |
EXPORT_SYMBOL(blk_queue_logical_block_size); |
86db1e297
|
387 |
|
c72758f33
|
388 389 390 391 392 393 394 395 396 397 |
/** * 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. */ |
892b6f90d
|
398 |
void blk_queue_physical_block_size(struct request_queue *q, unsigned int size) |
c72758f33
|
399 400 401 402 403 404 405 406 407 408 409 410 411 412 |
{ 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 |
8ebf97560
|
413 |
* @offset: alignment offset in bytes |
c72758f33
|
414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 |
* * 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); /** |
7c958e326
|
430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 |
* 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); /** |
c72758f33
|
453 454 |
* blk_queue_io_min - set minimum request size for the queue * @q: the request queue for the device |
8ebf97560
|
455 |
* @min: smallest I/O size in bytes |
c72758f33
|
456 457 |
* * Description: |
7e5f5fb09
|
458 459 460 461 462 463 464 |
* 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. |
c72758f33
|
465 466 467 |
*/ void blk_queue_io_min(struct request_queue *q, unsigned int min) { |
7c958e326
|
468 |
blk_limits_io_min(&q->limits, min); |
c72758f33
|
469 470 471 472 |
} EXPORT_SYMBOL(blk_queue_io_min); /** |
3c5820c74
|
473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 |
* 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); /** |
c72758f33
|
492 493 |
* blk_queue_io_opt - set optimal request size for the queue * @q: the request queue for the device |
8ebf97560
|
494 |
* @opt: optimal request size in bytes |
c72758f33
|
495 496 |
* * Description: |
7e5f5fb09
|
497 498 499 500 501 502 |
* 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. |
c72758f33
|
503 504 505 |
*/ void blk_queue_io_opt(struct request_queue *q, unsigned int opt) { |
3c5820c74
|
506 |
blk_limits_io_opt(&q->limits, opt); |
c72758f33
|
507 508 |
} EXPORT_SYMBOL(blk_queue_io_opt); |
86db1e297
|
509 510 511 512 513 514 515 |
/** * 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) { |
fef246672
|
516 |
blk_stack_limits(&t->limits, &b->limits, 0); |
86db1e297
|
517 |
} |
86db1e297
|
518 519 520 |
EXPORT_SYMBOL(blk_queue_stack_limits); /** |
c72758f33
|
521 |
* blk_stack_limits - adjust queue_limits for stacked devices |
81744ee44
|
522 523 |
* @t: the stacking driver limits (top device) * @b: the underlying queue limits (bottom, component device) |
e03a72e13
|
524 |
* @start: first data sector within component device |
c72758f33
|
525 526 |
* * Description: |
81744ee44
|
527 528 529 530 531 532 533 534 535 536 537 538 539 |
* 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. |
c72758f33
|
540 541 |
*/ int blk_stack_limits(struct queue_limits *t, struct queue_limits *b, |
e03a72e13
|
542 |
sector_t start) |
c72758f33
|
543 |
{ |
e03a72e13
|
544 |
unsigned int top, bottom, alignment, ret = 0; |
86b372814
|
545 |
|
c72758f33
|
546 547 |
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); |
ca369d51b
|
548 |
t->max_dev_sectors = min_not_zero(t->max_dev_sectors, b->max_dev_sectors); |
4363ac7c1
|
549 550 |
t->max_write_same_sectors = min(t->max_write_same_sectors, b->max_write_same_sectors); |
a6f0788ec
|
551 552 |
t->max_write_zeroes_sectors = min(t->max_write_zeroes_sectors, b->max_write_zeroes_sectors); |
77634f33d
|
553 |
t->bounce_pfn = min_not_zero(t->bounce_pfn, b->bounce_pfn); |
c72758f33
|
554 555 556 |
t->seg_boundary_mask = min_not_zero(t->seg_boundary_mask, b->seg_boundary_mask); |
03100aada
|
557 558 |
t->virt_boundary_mask = min_not_zero(t->virt_boundary_mask, b->virt_boundary_mask); |
c72758f33
|
559 |
|
8a78362c4
|
560 |
t->max_segments = min_not_zero(t->max_segments, b->max_segments); |
1e739730c
|
561 562 |
t->max_discard_segments = min_not_zero(t->max_discard_segments, b->max_discard_segments); |
13f05c8d8
|
563 564 |
t->max_integrity_segments = min_not_zero(t->max_integrity_segments, b->max_integrity_segments); |
c72758f33
|
565 566 567 |
t->max_segment_size = min_not_zero(t->max_segment_size, b->max_segment_size); |
fe0b393f2
|
568 |
t->misaligned |= b->misaligned; |
e03a72e13
|
569 |
alignment = queue_limit_alignment_offset(b, start); |
9504e0864
|
570 |
|
81744ee44
|
571 572 573 |
/* Bottom device has different alignment. Check that it is * compatible with the current top alignment. */ |
9504e0864
|
574 575 576 577 |
if (t->alignment_offset != alignment) { top = max(t->physical_block_size, t->io_min) + t->alignment_offset; |
81744ee44
|
578 |
bottom = max(b->physical_block_size, b->io_min) + alignment; |
9504e0864
|
579 |
|
81744ee44
|
580 |
/* Verify that top and bottom intervals line up */ |
b8839b8c5
|
581 |
if (max(top, bottom) % min(top, bottom)) { |
9504e0864
|
582 |
t->misaligned = 1; |
fe0b393f2
|
583 584 |
ret = -1; } |
9504e0864
|
585 |
} |
c72758f33
|
586 587 588 589 590 591 592 |
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); |
e9637415a
|
593 |
t->io_opt = lcm_not_zero(t->io_opt, b->io_opt); |
9504e0864
|
594 |
|
e692cb668
|
595 |
t->cluster &= b->cluster; |
c72758f33
|
596 |
|
81744ee44
|
597 |
/* Physical block size a multiple of the logical block size? */ |
9504e0864
|
598 599 |
if (t->physical_block_size & (t->logical_block_size - 1)) { t->physical_block_size = t->logical_block_size; |
c72758f33
|
600 |
t->misaligned = 1; |
fe0b393f2
|
601 |
ret = -1; |
86b372814
|
602 |
} |
81744ee44
|
603 |
/* Minimum I/O a multiple of the physical block size? */ |
9504e0864
|
604 605 606 |
if (t->io_min & (t->physical_block_size - 1)) { t->io_min = t->physical_block_size; t->misaligned = 1; |
fe0b393f2
|
607 |
ret = -1; |
c72758f33
|
608 |
} |
81744ee44
|
609 |
/* Optimal I/O a multiple of the physical block size? */ |
9504e0864
|
610 611 612 |
if (t->io_opt & (t->physical_block_size - 1)) { t->io_opt = 0; t->misaligned = 1; |
fe0b393f2
|
613 |
ret = -1; |
9504e0864
|
614 |
} |
c72758f33
|
615 |
|
c78afc626
|
616 617 618 |
t->raid_partial_stripes_expensive = max(t->raid_partial_stripes_expensive, b->raid_partial_stripes_expensive); |
81744ee44
|
619 |
/* Find lowest common alignment_offset */ |
e9637415a
|
620 |
t->alignment_offset = lcm_not_zero(t->alignment_offset, alignment) |
b8839b8c5
|
621 |
% max(t->physical_block_size, t->io_min); |
86b372814
|
622 |
|
81744ee44
|
623 |
/* Verify that new alignment_offset is on a logical block boundary */ |
fe0b393f2
|
624 |
if (t->alignment_offset & (t->logical_block_size - 1)) { |
c72758f33
|
625 |
t->misaligned = 1; |
fe0b393f2
|
626 627 |
ret = -1; } |
c72758f33
|
628 |
|
9504e0864
|
629 630 |
/* Discard alignment and granularity */ if (b->discard_granularity) { |
e03a72e13
|
631 |
alignment = queue_limit_discard_alignment(b, start); |
9504e0864
|
632 633 634 635 636 |
if (t->discard_granularity != 0 && t->discard_alignment != alignment) { top = t->discard_granularity + t->discard_alignment; bottom = b->discard_granularity + alignment; |
70dd5bf3b
|
637 |
|
9504e0864
|
638 |
/* Verify that top and bottom intervals line up */ |
8dd2cb7e8
|
639 |
if ((max(top, bottom) % min(top, bottom)) != 0) |
9504e0864
|
640 641 |
t->discard_misaligned = 1; } |
81744ee44
|
642 643 |
t->max_discard_sectors = min_not_zero(t->max_discard_sectors, b->max_discard_sectors); |
0034af036
|
644 645 |
t->max_hw_discard_sectors = min_not_zero(t->max_hw_discard_sectors, b->max_hw_discard_sectors); |
9504e0864
|
646 647 |
t->discard_granularity = max(t->discard_granularity, b->discard_granularity); |
e9637415a
|
648 |
t->discard_alignment = lcm_not_zero(t->discard_alignment, alignment) % |
8dd2cb7e8
|
649 |
t->discard_granularity; |
9504e0864
|
650 |
} |
70dd5bf3b
|
651 |
|
987b3b26e
|
652 653 654 |
if (b->chunk_sectors) t->chunk_sectors = min_not_zero(t->chunk_sectors, b->chunk_sectors); |
fe0b393f2
|
655 |
return ret; |
c72758f33
|
656 |
} |
5d85d3247
|
657 |
EXPORT_SYMBOL(blk_stack_limits); |
c72758f33
|
658 659 |
/** |
17be8c245
|
660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 |
* 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); |
e03a72e13
|
676 |
return blk_stack_limits(t, &bq->limits, start); |
17be8c245
|
677 678 679 680 |
} EXPORT_SYMBOL(bdev_stack_limits); /** |
c72758f33
|
681 |
* disk_stack_limits - adjust queue limits for stacked drivers |
77634f33d
|
682 |
* @disk: MD/DM gendisk (top) |
c72758f33
|
683 684 685 686 |
* @bdev: the underlying block device (bottom) * @offset: offset to beginning of data within component device * * Description: |
e03a72e13
|
687 688 |
* Merges the limits for a top level gendisk and a bottom level * block_device. |
c72758f33
|
689 690 691 692 693 |
*/ void disk_stack_limits(struct gendisk *disk, struct block_device *bdev, sector_t offset) { struct request_queue *t = disk->queue; |
c72758f33
|
694 |
|
e03a72e13
|
695 |
if (bdev_stack_limits(&t->limits, bdev, offset >> 9) < 0) { |
c72758f33
|
696 697 698 699 700 701 702 703 704 |
char top[BDEVNAME_SIZE], bottom[BDEVNAME_SIZE]; disk_name(disk, 0, top); bdevname(bdev, bottom); printk(KERN_NOTICE "%s: Warning: Device %s is misaligned ", top, bottom); } |
c72758f33
|
705 706 707 708 |
} EXPORT_SYMBOL(disk_stack_limits); /** |
e3790c7d4
|
709 710 711 712 |
* blk_queue_dma_pad - set pad mask * @q: the request queue for the device * @mask: pad mask * |
27f8221af
|
713 |
* Set dma pad mask. |
e3790c7d4
|
714 |
* |
27f8221af
|
715 716 |
* Appending pad buffer to a request modifies the last entry of a * scatter list such that it includes the pad buffer. |
e3790c7d4
|
717 718 719 720 721 722 723 724 |
**/ void blk_queue_dma_pad(struct request_queue *q, unsigned int mask) { q->dma_pad_mask = mask; } EXPORT_SYMBOL(blk_queue_dma_pad); /** |
27f8221af
|
725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 |
* 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); /** |
86db1e297
|
742 |
* blk_queue_dma_drain - Set up a drain buffer for excess dma. |
86db1e297
|
743 |
* @q: the request queue for the device |
2fb98e841
|
744 |
* @dma_drain_needed: fn which returns non-zero if drain is necessary |
86db1e297
|
745 746 747 748 749 750 751 752 753 754 755 756 |
* @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. * |
8a78362c4
|
757 758 759 760 |
* 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. |
86db1e297
|
761 |
*/ |
448da4d26
|
762 |
int blk_queue_dma_drain(struct request_queue *q, |
2fb98e841
|
763 764 |
dma_drain_needed_fn *dma_drain_needed, void *buf, unsigned int size) |
86db1e297
|
765 |
{ |
8a78362c4
|
766 |
if (queue_max_segments(q) < 2) |
86db1e297
|
767 768 |
return -EINVAL; /* make room for appending the drain */ |
8a78362c4
|
769 |
blk_queue_max_segments(q, queue_max_segments(q) - 1); |
2fb98e841
|
770 |
q->dma_drain_needed = dma_drain_needed; |
86db1e297
|
771 772 773 774 775 |
q->dma_drain_buffer = buf; q->dma_drain_size = size; return 0; } |
86db1e297
|
776 777 778 779 780 781 782 783 784 |
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) { |
09cbfeaf1
|
785 786 |
if (mask < PAGE_SIZE - 1) { mask = PAGE_SIZE - 1; |
24c03d47d
|
787 788 789 |
printk(KERN_INFO "%s: set to minimum %lx ", __func__, mask); |
86db1e297
|
790 |
} |
025146e13
|
791 |
q->limits.seg_boundary_mask = mask; |
86db1e297
|
792 |
} |
86db1e297
|
793 794 795 |
EXPORT_SYMBOL(blk_queue_segment_boundary); /** |
03100aada
|
796 797 798 799 800 801 802 803 804 805 806 |
* 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); /** |
86db1e297
|
807 808 809 810 811 |
* blk_queue_dma_alignment - set dma length and memory alignment * @q: the request queue for the device * @mask: alignment mask * * description: |
710027a48
|
812 |
* set required memory and length alignment for direct dma transactions. |
8feb4d20b
|
813 |
* this is used when building direct io requests for the queue. |
86db1e297
|
814 815 816 817 818 819 |
* **/ void blk_queue_dma_alignment(struct request_queue *q, int mask) { q->dma_alignment = mask; } |
86db1e297
|
820 821 822 823 824 825 826 827 |
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: |
710027a48
|
828 |
* update required memory and length alignment for direct dma transactions. |
86db1e297
|
829 830 831 832 833 834 835 836 837 838 839 840 841 842 |
* 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; } |
86db1e297
|
843 |
EXPORT_SYMBOL(blk_queue_update_dma_alignment); |
f38769309
|
844 845 |
void blk_queue_flush_queueable(struct request_queue *q, bool queueable) { |
c888a8f95
|
846 |
if (queueable) |
8814ce8a0
|
847 |
blk_queue_flag_clear(QUEUE_FLAG_FLUSH_NQ, q); |
c888a8f95
|
848 |
else |
8814ce8a0
|
849 |
blk_queue_flag_set(QUEUE_FLAG_FLUSH_NQ, q); |
f38769309
|
850 851 |
} EXPORT_SYMBOL_GPL(blk_queue_flush_queueable); |
93e9d8e83
|
852 |
/** |
d278d4a88
|
853 854 855 856 857 858 859 860 |
* 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; |
a79050434
|
861 |
wbt_set_queue_depth(q, depth); |
d278d4a88
|
862 863 864 865 |
} EXPORT_SYMBOL(blk_set_queue_depth); /** |
93e9d8e83
|
866 867 868 869 870 871 872 873 874 875 |
* 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); |
c888a8f95
|
876 |
if (wc) |
93e9d8e83
|
877 |
queue_flag_set(QUEUE_FLAG_WC, q); |
c888a8f95
|
878 |
else |
93e9d8e83
|
879 |
queue_flag_clear(QUEUE_FLAG_WC, q); |
c888a8f95
|
880 |
if (fua) |
93e9d8e83
|
881 |
queue_flag_set(QUEUE_FLAG_FUA, q); |
c888a8f95
|
882 |
else |
93e9d8e83
|
883 884 |
queue_flag_clear(QUEUE_FLAG_FUA, q); spin_unlock_irq(q->queue_lock); |
87760e5ee
|
885 |
|
a79050434
|
886 |
wbt_set_write_cache(q, test_bit(QUEUE_FLAG_WC, &q->queue_flags)); |
93e9d8e83
|
887 888 |
} EXPORT_SYMBOL_GPL(blk_queue_write_cache); |
aeb3d3a81
|
889 |
static int __init blk_settings_init(void) |
86db1e297
|
890 891 892 893 894 895 |
{ blk_max_low_pfn = max_low_pfn - 1; blk_max_pfn = max_pfn - 1; return 0; } subsys_initcall(blk_settings_init); |