scsi_lib.c 56.0 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
/*
 *  scsi_lib.c Copyright (C) 1999 Eric Youngdale
 *
 *  SCSI queueing library.
 *      Initial versions: Eric Youngdale (eric@andante.org).
 *                        Based upon conversations with large numbers
 *                        of people at Linux Expo.
 */

#include <linux/bio.h>
#include <linux/blkdev.h>
#include <linux/completion.h>
#include <linux/kernel.h>
#include <linux/mempool.h>
#include <linux/slab.h>
#include <linux/init.h>
#include <linux/pci.h>
#include <linux/delay.h>
19
#include <linux/hardirq.h>
L
Linus Torvalds 已提交
20 21

#include <scsi/scsi.h>
22
#include <scsi/scsi_cmnd.h>
L
Linus Torvalds 已提交
23 24 25 26 27 28 29 30 31 32
#include <scsi/scsi_dbg.h>
#include <scsi/scsi_device.h>
#include <scsi/scsi_driver.h>
#include <scsi/scsi_eh.h>
#include <scsi/scsi_host.h>

#include "scsi_priv.h"
#include "scsi_logging.h"


33
#define SG_MEMPOOL_NR		ARRAY_SIZE(scsi_sg_pools)
34
#define SG_MEMPOOL_SIZE		2
L
Linus Torvalds 已提交
35 36 37 38

struct scsi_host_sg_pool {
	size_t		size;
	char		*name; 
39
	struct kmem_cache	*slab;
L
Linus Torvalds 已提交
40 41 42 43 44 45 46 47
	mempool_t	*pool;
};

#if (SCSI_MAX_PHYS_SEGMENTS < 32)
#error SCSI_MAX_PHYS_SEGMENTS is too small
#endif

#define SP(x) { x, "sgpool-" #x } 
A
Adrian Bunk 已提交
48
static struct scsi_host_sg_pool scsi_sg_pools[] = {
L
Linus Torvalds 已提交
49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66
	SP(8),
	SP(16),
	SP(32),
#if (SCSI_MAX_PHYS_SEGMENTS > 32)
	SP(64),
#if (SCSI_MAX_PHYS_SEGMENTS > 64)
	SP(128),
#if (SCSI_MAX_PHYS_SEGMENTS > 128)
	SP(256),
#if (SCSI_MAX_PHYS_SEGMENTS > 256)
#error SCSI_MAX_PHYS_SEGMENTS is too large
#endif
#endif
#endif
#endif
}; 	
#undef SP

67
static void scsi_run_queue(struct request_queue *q);
68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84

/*
 * Function:	scsi_unprep_request()
 *
 * Purpose:	Remove all preparation done for a request, including its
 *		associated scsi_cmnd, so that it can be requeued.
 *
 * Arguments:	req	- request to unprepare
 *
 * Lock status:	Assumed that no locks are held upon entry.
 *
 * Returns:	Nothing.
 */
static void scsi_unprep_request(struct request *req)
{
	struct scsi_cmnd *cmd = req->special;

85
	req->cmd_flags &= ~REQ_DONTPREP;
86
	req->special = NULL;
87 88 89

	scsi_put_command(cmd);
}
90

L
Linus Torvalds 已提交
91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113
/*
 * Function:    scsi_queue_insert()
 *
 * Purpose:     Insert a command in the midlevel queue.
 *
 * Arguments:   cmd    - command that we are adding to queue.
 *              reason - why we are inserting command to queue.
 *
 * Lock status: Assumed that lock is not held upon entry.
 *
 * Returns:     Nothing.
 *
 * Notes:       We do this for one of two cases.  Either the host is busy
 *              and it cannot accept any more commands for the time being,
 *              or the device returned QUEUE_FULL and can accept no more
 *              commands.
 * Notes:       This could be called either from an interrupt context or a
 *              normal process context.
 */
int scsi_queue_insert(struct scsi_cmnd *cmd, int reason)
{
	struct Scsi_Host *host = cmd->device->host;
	struct scsi_device *device = cmd->device;
114 115
	struct request_queue *q = device->request_queue;
	unsigned long flags;
L
Linus Torvalds 已提交
116 117 118 119 120

	SCSI_LOG_MLQUEUE(1,
		 printk("Inserting command %p into mlqueue\n", cmd));

	/*
121
	 * Set the appropriate busy bit for the device/host.
L
Linus Torvalds 已提交
122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144
	 *
	 * If the host/device isn't busy, assume that something actually
	 * completed, and that we should be able to queue a command now.
	 *
	 * Note that the prior mid-layer assumption that any host could
	 * always queue at least one command is now broken.  The mid-layer
	 * will implement a user specifiable stall (see
	 * scsi_host.max_host_blocked and scsi_device.max_device_blocked)
	 * if a command is requeued with no other commands outstanding
	 * either for the device or for the host.
	 */
	if (reason == SCSI_MLQUEUE_HOST_BUSY)
		host->host_blocked = host->max_host_blocked;
	else if (reason == SCSI_MLQUEUE_DEVICE_BUSY)
		device->device_blocked = device->max_device_blocked;

	/*
	 * Decrement the counters, since these commands are no longer
	 * active on the host/device.
	 */
	scsi_device_unbusy(device);

	/*
145 146
	 * Requeue this command.  It will go before all other commands
	 * that are already in the queue.
L
Linus Torvalds 已提交
147 148 149 150
	 *
	 * NOTE: there is magic here about the way the queue is plugged if
	 * we have no outstanding commands.
	 * 
151
	 * Although we *don't* plug the queue, we call the request
L
Linus Torvalds 已提交
152 153
	 * function.  The SCSI request function detects the blocked condition
	 * and plugs the queue appropriately.
154 155
         */
	spin_lock_irqsave(q->queue_lock, flags);
J
James Bottomley 已提交
156
	blk_requeue_request(q, cmd->request);
157 158 159 160
	spin_unlock_irqrestore(q->queue_lock, flags);

	scsi_run_queue(q);

L
Linus Torvalds 已提交
161 162 163
	return 0;
}

164
/**
165
 * scsi_execute - insert request and wait for the result
166 167 168 169 170 171 172 173
 * @sdev:	scsi device
 * @cmd:	scsi command
 * @data_direction: data direction
 * @buffer:	data buffer
 * @bufflen:	len of buffer
 * @sense:	optional sense buffer
 * @timeout:	request timeout in seconds
 * @retries:	number of times to retry request
174
 * @flags:	or into request flags;
175
 *
176
 * returns the req->errors value which is the scsi_cmnd result
177
 * field.
178
 **/
179 180 181
int scsi_execute(struct scsi_device *sdev, const unsigned char *cmd,
		 int data_direction, void *buffer, unsigned bufflen,
		 unsigned char *sense, int timeout, int retries, int flags)
182 183 184 185 186 187 188 189 190 191 192 193 194 195 196
{
	struct request *req;
	int write = (data_direction == DMA_TO_DEVICE);
	int ret = DRIVER_ERROR << 24;

	req = blk_get_request(sdev->request_queue, write, __GFP_WAIT);

	if (bufflen &&	blk_rq_map_kern(sdev->request_queue, req,
					buffer, bufflen, __GFP_WAIT))
		goto out;

	req->cmd_len = COMMAND_SIZE(cmd[0]);
	memcpy(req->cmd, cmd, req->cmd_len);
	req->sense = sense;
	req->sense_len = 0;
197
	req->retries = retries;
198
	req->timeout = timeout;
199 200
	req->cmd_type = REQ_TYPE_BLOCK_PC;
	req->cmd_flags |= flags | REQ_QUIET | REQ_PREEMPT;
201 202 203 204 205 206 207 208 209 210 211 212

	/*
	 * head injection *required* here otherwise quiesce won't work
	 */
	blk_execute_rq(req->q, NULL, req, 1);

	ret = req->errors;
 out:
	blk_put_request(req);

	return ret;
}
213
EXPORT_SYMBOL(scsi_execute);
214

215 216 217 218 219 220

int scsi_execute_req(struct scsi_device *sdev, const unsigned char *cmd,
		     int data_direction, void *buffer, unsigned bufflen,
		     struct scsi_sense_hdr *sshdr, int timeout, int retries)
{
	char *sense = NULL;
221 222
	int result;
	
223
	if (sshdr) {
J
Jes Sorensen 已提交
224
		sense = kzalloc(SCSI_SENSE_BUFFERSIZE, GFP_NOIO);
225 226 227
		if (!sense)
			return DRIVER_ERROR << 24;
	}
228
	result = scsi_execute(sdev, cmd, data_direction, buffer, bufflen,
J
Jes Sorensen 已提交
229
			      sense, timeout, retries, 0);
230
	if (sshdr)
231
		scsi_normalize_sense(sense, SCSI_SENSE_BUFFERSIZE, sshdr);
232 233 234 235 236 237

	kfree(sense);
	return result;
}
EXPORT_SYMBOL(scsi_execute_req);

238 239 240 241 242 243
struct scsi_io_context {
	void *data;
	void (*done)(void *data, char *sense, int result, int resid);
	char sense[SCSI_SENSE_BUFFERSIZE];
};

244
static struct kmem_cache *scsi_io_context_cache;
245

246
static void scsi_end_async(struct request *req, int uptodate)
247 248 249 250 251 252
{
	struct scsi_io_context *sioc = req->end_io_data;

	if (sioc->done)
		sioc->done(sioc->data, sioc->sense, req->errors, req->data_len);

253
	kmem_cache_free(scsi_io_context_cache, sioc);
254 255 256 257 258 259 260 261 262 263 264 265
	__blk_put_request(req->q, req);
}

static int scsi_merge_bio(struct request *rq, struct bio *bio)
{
	struct request_queue *q = rq->q;

	bio->bi_flags &= ~(1 << BIO_SEG_VALID);
	if (rq_data_dir(rq) == WRITE)
		bio->bi_rw |= (1 << BIO_RW);
	blk_queue_bounce(q, &bio);

N
NeilBrown 已提交
266
	return blk_rq_append_bio(q, rq, bio);
267 268
}

269
static void scsi_bi_endio(struct bio *bio, int error)
270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289
{
	bio_put(bio);
}

/**
 * scsi_req_map_sg - map a scatterlist into a request
 * @rq:		request to fill
 * @sg:		scatterlist
 * @nsegs:	number of elements
 * @bufflen:	len of buffer
 * @gfp:	memory allocation flags
 *
 * scsi_req_map_sg maps a scatterlist into a request so that the
 * request can be sent to the block layer. We do not trust the scatterlist
 * sent to use, as some ULDs use that struct to only organize the pages.
 */
static int scsi_req_map_sg(struct request *rq, struct scatterlist *sgl,
			   int nsegs, unsigned bufflen, gfp_t gfp)
{
	struct request_queue *q = rq->q;
290
	int nr_pages = (bufflen + sgl[0].offset + PAGE_SIZE - 1) >> PAGE_SHIFT;
291
	unsigned int data_len = bufflen, len, bytes, off;
292 293 294 295 296 297 298 299 300
	struct page *page;
	struct bio *bio = NULL;
	int i, err, nr_vecs = 0;

	for (i = 0; i < nsegs; i++) {
		page = sgl[i].page;
		off = sgl[i].offset;
		len = sgl[i].length;

301 302 303 304 305 306
		while (len > 0 && data_len > 0) {
			/*
			 * sg sends a scatterlist that is larger than
			 * the data_len it wants transferred for certain
			 * IO sizes
			 */
307
			bytes = min_t(unsigned int, len, PAGE_SIZE - off);
308
			bytes = min(bytes, data_len);
309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331

			if (!bio) {
				nr_vecs = min_t(int, BIO_MAX_PAGES, nr_pages);
				nr_pages -= nr_vecs;

				bio = bio_alloc(gfp, nr_vecs);
				if (!bio) {
					err = -ENOMEM;
					goto free_bios;
				}
				bio->bi_end_io = scsi_bi_endio;
			}

			if (bio_add_pc_page(q, bio, page, bytes, off) !=
			    bytes) {
				bio_put(bio);
				err = -EINVAL;
				goto free_bios;
			}

			if (bio->bi_vcnt >= nr_vecs) {
				err = scsi_merge_bio(rq, bio);
				if (err) {
332
					bio_endio(bio, 0);
333 334 335 336 337 338 339
					goto free_bios;
				}
				bio = NULL;
			}

			page++;
			len -= bytes;
340
			data_len -=bytes;
341 342 343 344 345
			off = 0;
		}
	}

	rq->buffer = rq->data = NULL;
346
	rq->data_len = bufflen;
347 348 349 350 351 352 353 354
	return 0;

free_bios:
	while ((bio = rq->bio) != NULL) {
		rq->bio = bio->bi_next;
		/*
		 * call endio instead of bio_put incase it was bounced
		 */
355
		bio_endio(bio, 0);
356 357 358 359 360 361 362 363 364
	}

	return err;
}

/**
 * scsi_execute_async - insert request
 * @sdev:	scsi device
 * @cmd:	scsi command
365
 * @cmd_len:	length of scsi cdb
366 367 368 369 370 371 372 373 374
 * @data_direction: data direction
 * @buffer:	data buffer (this can be a kernel buffer or scatterlist)
 * @bufflen:	len of buffer
 * @use_sg:	if buffer is a scatterlist this is the number of elements
 * @timeout:	request timeout in seconds
 * @retries:	number of times to retry request
 * @flags:	or into request flags
 **/
int scsi_execute_async(struct scsi_device *sdev, const unsigned char *cmd,
375
		       int cmd_len, int data_direction, void *buffer, unsigned bufflen,
376 377 378 379 380 381 382 383
		       int use_sg, int timeout, int retries, void *privdata,
		       void (*done)(void *, char *, int, int), gfp_t gfp)
{
	struct request *req;
	struct scsi_io_context *sioc;
	int err = 0;
	int write = (data_direction == DMA_TO_DEVICE);

384
	sioc = kmem_cache_zalloc(scsi_io_context_cache, gfp);
385 386 387 388 389 390
	if (!sioc)
		return DRIVER_ERROR << 24;

	req = blk_get_request(sdev->request_queue, write, gfp);
	if (!req)
		goto free_sense;
391 392
	req->cmd_type = REQ_TYPE_BLOCK_PC;
	req->cmd_flags |= REQ_QUIET;
393 394 395 396 397 398 399 400 401

	if (use_sg)
		err = scsi_req_map_sg(req, buffer, use_sg, bufflen, gfp);
	else if (bufflen)
		err = blk_rq_map_kern(req->q, req, buffer, bufflen, gfp);

	if (err)
		goto free_req;

402
	req->cmd_len = cmd_len;
403
	memset(req->cmd, 0, BLK_MAX_CDB); /* ATAPI hates garbage after CDB */
404 405 406 407
	memcpy(req->cmd, cmd, req->cmd_len);
	req->sense = sioc->sense;
	req->sense_len = 0;
	req->timeout = timeout;
408
	req->retries = retries;
409 410 411 412 413 414 415 416 417 418 419
	req->end_io_data = sioc;

	sioc->data = privdata;
	sioc->done = done;

	blk_execute_rq_nowait(req->q, NULL, req, 1, scsi_end_async);
	return 0;

free_req:
	blk_put_request(req);
free_sense:
420
	kmem_cache_free(scsi_io_context_cache, sioc);
421 422 423 424
	return DRIVER_ERROR << 24;
}
EXPORT_SYMBOL_GPL(scsi_execute_async);

L
Linus Torvalds 已提交
425 426 427 428 429 430 431 432 433 434 435
/*
 * Function:    scsi_init_cmd_errh()
 *
 * Purpose:     Initialize cmd fields related to error handling.
 *
 * Arguments:   cmd	- command that is ready to be queued.
 *
 * Notes:       This function has the job of initializing a number of
 *              fields related to error handling.   Typically this will
 *              be called once for each command, as required.
 */
436
static void scsi_init_cmd_errh(struct scsi_cmnd *cmd)
L
Linus Torvalds 已提交
437 438
{
	cmd->serial_number = 0;
439
	cmd->resid = 0;
L
Linus Torvalds 已提交
440 441 442 443 444 445 446 447 448 449 450 451
	memset(cmd->sense_buffer, 0, sizeof cmd->sense_buffer);
	if (cmd->cmd_len == 0)
		cmd->cmd_len = COMMAND_SIZE(cmd->cmnd[0]);
}

void scsi_device_unbusy(struct scsi_device *sdev)
{
	struct Scsi_Host *shost = sdev->host;
	unsigned long flags;

	spin_lock_irqsave(shost->host_lock, flags);
	shost->host_busy--;
452
	if (unlikely(scsi_host_in_recovery(shost) &&
453
		     (shost->host_failed || shost->host_eh_scheduled)))
L
Linus Torvalds 已提交
454 455
		scsi_eh_wakeup(shost);
	spin_unlock(shost->host_lock);
已提交
456
	spin_lock(sdev->request_queue->queue_lock);
L
Linus Torvalds 已提交
457
	sdev->device_busy--;
已提交
458
	spin_unlock_irqrestore(sdev->request_queue->queue_lock, flags);
L
Linus Torvalds 已提交
459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547
}

/*
 * Called for single_lun devices on IO completion. Clear starget_sdev_user,
 * and call blk_run_queue for all the scsi_devices on the target -
 * including current_sdev first.
 *
 * Called with *no* scsi locks held.
 */
static void scsi_single_lun_run(struct scsi_device *current_sdev)
{
	struct Scsi_Host *shost = current_sdev->host;
	struct scsi_device *sdev, *tmp;
	struct scsi_target *starget = scsi_target(current_sdev);
	unsigned long flags;

	spin_lock_irqsave(shost->host_lock, flags);
	starget->starget_sdev_user = NULL;
	spin_unlock_irqrestore(shost->host_lock, flags);

	/*
	 * Call blk_run_queue for all LUNs on the target, starting with
	 * current_sdev. We race with others (to set starget_sdev_user),
	 * but in most cases, we will be first. Ideally, each LU on the
	 * target would get some limited time or requests on the target.
	 */
	blk_run_queue(current_sdev->request_queue);

	spin_lock_irqsave(shost->host_lock, flags);
	if (starget->starget_sdev_user)
		goto out;
	list_for_each_entry_safe(sdev, tmp, &starget->devices,
			same_target_siblings) {
		if (sdev == current_sdev)
			continue;
		if (scsi_device_get(sdev))
			continue;

		spin_unlock_irqrestore(shost->host_lock, flags);
		blk_run_queue(sdev->request_queue);
		spin_lock_irqsave(shost->host_lock, flags);
	
		scsi_device_put(sdev);
	}
 out:
	spin_unlock_irqrestore(shost->host_lock, flags);
}

/*
 * Function:	scsi_run_queue()
 *
 * Purpose:	Select a proper request queue to serve next
 *
 * Arguments:	q	- last request's queue
 *
 * Returns:     Nothing
 *
 * Notes:	The previous command was completely finished, start
 *		a new one if possible.
 */
static void scsi_run_queue(struct request_queue *q)
{
	struct scsi_device *sdev = q->queuedata;
	struct Scsi_Host *shost = sdev->host;
	unsigned long flags;

	if (sdev->single_lun)
		scsi_single_lun_run(sdev);

	spin_lock_irqsave(shost->host_lock, flags);
	while (!list_empty(&shost->starved_list) &&
	       !shost->host_blocked && !shost->host_self_blocked &&
		!((shost->can_queue > 0) &&
		  (shost->host_busy >= shost->can_queue))) {
		/*
		 * As long as shost is accepting commands and we have
		 * starved queues, call blk_run_queue. scsi_request_fn
		 * drops the queue_lock and can add us back to the
		 * starved_list.
		 *
		 * host_lock protects the starved_list and starved_entry.
		 * scsi_request_fn must get the host_lock before checking
		 * or modifying starved_list or starved_entry.
		 */
		sdev = list_entry(shost->starved_list.next,
					  struct scsi_device, starved_entry);
		list_del_init(&sdev->starved_entry);
		spin_unlock_irqrestore(shost->host_lock, flags);

548 549 550 551 552 553 554 555 556

		if (test_bit(QUEUE_FLAG_REENTER, &q->queue_flags) &&
		    !test_and_set_bit(QUEUE_FLAG_REENTER,
				      &sdev->request_queue->queue_flags)) {
			blk_run_queue(sdev->request_queue);
			clear_bit(QUEUE_FLAG_REENTER,
				  &sdev->request_queue->queue_flags);
		} else
			blk_run_queue(sdev->request_queue);
L
Linus Torvalds 已提交
557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587

		spin_lock_irqsave(shost->host_lock, flags);
		if (unlikely(!list_empty(&sdev->starved_entry)))
			/*
			 * sdev lost a race, and was put back on the
			 * starved list. This is unlikely but without this
			 * in theory we could loop forever.
			 */
			break;
	}
	spin_unlock_irqrestore(shost->host_lock, flags);

	blk_run_queue(q);
}

/*
 * Function:	scsi_requeue_command()
 *
 * Purpose:	Handle post-processing of completed commands.
 *
 * Arguments:	q	- queue to operate on
 *		cmd	- command that may need to be requeued.
 *
 * Returns:	Nothing
 *
 * Notes:	After command completion, there may be blocks left
 *		over which weren't finished by the previous command
 *		this can be for a number of reasons - the main one is
 *		I/O errors in the middle of the request, in which case
 *		we need to request the blocks that come after the bad
 *		sector.
588
 * Notes:	Upon return, cmd is a stale pointer.
L
Linus Torvalds 已提交
589 590 591
 */
static void scsi_requeue_command(struct request_queue *q, struct scsi_cmnd *cmd)
{
592
	struct request *req = cmd->request;
593 594
	unsigned long flags;

595
	scsi_unprep_request(req);
596
	spin_lock_irqsave(q->queue_lock, flags);
597
	blk_requeue_request(q, req);
598
	spin_unlock_irqrestore(q->queue_lock, flags);
L
Linus Torvalds 已提交
599 600 601 602 603 604

	scsi_run_queue(q);
}

void scsi_next_command(struct scsi_cmnd *cmd)
{
605 606 607 608 609
	struct scsi_device *sdev = cmd->device;
	struct request_queue *q = sdev->request_queue;

	/* need to hold a reference on the device before we let go of the cmd */
	get_device(&sdev->sdev_gendev);
L
Linus Torvalds 已提交
610 611 612

	scsi_put_command(cmd);
	scsi_run_queue(q);
613 614 615

	/* ok to remove device now */
	put_device(&sdev->sdev_gendev);
L
Linus Torvalds 已提交
616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638
}

void scsi_run_host_queues(struct Scsi_Host *shost)
{
	struct scsi_device *sdev;

	shost_for_each_device(sdev, shost)
		scsi_run_queue(sdev->request_queue);
}

/*
 * Function:    scsi_end_request()
 *
 * Purpose:     Post-processing of completed commands (usually invoked at end
 *		of upper level post-processing and scsi_io_completion).
 *
 * Arguments:   cmd	 - command that is complete.
 *              uptodate - 1 if I/O indicates success, <= 0 for I/O error.
 *              bytes    - number of bytes of completed I/O
 *		requeue  - indicates whether we should requeue leftovers.
 *
 * Lock status: Assumed that lock is not held upon entry.
 *
639
 * Returns:     cmd if requeue required, NULL otherwise.
L
Linus Torvalds 已提交
640 641 642 643 644 645
 *
 * Notes:       This is called for block device requests in order to
 *              mark some number of sectors as complete.
 * 
 *		We are guaranteeing that the request queue will be goosed
 *		at some point during this call.
646
 * Notes:	If cmd was requeued, upon return it will be a stale pointer.
L
Linus Torvalds 已提交
647 648 649 650
 */
static struct scsi_cmnd *scsi_end_request(struct scsi_cmnd *cmd, int uptodate,
					  int bytes, int requeue)
{
651
	struct request_queue *q = cmd->device->request_queue;
L
Linus Torvalds 已提交
652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668
	struct request *req = cmd->request;
	unsigned long flags;

	/*
	 * If there are blocks left over at the end, set up the command
	 * to queue the remainder of them.
	 */
	if (end_that_request_chunk(req, uptodate, bytes)) {
		int leftover = (req->hard_nr_sectors << 9);

		if (blk_pc_request(req))
			leftover = req->data_len;

		/* kill remainder if no retrys */
		if (!uptodate && blk_noretry_request(req))
			end_that_request_chunk(req, 0, leftover);
		else {
669
			if (requeue) {
L
Linus Torvalds 已提交
670 671 672 673 674 675
				/*
				 * Bleah.  Leftovers again.  Stick the
				 * leftovers in the front of the
				 * queue, and goose the queue again.
				 */
				scsi_requeue_command(q, cmd);
676 677
				cmd = NULL;
			}
L
Linus Torvalds 已提交
678 679 680 681 682 683 684 685 686
			return cmd;
		}
	}

	add_disk_randomness(req->rq_disk);

	spin_lock_irqsave(q->queue_lock, flags);
	if (blk_rq_tagged(req))
		blk_queue_end_tag(q, req);
687
	end_that_request_last(req, uptodate);
L
Linus Torvalds 已提交
688 689 690 691 692 693 694 695 696 697
	spin_unlock_irqrestore(q->queue_lock, flags);

	/*
	 * This will goose the queue request function at the end, so we don't
	 * need to worry about launching another command.
	 */
	scsi_next_command(cmd);
	return NULL;
}

698
struct scatterlist *scsi_alloc_sgtable(struct scsi_cmnd *cmd, gfp_t gfp_mask)
L
Linus Torvalds 已提交
699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738
{
	struct scsi_host_sg_pool *sgp;
	struct scatterlist *sgl;

	BUG_ON(!cmd->use_sg);

	switch (cmd->use_sg) {
	case 1 ... 8:
		cmd->sglist_len = 0;
		break;
	case 9 ... 16:
		cmd->sglist_len = 1;
		break;
	case 17 ... 32:
		cmd->sglist_len = 2;
		break;
#if (SCSI_MAX_PHYS_SEGMENTS > 32)
	case 33 ... 64:
		cmd->sglist_len = 3;
		break;
#if (SCSI_MAX_PHYS_SEGMENTS > 64)
	case 65 ... 128:
		cmd->sglist_len = 4;
		break;
#if (SCSI_MAX_PHYS_SEGMENTS  > 128)
	case 129 ... 256:
		cmd->sglist_len = 5;
		break;
#endif
#endif
#endif
	default:
		return NULL;
	}

	sgp = scsi_sg_pools + cmd->sglist_len;
	sgl = mempool_alloc(sgp->pool, gfp_mask);
	return sgl;
}

739 740 741
EXPORT_SYMBOL(scsi_alloc_sgtable);

void scsi_free_sgtable(struct scatterlist *sgl, int index)
L
Linus Torvalds 已提交
742 743 744
{
	struct scsi_host_sg_pool *sgp;

745
	BUG_ON(index >= SG_MEMPOOL_NR);
L
Linus Torvalds 已提交
746 747 748 749 750

	sgp = scsi_sg_pools + index;
	mempool_free(sgl, sgp->pool);
}

751 752
EXPORT_SYMBOL(scsi_free_sgtable);

L
Linus Torvalds 已提交
753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810
/*
 * Function:    scsi_release_buffers()
 *
 * Purpose:     Completion processing for block device I/O requests.
 *
 * Arguments:   cmd	- command that we are bailing.
 *
 * Lock status: Assumed that no lock is held upon entry.
 *
 * Returns:     Nothing
 *
 * Notes:       In the event that an upper level driver rejects a
 *		command, we must release resources allocated during
 *		the __init_io() function.  Primarily this would involve
 *		the scatter-gather table, and potentially any bounce
 *		buffers.
 */
static void scsi_release_buffers(struct scsi_cmnd *cmd)
{
	if (cmd->use_sg)
		scsi_free_sgtable(cmd->request_buffer, cmd->sglist_len);

	/*
	 * Zero these out.  They now point to freed memory, and it is
	 * dangerous to hang onto the pointers.
	 */
	cmd->request_buffer = NULL;
	cmd->request_bufflen = 0;
}

/*
 * Function:    scsi_io_completion()
 *
 * Purpose:     Completion processing for block device I/O requests.
 *
 * Arguments:   cmd   - command that is finished.
 *
 * Lock status: Assumed that no lock is held upon entry.
 *
 * Returns:     Nothing
 *
 * Notes:       This function is matched in terms of capabilities to
 *              the function that created the scatter-gather list.
 *              In other words, if there are no bounce buffers
 *              (the normal case for most drivers), we don't need
 *              the logic to deal with cleaning up afterwards.
 *
 *		We must do one of several things here:
 *
 *		a) Call scsi_end_request.  This will finish off the
 *		   specified number of sectors.  If we are done, the
 *		   command block will be released, and the queue
 *		   function will be goosed.  If we are not done, then
 *		   scsi_end_request will directly goose the queue.
 *
 *		b) We can just use scsi_requeue_command() here.  This would
 *		   be used if we just wanted to retry, for example.
 */
811
void scsi_io_completion(struct scsi_cmnd *cmd, unsigned int good_bytes)
L
Linus Torvalds 已提交
812 813
{
	int result = cmd->result;
814
	int this_count = cmd->request_bufflen;
815
	struct request_queue *q = cmd->device->request_queue;
L
Linus Torvalds 已提交
816 817 818 819 820 821
	struct request *req = cmd->request;
	int clear_errors = 1;
	struct scsi_sense_hdr sshdr;
	int sense_valid = 0;
	int sense_deferred = 0;

822
	scsi_release_buffers(cmd);
L
Linus Torvalds 已提交
823 824 825 826 827 828

	if (result) {
		sense_valid = scsi_command_normalize_sense(cmd, &sshdr);
		if (sense_valid)
			sense_deferred = scsi_sense_is_deferred(&sshdr);
	}
829

L
Linus Torvalds 已提交
830 831 832 833 834 835 836 837 838 839 840 841 842 843 844
	if (blk_pc_request(req)) { /* SG_IO ioctl from block level */
		req->errors = result;
		if (result) {
			clear_errors = 0;
			if (sense_valid && req->sense) {
				/*
				 * SG_IO wants current and deferred errors
				 */
				int len = 8 + cmd->sense_buffer[7];

				if (len > SCSI_SENSE_BUFFERSIZE)
					len = SCSI_SENSE_BUFFERSIZE;
				memcpy(req->sense, cmd->sense_buffer,  len);
				req->sense_len = len;
			}
845 846
		}
		req->data_len = cmd->resid;
L
Linus Torvalds 已提交
847 848 849 850 851 852
	}

	/*
	 * Next deal with any sectors which we were able to correctly
	 * handle.
	 */
853 854 855 856 857 858 859 860 861 862 863 864 865 866
	SCSI_LOG_HLCOMPLETE(1, printk("%ld sectors total, "
				      "%d bytes done.\n",
				      req->nr_sectors, good_bytes));
	SCSI_LOG_HLCOMPLETE(1, printk("use_sg is %d\n", cmd->use_sg));

	if (clear_errors)
		req->errors = 0;

	/* A number of bytes were successfully read.  If there
	 * are leftovers and there is some kind of error
	 * (result != 0), retry the rest.
	 */
	if (scsi_end_request(cmd, 1, good_bytes, result == 0) == NULL)
		return;
867 868 869

	/* good_bytes = 0, or (inclusive) there were leftovers and
	 * result = 0, so scsi_end_request couldn't retry.
L
Linus Torvalds 已提交
870 871 872 873 874
	 */
	if (sense_valid && !sense_deferred) {
		switch (sshdr.sense_key) {
		case UNIT_ATTENTION:
			if (cmd->device->removable) {
875
				/* Detected disc change.  Set a bit
L
Linus Torvalds 已提交
876 877 878
				 * and quietly refuse further access.
				 */
				cmd->device->changed = 1;
879
				scsi_end_request(cmd, 0, this_count, 1);
L
Linus Torvalds 已提交
880 881
				return;
			} else {
882 883 884 885 886
				/* Must have been a power glitch, or a
				 * bus reset.  Could not have been a
				 * media change, so we just retry the
				 * request and see what happens.
				 */
L
Linus Torvalds 已提交
887 888 889 890 891
				scsi_requeue_command(q, cmd);
				return;
			}
			break;
		case ILLEGAL_REQUEST:
892 893 894 895 896 897 898 899
			/* If we had an ILLEGAL REQUEST returned, then
			 * we may have performed an unsupported
			 * command.  The only thing this should be
			 * would be a ten byte read where only a six
			 * byte read was supported.  Also, on a system
			 * where READ CAPACITY failed, we may have
			 * read past the end of the disk.
			 */
900 901
			if ((cmd->device->use_10_for_rw &&
			    sshdr.asc == 0x20 && sshdr.ascq == 0x00) &&
L
Linus Torvalds 已提交
902 903 904
			    (cmd->cmnd[0] == READ_10 ||
			     cmd->cmnd[0] == WRITE_10)) {
				cmd->device->use_10_for_rw = 0;
905 906
				/* This will cause a retry with a
				 * 6-byte command.
L
Linus Torvalds 已提交
907 908
				 */
				scsi_requeue_command(q, cmd);
909
				return;
L
Linus Torvalds 已提交
910
			} else {
911
				scsi_end_request(cmd, 0, this_count, 1);
L
Linus Torvalds 已提交
912 913 914 915
				return;
			}
			break;
		case NOT_READY:
916
			/* If the device is in the process of becoming
J
James Bottomley 已提交
917
			 * ready, or has a temporary blockage, retry.
L
Linus Torvalds 已提交
918
			 */
J
James Bottomley 已提交
919 920 921 922 923 924 925 926 927 928 929 930 931 932
			if (sshdr.asc == 0x04) {
				switch (sshdr.ascq) {
				case 0x01: /* becoming ready */
				case 0x04: /* format in progress */
				case 0x05: /* rebuild in progress */
				case 0x06: /* recalculation in progress */
				case 0x07: /* operation in progress */
				case 0x08: /* Long write in progress */
				case 0x09: /* self test in progress */
					scsi_requeue_command(q, cmd);
					return;
				default:
					break;
				}
L
Linus Torvalds 已提交
933
			}
934 935 936 937 938
			if (!(req->cmd_flags & REQ_QUIET))
				scsi_cmd_print_sense_hdr(cmd,
							 "Device not ready",
							 &sshdr);

939
			scsi_end_request(cmd, 0, this_count, 1);
L
Linus Torvalds 已提交
940 941
			return;
		case VOLUME_OVERFLOW:
942
			if (!(req->cmd_flags & REQ_QUIET)) {
943
				scmd_printk(KERN_INFO, cmd,
944
					    "Volume overflow, CDB: ");
945
				__scsi_print_command(cmd->cmnd);
946 947
				scsi_print_sense("", cmd);
			}
948 949
			/* See SSC3rXX or current. */
			scsi_end_request(cmd, 0, this_count, 1);
L
Linus Torvalds 已提交
950 951 952 953
			return;
		default:
			break;
		}
954
	}
L
Linus Torvalds 已提交
955
	if (host_byte(result) == DID_RESET) {
956 957 958
		/* Third party bus reset or reset for error recovery
		 * reasons.  Just retry the request and see what
		 * happens.
L
Linus Torvalds 已提交
959 960 961 962 963
		 */
		scsi_requeue_command(q, cmd);
		return;
	}
	if (result) {
964
		if (!(req->cmd_flags & REQ_QUIET)) {
965
			scsi_print_result(cmd);
966 967 968
			if (driver_byte(result) & DRIVER_SENSE)
				scsi_print_sense("", cmd);
		}
L
Linus Torvalds 已提交
969
	}
970
	scsi_end_request(cmd, 0, this_count, !result);
L
Linus Torvalds 已提交
971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990
}

/*
 * Function:    scsi_init_io()
 *
 * Purpose:     SCSI I/O initialize function.
 *
 * Arguments:   cmd   - Command descriptor we wish to initialize
 *
 * Returns:     0 on success
 *		BLKPREP_DEFER if the failure is retryable
 *		BLKPREP_KILL if the failure is fatal
 */
static int scsi_init_io(struct scsi_cmnd *cmd)
{
	struct request     *req = cmd->request;
	struct scatterlist *sgpnt;
	int		   count;

	/*
C
Christoph Hellwig 已提交
991
	 * We used to not use scatter-gather for single segment request,
L
Linus Torvalds 已提交
992 993 994 995 996 997
	 * but now we do (it makes highmem I/O easier to support without
	 * kmapping pages)
	 */
	cmd->use_sg = req->nr_phys_segments;

	/*
C
Christoph Hellwig 已提交
998
	 * If sg table allocation fails, requeue request later.
L
Linus Torvalds 已提交
999 1000
	 */
	sgpnt = scsi_alloc_sgtable(cmd, GFP_ATOMIC);
A
Alan Stern 已提交
1001 1002
	if (unlikely(!sgpnt)) {
		scsi_unprep_request(req);
L
Linus Torvalds 已提交
1003
		return BLKPREP_DEFER;
A
Alan Stern 已提交
1004
	}
L
Linus Torvalds 已提交
1005

C
Christoph Hellwig 已提交
1006
	req->buffer = NULL;
L
Linus Torvalds 已提交
1007 1008 1009
	cmd->request_buffer = (char *) sgpnt;
	if (blk_pc_request(req))
		cmd->request_bufflen = req->data_len;
C
Christoph Hellwig 已提交
1010 1011
	else
		cmd->request_bufflen = req->nr_sectors << 9;
L
Linus Torvalds 已提交
1012 1013 1014 1015 1016 1017 1018 1019

	/* 
	 * Next, walk the list, and fill in the addresses and sizes of
	 * each segment.
	 */
	count = blk_rq_map_sg(req->q, req, cmd->request_buffer);
	if (likely(count <= cmd->use_sg)) {
		cmd->use_sg = count;
C
Christoph Hellwig 已提交
1020
		return BLKPREP_OK;
L
Linus Torvalds 已提交
1021 1022 1023 1024 1025 1026 1027 1028 1029 1030
	}

	printk(KERN_ERR "Incorrect number of segments after building list\n");
	printk(KERN_ERR "counted %d, received %d\n", count, cmd->use_sg);
	printk(KERN_ERR "req nr_sec %lu, cur_nr_sec %u\n", req->nr_sectors,
			req->current_nr_sectors);

	return BLKPREP_KILL;
}

C
Christoph Hellwig 已提交
1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051
static struct scsi_cmnd *scsi_get_cmd_from_req(struct scsi_device *sdev,
		struct request *req)
{
	struct scsi_cmnd *cmd;

	if (!req->special) {
		cmd = scsi_get_command(sdev, GFP_ATOMIC);
		if (unlikely(!cmd))
			return NULL;
		req->special = cmd;
	} else {
		cmd = req->special;
	}

	/* pull a tag out of the request if we have one */
	cmd->tag = req->tag;
	cmd->request = req;

	return cmd;
}

1052
int scsi_setup_blk_pc_cmnd(struct scsi_device *sdev, struct request *req)
J
James Bottomley 已提交
1053
{
C
Christoph Hellwig 已提交
1054
	struct scsi_cmnd *cmd;
1055 1056 1057 1058
	int ret = scsi_prep_state_check(sdev, req);

	if (ret != BLKPREP_OK)
		return ret;
C
Christoph Hellwig 已提交
1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086

	cmd = scsi_get_cmd_from_req(sdev, req);
	if (unlikely(!cmd))
		return BLKPREP_DEFER;

	/*
	 * BLOCK_PC requests may transfer data, in which case they must
	 * a bio attached to them.  Or they might contain a SCSI command
	 * that does not transfer data, in which case they may optionally
	 * submit a request without an attached bio.
	 */
	if (req->bio) {
		int ret;

		BUG_ON(!req->nr_phys_segments);

		ret = scsi_init_io(cmd);
		if (unlikely(ret))
			return ret;
	} else {
		BUG_ON(req->data_len);
		BUG_ON(req->data);

		cmd->request_bufflen = 0;
		cmd->request_buffer = NULL;
		cmd->use_sg = 0;
		req->buffer = NULL;
	}
J
James Bottomley 已提交
1087

1088
	BUILD_BUG_ON(sizeof(req->cmd) > sizeof(cmd->cmnd));
J
James Bottomley 已提交
1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100
	memcpy(cmd->cmnd, req->cmd, sizeof(cmd->cmnd));
	cmd->cmd_len = req->cmd_len;
	if (!req->data_len)
		cmd->sc_data_direction = DMA_NONE;
	else if (rq_data_dir(req) == WRITE)
		cmd->sc_data_direction = DMA_TO_DEVICE;
	else
		cmd->sc_data_direction = DMA_FROM_DEVICE;
	
	cmd->transfersize = req->data_len;
	cmd->allowed = req->retries;
	cmd->timeout_per_command = req->timeout;
C
Christoph Hellwig 已提交
1101
	return BLKPREP_OK;
J
James Bottomley 已提交
1102
}
1103
EXPORT_SYMBOL(scsi_setup_blk_pc_cmnd);
J
James Bottomley 已提交
1104

C
Christoph Hellwig 已提交
1105 1106 1107 1108 1109
/*
 * Setup a REQ_TYPE_FS command.  These are simple read/write request
 * from filesystems that still need to be translated to SCSI CDBs from
 * the ULD.
 */
1110
int scsi_setup_fs_cmnd(struct scsi_device *sdev, struct request *req)
L
Linus Torvalds 已提交
1111 1112
{
	struct scsi_cmnd *cmd;
1113
	int ret = scsi_prep_state_check(sdev, req);
L
Linus Torvalds 已提交
1114

1115 1116
	if (ret != BLKPREP_OK)
		return ret;
L
Linus Torvalds 已提交
1117
	/*
C
Christoph Hellwig 已提交
1118
	 * Filesystem requests must transfer data.
L
Linus Torvalds 已提交
1119
	 */
C
Christoph Hellwig 已提交
1120 1121 1122 1123 1124 1125
	BUG_ON(!req->nr_phys_segments);

	cmd = scsi_get_cmd_from_req(sdev, req);
	if (unlikely(!cmd))
		return BLKPREP_DEFER;

1126
	return scsi_init_io(cmd);
C
Christoph Hellwig 已提交
1127
}
1128
EXPORT_SYMBOL(scsi_setup_fs_cmnd);
C
Christoph Hellwig 已提交
1129

1130
int scsi_prep_state_check(struct scsi_device *sdev, struct request *req)
C
Christoph Hellwig 已提交
1131 1132 1133
{
	int ret = BLKPREP_OK;

L
Linus Torvalds 已提交
1134
	/*
C
Christoph Hellwig 已提交
1135 1136
	 * If the device is not in running state we will reject some
	 * or all commands.
L
Linus Torvalds 已提交
1137
	 */
C
Christoph Hellwig 已提交
1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154
	if (unlikely(sdev->sdev_state != SDEV_RUNNING)) {
		switch (sdev->sdev_state) {
		case SDEV_OFFLINE:
			/*
			 * If the device is offline we refuse to process any
			 * commands.  The device must be brought online
			 * before trying any recovery commands.
			 */
			sdev_printk(KERN_ERR, sdev,
				    "rejecting I/O to offline device\n");
			ret = BLKPREP_KILL;
			break;
		case SDEV_DEL:
			/*
			 * If the device is fully deleted, we refuse to
			 * process any commands as well.
			 */
1155
			sdev_printk(KERN_ERR, sdev,
C
Christoph Hellwig 已提交
1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175
				    "rejecting I/O to dead device\n");
			ret = BLKPREP_KILL;
			break;
		case SDEV_QUIESCE:
		case SDEV_BLOCK:
			/*
			 * If the devices is blocked we defer normal commands.
			 */
			if (!(req->cmd_flags & REQ_PREEMPT))
				ret = BLKPREP_DEFER;
			break;
		default:
			/*
			 * For any other not fully online state we only allow
			 * special commands.  In particular any user initiated
			 * command is not allowed.
			 */
			if (!(req->cmd_flags & REQ_PREEMPT))
				ret = BLKPREP_KILL;
			break;
L
Linus Torvalds 已提交
1176 1177
		}
	}
1178 1179 1180
	return ret;
}
EXPORT_SYMBOL(scsi_prep_state_check);
L
Linus Torvalds 已提交
1181

1182 1183 1184
int scsi_prep_return(struct request_queue *q, struct request *req, int ret)
{
	struct scsi_device *sdev = q->queuedata;
L
Linus Torvalds 已提交
1185

C
Christoph Hellwig 已提交
1186 1187 1188
	switch (ret) {
	case BLKPREP_KILL:
		req->errors = DID_NO_CONNECT << 16;
1189 1190 1191 1192 1193 1194 1195
		/* release the command and kill it */
		if (req->special) {
			struct scsi_cmnd *cmd = req->special;
			scsi_release_buffers(cmd);
			scsi_put_command(cmd);
			req->special = NULL;
		}
C
Christoph Hellwig 已提交
1196 1197
		break;
	case BLKPREP_DEFER:
L
Linus Torvalds 已提交
1198
		/*
C
Christoph Hellwig 已提交
1199 1200 1201
		 * If we defer, the elv_next_request() returns NULL, but the
		 * queue must be restarted, so we plug here if no returning
		 * command will automatically do that.
L
Linus Torvalds 已提交
1202
		 */
C
Christoph Hellwig 已提交
1203 1204 1205 1206 1207
		if (sdev->device_busy == 0)
			blk_plug_device(q);
		break;
	default:
		req->cmd_flags |= REQ_DONTPREP;
L
Linus Torvalds 已提交
1208 1209
	}

C
Christoph Hellwig 已提交
1210
	return ret;
L
Linus Torvalds 已提交
1211
}
1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222
EXPORT_SYMBOL(scsi_prep_return);

static int scsi_prep_fn(struct request_queue *q, struct request *req)
{
	struct scsi_device *sdev = q->queuedata;
	int ret = BLKPREP_KILL;

	if (req->cmd_type == REQ_TYPE_BLOCK_PC)
		ret = scsi_setup_blk_pc_cmnd(sdev, req);
	return scsi_prep_return(q, req, ret);
}
L
Linus Torvalds 已提交
1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240

/*
 * scsi_dev_queue_ready: if we can send requests to sdev, return 1 else
 * return 0.
 *
 * Called with the queue_lock held.
 */
static inline int scsi_dev_queue_ready(struct request_queue *q,
				  struct scsi_device *sdev)
{
	if (sdev->device_busy >= sdev->queue_depth)
		return 0;
	if (sdev->device_busy == 0 && sdev->device_blocked) {
		/*
		 * unblock after device_blocked iterates to zero
		 */
		if (--sdev->device_blocked == 0) {
			SCSI_LOG_MLQUEUE(3,
1241 1242
				   sdev_printk(KERN_INFO, sdev,
				   "unblocking device at zero depth\n"));
L
Linus Torvalds 已提交
1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264
		} else {
			blk_plug_device(q);
			return 0;
		}
	}
	if (sdev->device_blocked)
		return 0;

	return 1;
}

/*
 * scsi_host_queue_ready: if we can send requests to shost, return 1 else
 * return 0. We must end up running the queue again whenever 0 is
 * returned, else IO can hang.
 *
 * Called with host_lock held.
 */
static inline int scsi_host_queue_ready(struct request_queue *q,
				   struct Scsi_Host *shost,
				   struct scsi_device *sdev)
{
1265
	if (scsi_host_in_recovery(shost))
L
Linus Torvalds 已提交
1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294
		return 0;
	if (shost->host_busy == 0 && shost->host_blocked) {
		/*
		 * unblock after host_blocked iterates to zero
		 */
		if (--shost->host_blocked == 0) {
			SCSI_LOG_MLQUEUE(3,
				printk("scsi%d unblocking host at zero depth\n",
					shost->host_no));
		} else {
			blk_plug_device(q);
			return 0;
		}
	}
	if ((shost->can_queue > 0 && shost->host_busy >= shost->can_queue) ||
	    shost->host_blocked || shost->host_self_blocked) {
		if (list_empty(&sdev->starved_entry))
			list_add_tail(&sdev->starved_entry, &shost->starved_list);
		return 0;
	}

	/* We're OK to process the command, so we can't be starved */
	if (!list_empty(&sdev->starved_entry))
		list_del_init(&sdev->starved_entry);

	return 1;
}

/*
1295
 * Kill a request for a dead device
L
Linus Torvalds 已提交
1296
 */
1297
static void scsi_kill_request(struct request *req, struct request_queue *q)
L
Linus Torvalds 已提交
1298
{
1299
	struct scsi_cmnd *cmd = req->special;
1300 1301
	struct scsi_device *sdev = cmd->device;
	struct Scsi_Host *shost = sdev->host;
L
Linus Torvalds 已提交
1302

1303 1304
	blkdev_dequeue_request(req);

1305 1306 1307 1308
	if (unlikely(cmd == NULL)) {
		printk(KERN_CRIT "impossible request in %s.\n",
				 __FUNCTION__);
		BUG();
L
Linus Torvalds 已提交
1309
	}
1310 1311 1312 1313

	scsi_init_cmd_errh(cmd);
	cmd->result = DID_NO_CONNECT << 16;
	atomic_inc(&cmd->device->iorequest_cnt);
1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326

	/*
	 * SCSI request completion path will do scsi_device_unbusy(),
	 * bump busy counts.  To bump the counters, we need to dance
	 * with the locks as normal issue path does.
	 */
	sdev->device_busy++;
	spin_unlock(sdev->request_queue->queue_lock);
	spin_lock(shost->host_lock);
	shost->host_busy++;
	spin_unlock(shost->host_lock);
	spin_lock(sdev->request_queue->queue_lock);

1327
	__scsi_done(cmd);
L
Linus Torvalds 已提交
1328 1329
}

1330 1331 1332
static void scsi_softirq_done(struct request *rq)
{
	struct scsi_cmnd *cmd = rq->completion_data;
1333
	unsigned long wait_for = (cmd->allowed + 1) * cmd->timeout_per_command;
1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353
	int disposition;

	INIT_LIST_HEAD(&cmd->eh_entry);

	disposition = scsi_decide_disposition(cmd);
	if (disposition != SUCCESS &&
	    time_before(cmd->jiffies_at_alloc + wait_for, jiffies)) {
		sdev_printk(KERN_ERR, cmd->device,
			    "timing out command, waited %lus\n",
			    wait_for/HZ);
		disposition = SUCCESS;
	}
			
	scsi_log_completion(cmd, disposition);

	switch (disposition) {
		case SUCCESS:
			scsi_finish_command(cmd);
			break;
		case NEEDS_RETRY:
1354
			scsi_queue_insert(cmd, SCSI_MLQUEUE_EH_RETRY);
1355 1356 1357 1358 1359 1360 1361 1362 1363 1364
			break;
		case ADD_TO_MLQUEUE:
			scsi_queue_insert(cmd, SCSI_MLQUEUE_DEVICE_BUSY);
			break;
		default:
			if (!scsi_eh_scmd_add(cmd, 0))
				scsi_finish_command(cmd);
	}
}

L
Linus Torvalds 已提交
1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384
/*
 * Function:    scsi_request_fn()
 *
 * Purpose:     Main strategy routine for SCSI.
 *
 * Arguments:   q       - Pointer to actual queue.
 *
 * Returns:     Nothing
 *
 * Lock status: IO request lock assumed to be held when called.
 */
static void scsi_request_fn(struct request_queue *q)
{
	struct scsi_device *sdev = q->queuedata;
	struct Scsi_Host *shost;
	struct scsi_cmnd *cmd;
	struct request *req;

	if (!sdev) {
		printk("scsi: killing requests for dead queue\n");
1385 1386
		while ((req = elv_next_request(q)) != NULL)
			scsi_kill_request(req, q);
L
Linus Torvalds 已提交
1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410
		return;
	}

	if(!get_device(&sdev->sdev_gendev))
		/* We must be tearing the block queue down already */
		return;

	/*
	 * To start with, we keep looping until the queue is empty, or until
	 * the host is no longer able to accept any more requests.
	 */
	shost = sdev->host;
	while (!blk_queue_plugged(q)) {
		int rtn;
		/*
		 * get next queueable request.  We do this early to make sure
		 * that the request is fully prepared even if we cannot 
		 * accept it.
		 */
		req = elv_next_request(q);
		if (!req || !scsi_dev_queue_ready(q, sdev))
			break;

		if (unlikely(!scsi_device_online(sdev))) {
1411 1412
			sdev_printk(KERN_ERR, sdev,
				    "rejecting I/O to offline device\n");
1413
			scsi_kill_request(req, q);
L
Linus Torvalds 已提交
1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425
			continue;
		}


		/*
		 * Remove the request from the request list.
		 */
		if (!(blk_queue_tagged(q) && !blk_queue_start_tag(q, req)))
			blkdev_dequeue_request(req);
		sdev->device_busy++;

		spin_unlock(q->queue_lock);
1426 1427 1428 1429
		cmd = req->special;
		if (unlikely(cmd == NULL)) {
			printk(KERN_CRIT "impossible request in %s.\n"
					 "please mail a stack trace to "
1430
					 "linux-scsi@vger.kernel.org\n",
1431
					 __FUNCTION__);
1432
			blk_dump_rq_flags(req, "foo");
1433 1434
			BUG();
		}
L
Linus Torvalds 已提交
1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522
		spin_lock(shost->host_lock);

		if (!scsi_host_queue_ready(q, shost, sdev))
			goto not_ready;
		if (sdev->single_lun) {
			if (scsi_target(sdev)->starget_sdev_user &&
			    scsi_target(sdev)->starget_sdev_user != sdev)
				goto not_ready;
			scsi_target(sdev)->starget_sdev_user = sdev;
		}
		shost->host_busy++;

		/*
		 * XXX(hch): This is rather suboptimal, scsi_dispatch_cmd will
		 *		take the lock again.
		 */
		spin_unlock_irq(shost->host_lock);

		/*
		 * Finally, initialize any error handling parameters, and set up
		 * the timers for timeouts.
		 */
		scsi_init_cmd_errh(cmd);

		/*
		 * Dispatch the command to the low-level driver.
		 */
		rtn = scsi_dispatch_cmd(cmd);
		spin_lock_irq(q->queue_lock);
		if(rtn) {
			/* we're refusing the command; because of
			 * the way locks get dropped, we need to 
			 * check here if plugging is required */
			if(sdev->device_busy == 0)
				blk_plug_device(q);

			break;
		}
	}

	goto out;

 not_ready:
	spin_unlock_irq(shost->host_lock);

	/*
	 * lock q, handle tag, requeue req, and decrement device_busy. We
	 * must return with queue_lock held.
	 *
	 * Decrementing device_busy without checking it is OK, as all such
	 * cases (host limits or settings) should run the queue at some
	 * later time.
	 */
	spin_lock_irq(q->queue_lock);
	blk_requeue_request(q, req);
	sdev->device_busy--;
	if(sdev->device_busy == 0)
		blk_plug_device(q);
 out:
	/* must be careful here...if we trigger the ->remove() function
	 * we cannot be holding the q lock */
	spin_unlock_irq(q->queue_lock);
	put_device(&sdev->sdev_gendev);
	spin_lock_irq(q->queue_lock);
}

u64 scsi_calculate_bounce_limit(struct Scsi_Host *shost)
{
	struct device *host_dev;
	u64 bounce_limit = 0xffffffff;

	if (shost->unchecked_isa_dma)
		return BLK_BOUNCE_ISA;
	/*
	 * Platforms with virtual-DMA translation
	 * hardware have no practical limit.
	 */
	if (!PCI_DMA_BUS_IS_PHYS)
		return BLK_BOUNCE_ANY;

	host_dev = scsi_get_device(shost);
	if (host_dev && host_dev->dma_mask)
		bounce_limit = *host_dev->dma_mask;

	return bounce_limit;
}
EXPORT_SYMBOL(scsi_calculate_bounce_limit);

1523 1524
struct request_queue *__scsi_alloc_queue(struct Scsi_Host *shost,
					 request_fn_proc *request_fn)
L
Linus Torvalds 已提交
1525 1526 1527
{
	struct request_queue *q;

1528
	q = blk_init_queue(request_fn, NULL);
L
Linus Torvalds 已提交
1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541
	if (!q)
		return NULL;

	blk_queue_max_hw_segments(q, shost->sg_tablesize);
	blk_queue_max_phys_segments(q, SCSI_MAX_PHYS_SEGMENTS);
	blk_queue_max_sectors(q, shost->max_sectors);
	blk_queue_bounce_limit(q, scsi_calculate_bounce_limit(shost));
	blk_queue_segment_boundary(q, shost->dma_boundary);

	if (!shost->use_clustering)
		clear_bit(QUEUE_FLAG_CLUSTER, &q->queue_flags);
	return q;
}
1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555
EXPORT_SYMBOL(__scsi_alloc_queue);

struct request_queue *scsi_alloc_queue(struct scsi_device *sdev)
{
	struct request_queue *q;

	q = __scsi_alloc_queue(sdev->host, scsi_request_fn);
	if (!q)
		return NULL;

	blk_queue_prep_rq(q, scsi_prep_fn);
	blk_queue_softirq_done(q, scsi_softirq_done);
	return q;
}
L
Linus Torvalds 已提交
1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614

void scsi_free_queue(struct request_queue *q)
{
	blk_cleanup_queue(q);
}

/*
 * Function:    scsi_block_requests()
 *
 * Purpose:     Utility function used by low-level drivers to prevent further
 *		commands from being queued to the device.
 *
 * Arguments:   shost       - Host in question
 *
 * Returns:     Nothing
 *
 * Lock status: No locks are assumed held.
 *
 * Notes:       There is no timer nor any other means by which the requests
 *		get unblocked other than the low-level driver calling
 *		scsi_unblock_requests().
 */
void scsi_block_requests(struct Scsi_Host *shost)
{
	shost->host_self_blocked = 1;
}
EXPORT_SYMBOL(scsi_block_requests);

/*
 * Function:    scsi_unblock_requests()
 *
 * Purpose:     Utility function used by low-level drivers to allow further
 *		commands from being queued to the device.
 *
 * Arguments:   shost       - Host in question
 *
 * Returns:     Nothing
 *
 * Lock status: No locks are assumed held.
 *
 * Notes:       There is no timer nor any other means by which the requests
 *		get unblocked other than the low-level driver calling
 *		scsi_unblock_requests().
 *
 *		This is done as an API function so that changes to the
 *		internals of the scsi mid-layer won't require wholesale
 *		changes to drivers that use this feature.
 */
void scsi_unblock_requests(struct Scsi_Host *shost)
{
	shost->host_self_blocked = 0;
	scsi_run_host_queues(shost);
}
EXPORT_SYMBOL(scsi_unblock_requests);

int __init scsi_init_queue(void)
{
	int i;

1615 1616
	scsi_io_context_cache = kmem_cache_create("scsi_io_context",
					sizeof(struct scsi_io_context),
1617
					0, 0, NULL);
1618 1619 1620 1621 1622
	if (!scsi_io_context_cache) {
		printk(KERN_ERR "SCSI: can't init scsi io context cache\n");
		return -ENOMEM;
	}

L
Linus Torvalds 已提交
1623 1624 1625 1626 1627
	for (i = 0; i < SG_MEMPOOL_NR; i++) {
		struct scsi_host_sg_pool *sgp = scsi_sg_pools + i;
		int size = sgp->size * sizeof(struct scatterlist);

		sgp->slab = kmem_cache_create(sgp->name, size, 0,
1628
				SLAB_HWCACHE_ALIGN, NULL);
L
Linus Torvalds 已提交
1629 1630 1631 1632 1633
		if (!sgp->slab) {
			printk(KERN_ERR "SCSI: can't init sg slab %s\n",
					sgp->name);
		}

1634 1635
		sgp->pool = mempool_create_slab_pool(SG_MEMPOOL_SIZE,
						     sgp->slab);
L
Linus Torvalds 已提交
1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648
		if (!sgp->pool) {
			printk(KERN_ERR "SCSI: can't init sg mempool %s\n",
					sgp->name);
		}
	}

	return 0;
}

void scsi_exit_queue(void)
{
	int i;

1649 1650
	kmem_cache_destroy(scsi_io_context_cache);

L
Linus Torvalds 已提交
1651 1652 1653 1654 1655 1656
	for (i = 0; i < SG_MEMPOOL_NR; i++) {
		struct scsi_host_sg_pool *sgp = scsi_sg_pools + i;
		mempool_destroy(sgp->pool);
		kmem_cache_destroy(sgp->slab);
	}
}
1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734

/**
 *	scsi_mode_select - issue a mode select
 *	@sdev:	SCSI device to be queried
 *	@pf:	Page format bit (1 == standard, 0 == vendor specific)
 *	@sp:	Save page bit (0 == don't save, 1 == save)
 *	@modepage: mode page being requested
 *	@buffer: request buffer (may not be smaller than eight bytes)
 *	@len:	length of request buffer.
 *	@timeout: command timeout
 *	@retries: number of retries before failing
 *	@data: returns a structure abstracting the mode header data
 *	@sense: place to put sense data (or NULL if no sense to be collected).
 *		must be SCSI_SENSE_BUFFERSIZE big.
 *
 *	Returns zero if successful; negative error number or scsi
 *	status on error
 *
 */
int
scsi_mode_select(struct scsi_device *sdev, int pf, int sp, int modepage,
		 unsigned char *buffer, int len, int timeout, int retries,
		 struct scsi_mode_data *data, struct scsi_sense_hdr *sshdr)
{
	unsigned char cmd[10];
	unsigned char *real_buffer;
	int ret;

	memset(cmd, 0, sizeof(cmd));
	cmd[1] = (pf ? 0x10 : 0) | (sp ? 0x01 : 0);

	if (sdev->use_10_for_ms) {
		if (len > 65535)
			return -EINVAL;
		real_buffer = kmalloc(8 + len, GFP_KERNEL);
		if (!real_buffer)
			return -ENOMEM;
		memcpy(real_buffer + 8, buffer, len);
		len += 8;
		real_buffer[0] = 0;
		real_buffer[1] = 0;
		real_buffer[2] = data->medium_type;
		real_buffer[3] = data->device_specific;
		real_buffer[4] = data->longlba ? 0x01 : 0;
		real_buffer[5] = 0;
		real_buffer[6] = data->block_descriptor_length >> 8;
		real_buffer[7] = data->block_descriptor_length;

		cmd[0] = MODE_SELECT_10;
		cmd[7] = len >> 8;
		cmd[8] = len;
	} else {
		if (len > 255 || data->block_descriptor_length > 255 ||
		    data->longlba)
			return -EINVAL;

		real_buffer = kmalloc(4 + len, GFP_KERNEL);
		if (!real_buffer)
			return -ENOMEM;
		memcpy(real_buffer + 4, buffer, len);
		len += 4;
		real_buffer[0] = 0;
		real_buffer[1] = data->medium_type;
		real_buffer[2] = data->device_specific;
		real_buffer[3] = data->block_descriptor_length;
		

		cmd[0] = MODE_SELECT;
		cmd[4] = len;
	}

	ret = scsi_execute_req(sdev, cmd, DMA_TO_DEVICE, real_buffer, len,
			       sshdr, timeout, retries);
	kfree(real_buffer);
	return ret;
}
EXPORT_SYMBOL_GPL(scsi_mode_select);

L
Linus Torvalds 已提交
1735
/**
1736
 *	scsi_mode_sense - issue a mode sense, falling back from 10 to 
L
Linus Torvalds 已提交
1737
 *		six bytes if necessary.
1738
 *	@sdev:	SCSI device to be queried
L
Linus Torvalds 已提交
1739 1740 1741 1742 1743 1744 1745
 *	@dbd:	set if mode sense will allow block descriptors to be returned
 *	@modepage: mode page being requested
 *	@buffer: request buffer (may not be smaller than eight bytes)
 *	@len:	length of request buffer.
 *	@timeout: command timeout
 *	@retries: number of retries before failing
 *	@data: returns a structure abstracting the mode header data
1746 1747
 *	@sense: place to put sense data (or NULL if no sense to be collected).
 *		must be SCSI_SENSE_BUFFERSIZE big.
L
Linus Torvalds 已提交
1748 1749 1750 1751 1752 1753
 *
 *	Returns zero if unsuccessful, or the header offset (either 4
 *	or 8 depending on whether a six or ten byte command was
 *	issued) if successful.
 **/
int
1754
scsi_mode_sense(struct scsi_device *sdev, int dbd, int modepage,
L
Linus Torvalds 已提交
1755
		  unsigned char *buffer, int len, int timeout, int retries,
1756 1757
		  struct scsi_mode_data *data, struct scsi_sense_hdr *sshdr)
{
L
Linus Torvalds 已提交
1758 1759 1760
	unsigned char cmd[12];
	int use_10_for_ms;
	int header_length;
1761
	int result;
1762
	struct scsi_sense_hdr my_sshdr;
L
Linus Torvalds 已提交
1763 1764 1765 1766 1767 1768

	memset(data, 0, sizeof(*data));
	memset(&cmd[0], 0, 12);
	cmd[1] = dbd & 0x18;	/* allows DBD and LLBA bits */
	cmd[2] = modepage;

1769 1770 1771 1772
	/* caller might not be interested in sense, but we need it */
	if (!sshdr)
		sshdr = &my_sshdr;

L
Linus Torvalds 已提交
1773
 retry:
1774
	use_10_for_ms = sdev->use_10_for_ms;
L
Linus Torvalds 已提交
1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793

	if (use_10_for_ms) {
		if (len < 8)
			len = 8;

		cmd[0] = MODE_SENSE_10;
		cmd[8] = len;
		header_length = 8;
	} else {
		if (len < 4)
			len = 4;

		cmd[0] = MODE_SENSE;
		cmd[4] = len;
		header_length = 4;
	}

	memset(buffer, 0, len);

1794
	result = scsi_execute_req(sdev, cmd, DMA_FROM_DEVICE, buffer, len,
1795
				  sshdr, timeout, retries);
L
Linus Torvalds 已提交
1796 1797 1798 1799 1800 1801

	/* This code looks awful: what it's doing is making sure an
	 * ILLEGAL REQUEST sense return identifies the actual command
	 * byte as the problem.  MODE_SENSE commands can return
	 * ILLEGAL REQUEST if the code page isn't supported */

1802 1803
	if (use_10_for_ms && !scsi_status_is_good(result) &&
	    (driver_byte(result) & DRIVER_SENSE)) {
1804 1805 1806
		if (scsi_sense_valid(sshdr)) {
			if ((sshdr->sense_key == ILLEGAL_REQUEST) &&
			    (sshdr->asc == 0x20) && (sshdr->ascq == 0)) {
L
Linus Torvalds 已提交
1807 1808 1809
				/* 
				 * Invalid command operation code
				 */
1810
				sdev->use_10_for_ms = 0;
L
Linus Torvalds 已提交
1811 1812 1813 1814 1815
				goto retry;
			}
		}
	}

1816
	if(scsi_status_is_good(result)) {
1817 1818 1819 1820 1821 1822 1823 1824 1825 1826
		if (unlikely(buffer[0] == 0x86 && buffer[1] == 0x0b &&
			     (modepage == 6 || modepage == 8))) {
			/* Initio breakage? */
			header_length = 0;
			data->length = 13;
			data->medium_type = 0;
			data->device_specific = 0;
			data->longlba = 0;
			data->block_descriptor_length = 0;
		} else if(use_10_for_ms) {
L
Linus Torvalds 已提交
1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838
			data->length = buffer[0]*256 + buffer[1] + 2;
			data->medium_type = buffer[2];
			data->device_specific = buffer[3];
			data->longlba = buffer[4] & 0x01;
			data->block_descriptor_length = buffer[6]*256
				+ buffer[7];
		} else {
			data->length = buffer[0] + 1;
			data->medium_type = buffer[1];
			data->device_specific = buffer[2];
			data->block_descriptor_length = buffer[3];
		}
1839
		data->header_length = header_length;
L
Linus Torvalds 已提交
1840 1841
	}

1842
	return result;
L
Linus Torvalds 已提交
1843 1844 1845 1846 1847 1848 1849 1850 1851
}
EXPORT_SYMBOL(scsi_mode_sense);

int
scsi_test_unit_ready(struct scsi_device *sdev, int timeout, int retries)
{
	char cmd[] = {
		TEST_UNIT_READY, 0, 0, 0, 0, 0,
	};
1852
	struct scsi_sense_hdr sshdr;
L
Linus Torvalds 已提交
1853 1854
	int result;
	
1855
	result = scsi_execute_req(sdev, cmd, DMA_NONE, NULL, 0, &sshdr,
1856
				  timeout, retries);
L
Linus Torvalds 已提交
1857

1858
	if ((driver_byte(result) & DRIVER_SENSE) && sdev->removable) {
L
Linus Torvalds 已提交
1859

1860
		if ((scsi_sense_valid(&sshdr)) &&
L
Linus Torvalds 已提交
1861 1862 1863
		    ((sshdr.sense_key == UNIT_ATTENTION) ||
		     (sshdr.sense_key == NOT_READY))) {
			sdev->changed = 1;
1864
			result = 0;
L
Linus Torvalds 已提交
1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942
		}
	}
	return result;
}
EXPORT_SYMBOL(scsi_test_unit_ready);

/**
 *	scsi_device_set_state - Take the given device through the device
 *		state model.
 *	@sdev:	scsi device to change the state of.
 *	@state:	state to change to.
 *
 *	Returns zero if unsuccessful or an error if the requested 
 *	transition is illegal.
 **/
int
scsi_device_set_state(struct scsi_device *sdev, enum scsi_device_state state)
{
	enum scsi_device_state oldstate = sdev->sdev_state;

	if (state == oldstate)
		return 0;

	switch (state) {
	case SDEV_CREATED:
		/* There are no legal states that come back to
		 * created.  This is the manually initialised start
		 * state */
		goto illegal;
			
	case SDEV_RUNNING:
		switch (oldstate) {
		case SDEV_CREATED:
		case SDEV_OFFLINE:
		case SDEV_QUIESCE:
		case SDEV_BLOCK:
			break;
		default:
			goto illegal;
		}
		break;

	case SDEV_QUIESCE:
		switch (oldstate) {
		case SDEV_RUNNING:
		case SDEV_OFFLINE:
			break;
		default:
			goto illegal;
		}
		break;

	case SDEV_OFFLINE:
		switch (oldstate) {
		case SDEV_CREATED:
		case SDEV_RUNNING:
		case SDEV_QUIESCE:
		case SDEV_BLOCK:
			break;
		default:
			goto illegal;
		}
		break;

	case SDEV_BLOCK:
		switch (oldstate) {
		case SDEV_CREATED:
		case SDEV_RUNNING:
			break;
		default:
			goto illegal;
		}
		break;

	case SDEV_CANCEL:
		switch (oldstate) {
		case SDEV_CREATED:
		case SDEV_RUNNING:
1943
		case SDEV_QUIESCE:
L
Linus Torvalds 已提交
1944 1945 1946 1947 1948 1949 1950 1951 1952 1953
		case SDEV_OFFLINE:
		case SDEV_BLOCK:
			break;
		default:
			goto illegal;
		}
		break;

	case SDEV_DEL:
		switch (oldstate) {
1954 1955 1956
		case SDEV_CREATED:
		case SDEV_RUNNING:
		case SDEV_OFFLINE:
L
Linus Torvalds 已提交
1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969
		case SDEV_CANCEL:
			break;
		default:
			goto illegal;
		}
		break;

	}
	sdev->sdev_state = state;
	return 0;

 illegal:
	SCSI_LOG_ERROR_RECOVERY(1, 
1970 1971 1972 1973
				sdev_printk(KERN_ERR, sdev,
					    "Illegal state transition %s->%s\n",
					    scsi_device_state_name(oldstate),
					    scsi_device_state_name(state))
L
Linus Torvalds 已提交
1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074
				);
	return -EINVAL;
}
EXPORT_SYMBOL(scsi_device_set_state);

/**
 *	scsi_device_quiesce - Block user issued commands.
 *	@sdev:	scsi device to quiesce.
 *
 *	This works by trying to transition to the SDEV_QUIESCE state
 *	(which must be a legal transition).  When the device is in this
 *	state, only special requests will be accepted, all others will
 *	be deferred.  Since special requests may also be requeued requests,
 *	a successful return doesn't guarantee the device will be 
 *	totally quiescent.
 *
 *	Must be called with user context, may sleep.
 *
 *	Returns zero if unsuccessful or an error if not.
 **/
int
scsi_device_quiesce(struct scsi_device *sdev)
{
	int err = scsi_device_set_state(sdev, SDEV_QUIESCE);
	if (err)
		return err;

	scsi_run_queue(sdev->request_queue);
	while (sdev->device_busy) {
		msleep_interruptible(200);
		scsi_run_queue(sdev->request_queue);
	}
	return 0;
}
EXPORT_SYMBOL(scsi_device_quiesce);

/**
 *	scsi_device_resume - Restart user issued commands to a quiesced device.
 *	@sdev:	scsi device to resume.
 *
 *	Moves the device from quiesced back to running and restarts the
 *	queues.
 *
 *	Must be called with user context, may sleep.
 **/
void
scsi_device_resume(struct scsi_device *sdev)
{
	if(scsi_device_set_state(sdev, SDEV_RUNNING))
		return;
	scsi_run_queue(sdev->request_queue);
}
EXPORT_SYMBOL(scsi_device_resume);

static void
device_quiesce_fn(struct scsi_device *sdev, void *data)
{
	scsi_device_quiesce(sdev);
}

void
scsi_target_quiesce(struct scsi_target *starget)
{
	starget_for_each_device(starget, NULL, device_quiesce_fn);
}
EXPORT_SYMBOL(scsi_target_quiesce);

static void
device_resume_fn(struct scsi_device *sdev, void *data)
{
	scsi_device_resume(sdev);
}

void
scsi_target_resume(struct scsi_target *starget)
{
	starget_for_each_device(starget, NULL, device_resume_fn);
}
EXPORT_SYMBOL(scsi_target_resume);

/**
 * scsi_internal_device_block - internal function to put a device
 *				temporarily into the SDEV_BLOCK state
 * @sdev:	device to block
 *
 * Block request made by scsi lld's to temporarily stop all
 * scsi commands on the specified device.  Called from interrupt
 * or normal process context.
 *
 * Returns zero if successful or error if not
 *
 * Notes:       
 *	This routine transitions the device to the SDEV_BLOCK state
 *	(which must be a legal transition).  When the device is in this
 *	state, all commands are deferred until the scsi lld reenables
 *	the device with scsi_device_unblock or device_block_tmo fires.
 *	This routine assumes the host_lock is held on entry.
 **/
int
scsi_internal_device_block(struct scsi_device *sdev)
{
2075
	struct request_queue *q = sdev->request_queue;
L
Linus Torvalds 已提交
2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114
	unsigned long flags;
	int err = 0;

	err = scsi_device_set_state(sdev, SDEV_BLOCK);
	if (err)
		return err;

	/* 
	 * The device has transitioned to SDEV_BLOCK.  Stop the
	 * block layer from calling the midlayer with this device's
	 * request queue. 
	 */
	spin_lock_irqsave(q->queue_lock, flags);
	blk_stop_queue(q);
	spin_unlock_irqrestore(q->queue_lock, flags);

	return 0;
}
EXPORT_SYMBOL_GPL(scsi_internal_device_block);
 
/**
 * scsi_internal_device_unblock - resume a device after a block request
 * @sdev:	device to resume
 *
 * Called by scsi lld's or the midlayer to restart the device queue
 * for the previously suspended scsi device.  Called from interrupt or
 * normal process context.
 *
 * Returns zero if successful or error if not.
 *
 * Notes:       
 *	This routine transitions the device to the SDEV_RUNNING state
 *	(which must be a legal transition) allowing the midlayer to
 *	goose the queue for this device.  This routine assumes the 
 *	host_lock is held upon entry.
 **/
int
scsi_internal_device_unblock(struct scsi_device *sdev)
{
2115
	struct request_queue *q = sdev->request_queue; 
L
Linus Torvalds 已提交
2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185
	int err;
	unsigned long flags;
	
	/* 
	 * Try to transition the scsi device to SDEV_RUNNING
	 * and goose the device queue if successful.  
	 */
	err = scsi_device_set_state(sdev, SDEV_RUNNING);
	if (err)
		return err;

	spin_lock_irqsave(q->queue_lock, flags);
	blk_start_queue(q);
	spin_unlock_irqrestore(q->queue_lock, flags);

	return 0;
}
EXPORT_SYMBOL_GPL(scsi_internal_device_unblock);

static void
device_block(struct scsi_device *sdev, void *data)
{
	scsi_internal_device_block(sdev);
}

static int
target_block(struct device *dev, void *data)
{
	if (scsi_is_target_device(dev))
		starget_for_each_device(to_scsi_target(dev), NULL,
					device_block);
	return 0;
}

void
scsi_target_block(struct device *dev)
{
	if (scsi_is_target_device(dev))
		starget_for_each_device(to_scsi_target(dev), NULL,
					device_block);
	else
		device_for_each_child(dev, NULL, target_block);
}
EXPORT_SYMBOL_GPL(scsi_target_block);

static void
device_unblock(struct scsi_device *sdev, void *data)
{
	scsi_internal_device_unblock(sdev);
}

static int
target_unblock(struct device *dev, void *data)
{
	if (scsi_is_target_device(dev))
		starget_for_each_device(to_scsi_target(dev), NULL,
					device_unblock);
	return 0;
}

void
scsi_target_unblock(struct device *dev)
{
	if (scsi_is_target_device(dev))
		starget_for_each_device(to_scsi_target(dev), NULL,
					device_unblock);
	else
		device_for_each_child(dev, NULL, target_unblock);
}
EXPORT_SYMBOL_GPL(scsi_target_unblock);
2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202

/**
 * scsi_kmap_atomic_sg - find and atomically map an sg-elemnt
 * @sg:		scatter-gather list
 * @sg_count:	number of segments in sg
 * @offset:	offset in bytes into sg, on return offset into the mapped area
 * @len:	bytes to map, on return number of bytes mapped
 *
 * Returns virtual address of the start of the mapped page
 */
void *scsi_kmap_atomic_sg(struct scatterlist *sg, int sg_count,
			  size_t *offset, size_t *len)
{
	int i;
	size_t sg_len = 0, len_complete = 0;
	struct page *page;

2203 2204
	WARN_ON(!irqs_disabled());

2205 2206 2207 2208 2209 2210 2211 2212
	for (i = 0; i < sg_count; i++) {
		len_complete = sg_len; /* Complete sg-entries */
		sg_len += sg[i].length;
		if (sg_len > *offset)
			break;
	}

	if (unlikely(i == sg_count)) {
2213 2214
		printk(KERN_ERR "%s: Bytes in sg: %zu, requested offset %zu, "
			"elements %d\n",
2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245
		       __FUNCTION__, sg_len, *offset, sg_count);
		WARN_ON(1);
		return NULL;
	}

	/* Offset starting from the beginning of first page in this sg-entry */
	*offset = *offset - len_complete + sg[i].offset;

	/* Assumption: contiguous pages can be accessed as "page + i" */
	page = nth_page(sg[i].page, (*offset >> PAGE_SHIFT));
	*offset &= ~PAGE_MASK;

	/* Bytes in this sg-entry from *offset to the end of the page */
	sg_len = PAGE_SIZE - *offset;
	if (*len > sg_len)
		*len = sg_len;

	return kmap_atomic(page, KM_BIO_SRC_IRQ);
}
EXPORT_SYMBOL(scsi_kmap_atomic_sg);

/**
 * scsi_kunmap_atomic_sg - atomically unmap a virtual address, previously
 *			   mapped with scsi_kmap_atomic_sg
 * @virt:	virtual address to be unmapped
 */
void scsi_kunmap_atomic_sg(void *virt)
{
	kunmap_atomic(virt, KM_BIO_SRC_IRQ);
}
EXPORT_SYMBOL(scsi_kunmap_atomic_sg);