virtio_blk.c 19.3 KB
Newer Older
R
Rusty Russell 已提交
1 2
//#define DEBUG
#include <linux/spinlock.h>
3
#include <linux/slab.h>
R
Rusty Russell 已提交
4 5
#include <linux/blkdev.h>
#include <linux/hdreg.h>
6
#include <linux/module.h>
7
#include <linux/mutex.h>
R
Rusty Russell 已提交
8 9
#include <linux/virtio.h>
#include <linux/virtio_blk.h>
10
#include <linux/scatterlist.h>
11
#include <linux/string_helpers.h>
12
#include <scsi/scsi_cmnd.h>
13
#include <linux/idr.h>
14

15
#define PART_BITS 4
R
Rusty Russell 已提交
16

17 18 19
static int major;
static DEFINE_IDA(vd_index_ida);

20
struct workqueue_struct *virtblk_wq;
21

R
Rusty Russell 已提交
22 23 24 25 26 27 28 29 30 31
struct virtio_blk
{
	struct virtio_device *vdev;
	struct virtqueue *vq;

	/* The disk structure for the kernel. */
	struct gendisk *disk;

	mempool_t *pool;

32 33 34
	/* Process context for config space updates */
	struct work_struct config_work;

35 36 37 38 39 40
	/* Lock for config space updates */
	struct mutex config_lock;

	/* enable config space updates */
	bool config_enable;

41 42 43
	/* What host tells us, plus 2 for header & tailer. */
	unsigned int sg_elems;

44 45 46
	/* Ida index - used to track minor number allocations. */
	int index;

R
Rusty Russell 已提交
47
	/* Scatterlist: can be too big for stack. */
48
	struct scatterlist sg[/*sg_elems*/];
R
Rusty Russell 已提交
49 50 51 52 53 54
};

struct virtblk_req
{
	struct request *req;
	struct virtio_blk_outhdr out_hdr;
55
	struct virtio_scsi_inhdr in_hdr;
56
	u8 status;
R
Rusty Russell 已提交
57 58
};

59
static void blk_done(struct virtqueue *vq)
R
Rusty Russell 已提交
60 61 62 63 64 65
{
	struct virtio_blk *vblk = vq->vdev->priv;
	struct virtblk_req *vbr;
	unsigned int len;
	unsigned long flags;

66
	spin_lock_irqsave(vblk->disk->queue->queue_lock, flags);
67
	while ((vbr = virtqueue_get_buf(vblk->vq, &len)) != NULL) {
68
		int error;
69

70
		switch (vbr->status) {
R
Rusty Russell 已提交
71
		case VIRTIO_BLK_S_OK:
72
			error = 0;
R
Rusty Russell 已提交
73 74
			break;
		case VIRTIO_BLK_S_UNSUPP:
75
			error = -ENOTTY;
R
Rusty Russell 已提交
76 77
			break;
		default:
78
			error = -EIO;
R
Rusty Russell 已提交
79 80 81
			break;
		}

82 83
		switch (vbr->req->cmd_type) {
		case REQ_TYPE_BLOCK_PC:
84 85 86
			vbr->req->resid_len = vbr->in_hdr.residual;
			vbr->req->sense_len = vbr->in_hdr.sense_len;
			vbr->req->errors = vbr->in_hdr.errors;
87 88
			break;
		case REQ_TYPE_SPECIAL:
89
			vbr->req->errors = (error != 0);
90
			break;
91 92
		default:
			break;
93
		}
94

95
		__blk_end_request_all(vbr->req, error);
R
Rusty Russell 已提交
96 97 98 99
		mempool_free(vbr, vblk->pool);
	}
	/* In case queue is stopped waiting for more buffers. */
	blk_start_queue(vblk->disk->queue);
100
	spin_unlock_irqrestore(vblk->disk->queue->queue_lock, flags);
R
Rusty Russell 已提交
101 102 103 104 105
}

static bool do_req(struct request_queue *q, struct virtio_blk *vblk,
		   struct request *req)
{
106
	unsigned long num, out = 0, in = 0;
R
Rusty Russell 已提交
107 108 109 110 111 112 113 114
	struct virtblk_req *vbr;

	vbr = mempool_alloc(vblk->pool, GFP_ATOMIC);
	if (!vbr)
		/* When another request finishes we'll try again. */
		return false;

	vbr->req = req;
115 116 117

	if (req->cmd_flags & REQ_FLUSH) {
		vbr->out_hdr.type = VIRTIO_BLK_T_FLUSH;
118 119
		vbr->out_hdr.sector = 0;
		vbr->out_hdr.ioprio = req_get_ioprio(vbr->req);
120 121 122 123 124 125 126 127 128
	} else {
		switch (req->cmd_type) {
		case REQ_TYPE_FS:
			vbr->out_hdr.type = 0;
			vbr->out_hdr.sector = blk_rq_pos(vbr->req);
			vbr->out_hdr.ioprio = req_get_ioprio(vbr->req);
			break;
		case REQ_TYPE_BLOCK_PC:
			vbr->out_hdr.type = VIRTIO_BLK_T_SCSI_CMD;
129 130 131
			vbr->out_hdr.sector = 0;
			vbr->out_hdr.ioprio = req_get_ioprio(vbr->req);
			break;
132 133 134 135 136 137 138 139
		case REQ_TYPE_SPECIAL:
			vbr->out_hdr.type = VIRTIO_BLK_T_GET_ID;
			vbr->out_hdr.sector = 0;
			vbr->out_hdr.ioprio = req_get_ioprio(vbr->req);
			break;
		default:
			/* We don't put anything else in the queue. */
			BUG();
140
		}
R
Rusty Russell 已提交
141 142
	}

143
	sg_set_buf(&vblk->sg[out++], &vbr->out_hdr, sizeof(vbr->out_hdr));
R
Rusty Russell 已提交
144

145 146 147 148 149 150
	/*
	 * If this is a packet command we need a couple of additional headers.
	 * Behind the normal outhdr we put a segment with the scsi command
	 * block, and before the normal inhdr we put the sense data and the
	 * inhdr with additional status information before the normal inhdr.
	 */
151
	if (vbr->req->cmd_type == REQ_TYPE_BLOCK_PC)
152 153 154 155
		sg_set_buf(&vblk->sg[out++], vbr->req->cmd, vbr->req->cmd_len);

	num = blk_rq_map_sg(q, vbr->req, vblk->sg + out);

156
	if (vbr->req->cmd_type == REQ_TYPE_BLOCK_PC) {
157
		sg_set_buf(&vblk->sg[num + out + in++], vbr->req->sense, SCSI_SENSE_BUFFERSIZE);
158 159 160 161 162 163 164 165 166 167 168 169 170 171 172
		sg_set_buf(&vblk->sg[num + out + in++], &vbr->in_hdr,
			   sizeof(vbr->in_hdr));
	}

	sg_set_buf(&vblk->sg[num + out + in++], &vbr->status,
		   sizeof(vbr->status));

	if (num) {
		if (rq_data_dir(vbr->req) == WRITE) {
			vbr->out_hdr.type |= VIRTIO_BLK_T_OUT;
			out += num;
		} else {
			vbr->out_hdr.type |= VIRTIO_BLK_T_IN;
			in += num;
		}
R
Rusty Russell 已提交
173 174
	}

175
	if (virtqueue_add_buf(vblk->vq, vblk->sg, out, in, vbr, GFP_ATOMIC)<0) {
R
Rusty Russell 已提交
176 177 178 179 180 181 182 183 184
		mempool_free(vbr, vblk->pool);
		return false;
	}

	return true;
}

static void do_virtblk_request(struct request_queue *q)
{
185
	struct virtio_blk *vblk = q->queuedata;
R
Rusty Russell 已提交
186 187 188
	struct request *req;
	unsigned int issued = 0;

189
	while ((req = blk_peek_request(q)) != NULL) {
190
		BUG_ON(req->nr_phys_segments + 2 > vblk->sg_elems);
R
Rusty Russell 已提交
191 192 193 194 195 196 197

		/* If this request fails, stop queue and wait for something to
		   finish to restart it. */
		if (!do_req(q, vblk, req)) {
			blk_stop_queue(q);
			break;
		}
198
		blk_start_request(req);
R
Rusty Russell 已提交
199 200 201 202
		issued++;
	}

	if (issued)
203
		virtqueue_kick(vblk->vq);
R
Rusty Russell 已提交
204 205
}

206 207 208 209 210 211 212
/* return id (s/n) string for *disk to *id_str
 */
static int virtblk_get_id(struct gendisk *disk, char *id_str)
{
	struct virtio_blk *vblk = disk->private_data;
	struct request *req;
	struct bio *bio;
M
Mike Snitzer 已提交
213
	int err;
214 215 216 217 218 219 220 221 222 223 224 225 226

	bio = bio_map_kern(vblk->disk->queue, id_str, VIRTIO_BLK_ID_BYTES,
			   GFP_KERNEL);
	if (IS_ERR(bio))
		return PTR_ERR(bio);

	req = blk_make_request(vblk->disk->queue, bio, GFP_KERNEL);
	if (IS_ERR(req)) {
		bio_put(bio);
		return PTR_ERR(req);
	}

	req->cmd_type = REQ_TYPE_SPECIAL;
M
Mike Snitzer 已提交
227 228 229 230
	err = blk_execute_rq(vblk->disk->queue, vblk->disk, req, false);
	blk_put_request(req);

	return err;
231 232
}

233 234
static int virtblk_ioctl(struct block_device *bdev, fmode_t mode,
			     unsigned int cmd, unsigned long data)
R
Rusty Russell 已提交
235
{
236 237 238 239 240 241 242
	struct gendisk *disk = bdev->bd_disk;
	struct virtio_blk *vblk = disk->private_data;

	/*
	 * Only allow the generic SCSI ioctls if the host can support it.
	 */
	if (!virtio_has_feature(vblk->vdev, VIRTIO_BLK_F_SCSI))
243
		return -ENOTTY;
244

245 246
	return scsi_cmd_blk_ioctl(bdev, mode, cmd,
				  (void __user *)data);
R
Rusty Russell 已提交
247 248
}

249 250 251
/* We provide getgeo only to please some old bootloader/partitioning tools */
static int virtblk_getgeo(struct block_device *bd, struct hd_geometry *geo)
{
252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270
	struct virtio_blk *vblk = bd->bd_disk->private_data;
	struct virtio_blk_geometry vgeo;
	int err;

	/* see if the host passed in geometry config */
	err = virtio_config_val(vblk->vdev, VIRTIO_BLK_F_GEOMETRY,
				offsetof(struct virtio_blk_config, geometry),
				&vgeo);

	if (!err) {
		geo->heads = vgeo.heads;
		geo->sectors = vgeo.sectors;
		geo->cylinders = vgeo.cylinders;
	} else {
		/* some standard values, similar to sd */
		geo->heads = 1 << 6;
		geo->sectors = 1 << 5;
		geo->cylinders = get_capacity(bd->bd_disk) >> 11;
	}
271 272 273
	return 0;
}

274
static const struct block_device_operations virtblk_fops = {
275
	.ioctl  = virtblk_ioctl,
276 277
	.owner  = THIS_MODULE,
	.getgeo = virtblk_getgeo,
R
Rusty Russell 已提交
278 279
};

280 281 282 283 284
static int index_to_minor(int index)
{
	return index << PART_BITS;
}

285 286 287 288 289
static int minor_to_index(int minor)
{
	return minor >> PART_BITS;
}

290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310
static ssize_t virtblk_serial_show(struct device *dev,
				struct device_attribute *attr, char *buf)
{
	struct gendisk *disk = dev_to_disk(dev);
	int err;

	/* sysfs gives us a PAGE_SIZE buffer */
	BUILD_BUG_ON(PAGE_SIZE < VIRTIO_BLK_ID_BYTES);

	buf[VIRTIO_BLK_ID_BYTES] = '\0';
	err = virtblk_get_id(disk, buf);
	if (!err)
		return strlen(buf);

	if (err == -EIO) /* Unsupported? Make it empty. */
		return 0;

	return err;
}
DEVICE_ATTR(serial, S_IRUGO, virtblk_serial_show, NULL);

311 312 313 314 315 316 317 318 319
static void virtblk_config_changed_work(struct work_struct *work)
{
	struct virtio_blk *vblk =
		container_of(work, struct virtio_blk, config_work);
	struct virtio_device *vdev = vblk->vdev;
	struct request_queue *q = vblk->disk->queue;
	char cap_str_2[10], cap_str_10[10];
	u64 capacity, size;

320 321 322 323
	mutex_lock(&vblk->config_lock);
	if (!vblk->config_enable)
		goto done;

324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345
	/* Host must always specify the capacity. */
	vdev->config->get(vdev, offsetof(struct virtio_blk_config, capacity),
			  &capacity, sizeof(capacity));

	/* If capacity is too big, truncate with warning. */
	if ((sector_t)capacity != capacity) {
		dev_warn(&vdev->dev, "Capacity %llu too large: truncating\n",
			 (unsigned long long)capacity);
		capacity = (sector_t)-1;
	}

	size = capacity * queue_logical_block_size(q);
	string_get_size(size, STRING_UNITS_2, cap_str_2, sizeof(cap_str_2));
	string_get_size(size, STRING_UNITS_10, cap_str_10, sizeof(cap_str_10));

	dev_notice(&vdev->dev,
		  "new size: %llu %d-byte logical blocks (%s/%s)\n",
		  (unsigned long long)capacity,
		  queue_logical_block_size(q),
		  cap_str_10, cap_str_2);

	set_capacity(vblk->disk, capacity);
346
	revalidate_disk(vblk->disk);
347 348
done:
	mutex_unlock(&vblk->config_lock);
349 350 351 352 353 354 355 356 357
}

static void virtblk_config_changed(struct virtio_device *vdev)
{
	struct virtio_blk *vblk = vdev->priv;

	queue_work(virtblk_wq, &vblk->config_work);
}

358 359 360 361 362 363 364 365 366 367 368 369
static int init_vq(struct virtio_blk *vblk)
{
	int err = 0;

	/* We expect one virtqueue, for output. */
	vblk->vq = virtio_find_single_vq(vblk->vdev, blk_done, "requests");
	if (IS_ERR(vblk->vq))
		err = PTR_ERR(vblk->vq);

	return err;
}

370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397
/*
 * Legacy naming scheme used for virtio devices.  We are stuck with it for
 * virtio blk but don't ever use it for any new driver.
 */
static int virtblk_name_format(char *prefix, int index, char *buf, int buflen)
{
	const int base = 'z' - 'a' + 1;
	char *begin = buf + strlen(prefix);
	char *end = buf + buflen;
	char *p;
	int unit;

	p = end - 1;
	*p = '\0';
	unit = base;
	do {
		if (p == begin)
			return -EINVAL;
		*--p = 'a' + (index % unit);
		index = (index / unit) - 1;
	} while (index >= 0);

	memmove(begin, p, end - p);
	memcpy(buf, prefix, strlen(prefix));

	return 0;
}

398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474
static int virtblk_get_cache_mode(struct virtio_device *vdev)
{
	u8 writeback;
	int err;

	err = virtio_config_val(vdev, VIRTIO_BLK_F_CONFIG_WCE,
				offsetof(struct virtio_blk_config, wce),
				&writeback);
	if (err)
		writeback = virtio_has_feature(vdev, VIRTIO_BLK_F_WCE);

	return writeback;
}

static void virtblk_update_cache_mode(struct virtio_device *vdev)
{
	u8 writeback = virtblk_get_cache_mode(vdev);
	struct virtio_blk *vblk = vdev->priv;

	if (writeback)
		blk_queue_flush(vblk->disk->queue, REQ_FLUSH);
	else
		blk_queue_flush(vblk->disk->queue, 0);

	revalidate_disk(vblk->disk);
}

static const char *const virtblk_cache_types[] = {
	"write through", "write back"
};

static ssize_t
virtblk_cache_type_store(struct device *dev, struct device_attribute *attr,
			 const char *buf, size_t count)
{
	struct gendisk *disk = dev_to_disk(dev);
	struct virtio_blk *vblk = disk->private_data;
	struct virtio_device *vdev = vblk->vdev;
	int i;
	u8 writeback;

	BUG_ON(!virtio_has_feature(vblk->vdev, VIRTIO_BLK_F_CONFIG_WCE));
	for (i = ARRAY_SIZE(virtblk_cache_types); --i >= 0; )
		if (sysfs_streq(buf, virtblk_cache_types[i]))
			break;

	if (i < 0)
		return -EINVAL;

	writeback = i;
	vdev->config->set(vdev,
			  offsetof(struct virtio_blk_config, wce),
			  &writeback, sizeof(writeback));

	virtblk_update_cache_mode(vdev);
	return count;
}

static ssize_t
virtblk_cache_type_show(struct device *dev, struct device_attribute *attr,
			 char *buf)
{
	struct gendisk *disk = dev_to_disk(dev);
	struct virtio_blk *vblk = disk->private_data;
	u8 writeback = virtblk_get_cache_mode(vblk->vdev);

	BUG_ON(writeback >= ARRAY_SIZE(virtblk_cache_types));
	return snprintf(buf, 40, "%s\n", virtblk_cache_types[writeback]);
}

static const struct device_attribute dev_attr_cache_type_ro =
	__ATTR(cache_type, S_IRUGO,
	       virtblk_cache_type_show, NULL);
static const struct device_attribute dev_attr_cache_type_rw =
	__ATTR(cache_type, S_IRUGO|S_IWUSR,
	       virtblk_cache_type_show, virtblk_cache_type_store);

475
static int __devinit virtblk_probe(struct virtio_device *vdev)
R
Rusty Russell 已提交
476 477
{
	struct virtio_blk *vblk;
478
	struct request_queue *q;
479
	int err, index;
R
Rusty Russell 已提交
480
	u64 cap;
481 482 483
	u32 v, blk_size, sg_elems, opt_io_size;
	u16 min_io_size;
	u8 physical_block_exp, alignment_offset;
R
Rusty Russell 已提交
484

485 486 487 488 489
	err = ida_simple_get(&vd_index_ida, 0, minor_to_index(1 << MINORBITS),
			     GFP_KERNEL);
	if (err < 0)
		goto out;
	index = err;
490

491 492 493 494
	/* We need to know how many segments before we allocate. */
	err = virtio_config_val(vdev, VIRTIO_BLK_F_SEG_MAX,
				offsetof(struct virtio_blk_config, seg_max),
				&sg_elems);
495 496 497

	/* We need at least one SG element, whatever they say. */
	if (err || !sg_elems)
498 499 500 501 502 503
		sg_elems = 1;

	/* We need an extra sg elements at head and tail. */
	sg_elems += 2;
	vdev->priv = vblk = kmalloc(sizeof(*vblk) +
				    sizeof(vblk->sg[0]) * sg_elems, GFP_KERNEL);
R
Rusty Russell 已提交
504 505
	if (!vblk) {
		err = -ENOMEM;
506
		goto out_free_index;
R
Rusty Russell 已提交
507 508 509
	}

	vblk->vdev = vdev;
510 511
	vblk->sg_elems = sg_elems;
	sg_init_table(vblk->sg, vblk->sg_elems);
512
	mutex_init(&vblk->config_lock);
513
	INIT_WORK(&vblk->config_work, virtblk_config_changed_work);
514
	vblk->config_enable = true;
R
Rusty Russell 已提交
515

516 517
	err = init_vq(vblk);
	if (err)
R
Rusty Russell 已提交
518 519 520 521 522 523 524 525 526
		goto out_free_vblk;

	vblk->pool = mempool_create_kmalloc_pool(1,sizeof(struct virtblk_req));
	if (!vblk->pool) {
		err = -ENOMEM;
		goto out_free_vq;
	}

	/* FIXME: How many partitions?  How long is a piece of string? */
527
	vblk->disk = alloc_disk(1 << PART_BITS);
R
Rusty Russell 已提交
528 529
	if (!vblk->disk) {
		err = -ENOMEM;
530
		goto out_mempool;
R
Rusty Russell 已提交
531 532
	}

533
	q = vblk->disk->queue = blk_init_queue(do_virtblk_request, NULL);
534
	if (!q) {
R
Rusty Russell 已提交
535 536 537 538
		err = -ENOMEM;
		goto out_put_disk;
	}

539
	q->queuedata = vblk;
540

541
	virtblk_name_format("vd", index, vblk->disk->disk_name, DISK_NAME_LEN);
542

R
Rusty Russell 已提交
543
	vblk->disk->major = major;
544
	vblk->disk->first_minor = index_to_minor(index);
R
Rusty Russell 已提交
545 546
	vblk->disk->private_data = vblk;
	vblk->disk->fops = &virtblk_fops;
547
	vblk->disk->driverfs_dev = &vdev->dev;
548
	vblk->index = index;
549

550
	/* configure queue flush support */
551
	virtblk_update_cache_mode(vdev);
R
Rusty Russell 已提交
552

553 554 555 556
	/* If disk is read-only in the host, the guest should obey */
	if (virtio_has_feature(vdev, VIRTIO_BLK_F_RO))
		set_disk_ro(vblk->disk, 1);

557
	/* Host must always specify the capacity. */
558 559
	vdev->config->get(vdev, offsetof(struct virtio_blk_config, capacity),
			  &cap, sizeof(cap));
R
Rusty Russell 已提交
560 561 562 563 564 565 566 567 568

	/* If capacity is too big, truncate with warning. */
	if ((sector_t)cap != cap) {
		dev_warn(&vdev->dev, "Capacity %llu too large: truncating\n",
			 (unsigned long long)cap);
		cap = (sector_t)-1;
	}
	set_capacity(vblk->disk, cap);

569
	/* We can handle whatever the host told us to handle. */
570
	blk_queue_max_segments(q, vblk->sg_elems-2);
571

572
	/* No need to bounce any requests */
573
	blk_queue_bounce_limit(q, BLK_BOUNCE_ANY);
574

575
	/* No real sector limit. */
576
	blk_queue_max_hw_sectors(q, -1U);
577

578 579 580 581 582
	/* Host can optionally specify maximum segment size and number of
	 * segments. */
	err = virtio_config_val(vdev, VIRTIO_BLK_F_SIZE_MAX,
				offsetof(struct virtio_blk_config, size_max),
				&v);
R
Rusty Russell 已提交
583
	if (!err)
584
		blk_queue_max_segment_size(q, v);
585
	else
586
		blk_queue_max_segment_size(q, -1U);
R
Rusty Russell 已提交
587

588 589 590 591 592
	/* Host can optionally specify the block size of the device */
	err = virtio_config_val(vdev, VIRTIO_BLK_F_BLK_SIZE,
				offsetof(struct virtio_blk_config, blk_size),
				&blk_size);
	if (!err)
593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622
		blk_queue_logical_block_size(q, blk_size);
	else
		blk_size = queue_logical_block_size(q);

	/* Use topology information if available */
	err = virtio_config_val(vdev, VIRTIO_BLK_F_TOPOLOGY,
			offsetof(struct virtio_blk_config, physical_block_exp),
			&physical_block_exp);
	if (!err && physical_block_exp)
		blk_queue_physical_block_size(q,
				blk_size * (1 << physical_block_exp));

	err = virtio_config_val(vdev, VIRTIO_BLK_F_TOPOLOGY,
			offsetof(struct virtio_blk_config, alignment_offset),
			&alignment_offset);
	if (!err && alignment_offset)
		blk_queue_alignment_offset(q, blk_size * alignment_offset);

	err = virtio_config_val(vdev, VIRTIO_BLK_F_TOPOLOGY,
			offsetof(struct virtio_blk_config, min_io_size),
			&min_io_size);
	if (!err && min_io_size)
		blk_queue_io_min(q, blk_size * min_io_size);

	err = virtio_config_val(vdev, VIRTIO_BLK_F_TOPOLOGY,
			offsetof(struct virtio_blk_config, opt_io_size),
			&opt_io_size);
	if (!err && opt_io_size)
		blk_queue_io_opt(q, blk_size * opt_io_size);

623

R
Rusty Russell 已提交
624
	add_disk(vblk->disk);
625 626 627 628
	err = device_create_file(disk_to_dev(vblk->disk), &dev_attr_serial);
	if (err)
		goto out_del_disk;

629 630 631 632 633 634 635 636
	if (virtio_has_feature(vdev, VIRTIO_BLK_F_CONFIG_WCE))
		err = device_create_file(disk_to_dev(vblk->disk),
					 &dev_attr_cache_type_rw);
	else
		err = device_create_file(disk_to_dev(vblk->disk),
					 &dev_attr_cache_type_ro);
	if (err)
		goto out_del_disk;
R
Rusty Russell 已提交
637 638
	return 0;

639 640 641
out_del_disk:
	del_gendisk(vblk->disk);
	blk_cleanup_queue(vblk->disk->queue);
R
Rusty Russell 已提交
642 643 644 645 646
out_put_disk:
	put_disk(vblk->disk);
out_mempool:
	mempool_destroy(vblk->pool);
out_free_vq:
647
	vdev->config->del_vqs(vdev);
R
Rusty Russell 已提交
648 649
out_free_vblk:
	kfree(vblk);
650 651
out_free_index:
	ida_simple_remove(&vd_index_ida, index);
R
Rusty Russell 已提交
652 653 654 655
out:
	return err;
}

656
static void __devexit virtblk_remove(struct virtio_device *vdev)
R
Rusty Russell 已提交
657 658
{
	struct virtio_blk *vblk = vdev->priv;
659
	int index = vblk->index;
R
Rusty Russell 已提交
660

661 662 663 664
	/* Prevent config work handler from accessing the device. */
	mutex_lock(&vblk->config_lock);
	vblk->config_enable = false;
	mutex_unlock(&vblk->config_lock);
665

666
	del_gendisk(vblk->disk);
667
	blk_cleanup_queue(vblk->disk->queue);
668

R
Rusty Russell 已提交
669 670 671
	/* Stop all the virtqueues. */
	vdev->config->reset(vdev);

672 673
	flush_work(&vblk->config_work);

R
Rusty Russell 已提交
674 675
	put_disk(vblk->disk);
	mempool_destroy(vblk->pool);
676
	vdev->config->del_vqs(vdev);
R
Rusty Russell 已提交
677
	kfree(vblk);
678
	ida_simple_remove(&vd_index_ida, index);
R
Rusty Russell 已提交
679 680
}

681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720
#ifdef CONFIG_PM
static int virtblk_freeze(struct virtio_device *vdev)
{
	struct virtio_blk *vblk = vdev->priv;

	/* Ensure we don't receive any more interrupts */
	vdev->config->reset(vdev);

	/* Prevent config work handler from accessing the device. */
	mutex_lock(&vblk->config_lock);
	vblk->config_enable = false;
	mutex_unlock(&vblk->config_lock);

	flush_work(&vblk->config_work);

	spin_lock_irq(vblk->disk->queue->queue_lock);
	blk_stop_queue(vblk->disk->queue);
	spin_unlock_irq(vblk->disk->queue->queue_lock);
	blk_sync_queue(vblk->disk->queue);

	vdev->config->del_vqs(vdev);
	return 0;
}

static int virtblk_restore(struct virtio_device *vdev)
{
	struct virtio_blk *vblk = vdev->priv;
	int ret;

	vblk->config_enable = true;
	ret = init_vq(vdev->priv);
	if (!ret) {
		spin_lock_irq(vblk->disk->queue->queue_lock);
		blk_start_queue(vblk->disk->queue);
		spin_unlock_irq(vblk->disk->queue->queue_lock);
	}
	return ret;
}
#endif

721
static const struct virtio_device_id id_table[] = {
R
Rusty Russell 已提交
722 723 724 725
	{ VIRTIO_ID_BLOCK, VIRTIO_DEV_ANY_ID },
	{ 0 },
};

726
static unsigned int features[] = {
727 728
	VIRTIO_BLK_F_SEG_MAX, VIRTIO_BLK_F_SIZE_MAX, VIRTIO_BLK_F_GEOMETRY,
	VIRTIO_BLK_F_RO, VIRTIO_BLK_F_BLK_SIZE, VIRTIO_BLK_F_SCSI,
729
	VIRTIO_BLK_F_WCE, VIRTIO_BLK_F_TOPOLOGY, VIRTIO_BLK_F_CONFIG_WCE
730 731
};

732 733 734 735 736 737
/*
 * virtio_blk causes spurious section mismatch warning by
 * simultaneously referring to a __devinit and a __devexit function.
 * Use __refdata to avoid this warning.
 */
static struct virtio_driver __refdata virtio_blk = {
738 739 740 741 742 743 744 745
	.feature_table		= features,
	.feature_table_size	= ARRAY_SIZE(features),
	.driver.name		= KBUILD_MODNAME,
	.driver.owner		= THIS_MODULE,
	.id_table		= id_table,
	.probe			= virtblk_probe,
	.remove			= __devexit_p(virtblk_remove),
	.config_changed		= virtblk_config_changed,
746 747 748 749
#ifdef CONFIG_PM
	.freeze			= virtblk_freeze,
	.restore		= virtblk_restore,
#endif
R
Rusty Russell 已提交
750 751 752 753
};

static int __init init(void)
{
754 755 756 757 758 759
	int error;

	virtblk_wq = alloc_workqueue("virtio-blk", 0, 0);
	if (!virtblk_wq)
		return -ENOMEM;

760
	major = register_blkdev(0, "virtblk");
761 762 763 764 765 766 767 768 769 770 771 772 773 774 775
	if (major < 0) {
		error = major;
		goto out_destroy_workqueue;
	}

	error = register_virtio_driver(&virtio_blk);
	if (error)
		goto out_unregister_blkdev;
	return 0;

out_unregister_blkdev:
	unregister_blkdev(major, "virtblk");
out_destroy_workqueue:
	destroy_workqueue(virtblk_wq);
	return error;
R
Rusty Russell 已提交
776 777 778 779
}

static void __exit fini(void)
{
780
	unregister_blkdev(major, "virtblk");
R
Rusty Russell 已提交
781
	unregister_virtio_driver(&virtio_blk);
782
	destroy_workqueue(virtblk_wq);
R
Rusty Russell 已提交
783 784 785 786 787 788 789
}
module_init(init);
module_exit(fini);

MODULE_DEVICE_TABLE(virtio, id_table);
MODULE_DESCRIPTION("Virtio block driver");
MODULE_LICENSE("GPL");