xen-blkfront.c 37.4 KB
Newer Older
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
/*
 * blkfront.c
 *
 * XenLinux virtual block device driver.
 *
 * Copyright (c) 2003-2004, Keir Fraser & Steve Hand
 * Modifications by Mark A. Williamson are (c) Intel Research Cambridge
 * Copyright (c) 2004, Christian Limpach
 * Copyright (c) 2004, Andrew Warfield
 * Copyright (c) 2005, Christopher Clark
 * Copyright (c) 2005, XenSource Ltd
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of the GNU General Public License version 2
 * as published by the Free Software Foundation; or, when distributed
 * separately from the Linux kernel or incorporated into other
 * software packages, subject to the following license:
 *
 * Permission is hereby granted, free of charge, to any person obtaining a copy
 * of this source file (the "Software"), to deal in the Software without
 * restriction, including without limitation the rights to use, copy, modify,
 * merge, publish, distribute, sublicense, and/or sell copies of the Software,
 * and to permit persons to whom the Software is furnished to do so, subject to
 * the following conditions:
 *
 * The above copyright notice and this permission notice shall be included in
 * all copies or substantial portions of the Software.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
 * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
 * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
 * IN THE SOFTWARE.
 */

#include <linux/interrupt.h>
#include <linux/blkdev.h>
40
#include <linux/hdreg.h>
41
#include <linux/cdrom.h>
42
#include <linux/module.h>
43
#include <linux/slab.h>
44
#include <linux/mutex.h>
45
#include <linux/scatterlist.h>
46

47
#include <xen/xen.h>
48
49
50
51
#include <xen/xenbus.h>
#include <xen/grant_table.h>
#include <xen/events.h>
#include <xen/page.h>
52
#include <xen/platform_pci.h>
53
54
55

#include <xen/interface/grant_table.h>
#include <xen/interface/io/blkif.h>
56
#include <xen/interface/io/protocols.h>
57
58
59
60
61
62
63
64
65
66
67

#include <asm/xen/hypervisor.h>

enum blkif_state {
	BLKIF_STATE_DISCONNECTED,
	BLKIF_STATE_CONNECTED,
	BLKIF_STATE_SUSPENDED,
};

struct blk_shadow {
	struct blkif_request req;
68
	struct request *request;
69
70
71
	unsigned long frame[BLKIF_MAX_SEGMENTS_PER_REQUEST];
};

72
static DEFINE_MUTEX(blkfront_mutex);
73
static const struct block_device_operations xlvbd_block_fops;
74

75
#define BLK_RING_SIZE __CONST_RING_SIZE(blkif, PAGE_SIZE)
76
77
78
79
80
81
82
83

/*
 * We have one of these per vbd, whether ide, scsi or 'other'.  They
 * hang in private_data off the gendisk structure. We may end up
 * putting all kinds of interesting stuff here :-)
 */
struct blkfront_info
{
84
	struct mutex mutex;
85
86
87
88
89
90
91
	struct xenbus_device *xbdev;
	struct gendisk *gd;
	int vdevice;
	blkif_vdev_t handle;
	enum blkif_state connected;
	int ring_ref;
	struct blkif_front_ring ring;
92
	struct scatterlist sg[BLKIF_MAX_SEGMENTS_PER_REQUEST];
93
94
95
96
97
98
	unsigned int evtchn, irq;
	struct request_queue *rq;
	struct work_struct work;
	struct gnttab_free_callback callback;
	struct blk_shadow shadow[BLK_RING_SIZE];
	unsigned long shadow_free;
99
	unsigned int feature_flush;
100
	unsigned int flush_op;
101
102
	unsigned int feature_discard:1;
	unsigned int feature_secdiscard:1;
103
104
	unsigned int discard_granularity;
	unsigned int discard_alignment;
105
	int is_ready;
106
107
108
109
};

static DEFINE_SPINLOCK(blkif_io_lock);

110
111
112
113
static unsigned int nr_minors;
static unsigned long *minors;
static DEFINE_SPINLOCK(minor_lock);

114
115
116
117
118
#define MAXIMUM_OUTSTANDING_BLOCK_REQS \
	(BLKIF_MAX_SEGMENTS_PER_REQUEST * BLK_RING_SIZE)
#define GRANT_INVALID_REF	0

#define PARTS_PER_DISK		16
119
#define PARTS_PER_EXT_DISK      256
120
121
122
123

#define BLKIF_MAJOR(dev) ((dev)>>8)
#define BLKIF_MINOR(dev) ((dev) & 0xff)

124
125
126
127
#define EXT_SHIFT 28
#define EXTENDED (1<<EXT_SHIFT)
#define VDEV_IS_EXTENDED(dev) ((dev)&(EXTENDED))
#define BLKIF_MINOR_EXT(dev) ((dev)&(~EXTENDED))
128
129
#define EMULATED_HD_DISK_MINOR_OFFSET (0)
#define EMULATED_HD_DISK_NAME_OFFSET (EMULATED_HD_DISK_MINOR_OFFSET / 256)
130
131
#define EMULATED_SD_DISK_MINOR_OFFSET (0)
#define EMULATED_SD_DISK_NAME_OFFSET (EMULATED_SD_DISK_MINOR_OFFSET / 256)
132

133
#define DEV_NAME	"xvd"	/* name in /dev */
134
135
136
137

static int get_id_from_freelist(struct blkfront_info *info)
{
	unsigned long free = info->shadow_free;
138
	BUG_ON(free >= BLK_RING_SIZE);
139
140
	info->shadow_free = info->shadow[free].req.u.rw.id;
	info->shadow[free].req.u.rw.id = 0x0fffffee; /* debug */
141
142
143
144
145
146
	return free;
}

static void add_id_to_freelist(struct blkfront_info *info,
			       unsigned long id)
{
147
	info->shadow[id].req.u.rw.id  = info->shadow_free;
148
	info->shadow[id].request = NULL;
149
150
151
	info->shadow_free = id;
}

152
153
154
155
156
157
158
159
static int xlbd_reserve_minors(unsigned int minor, unsigned int nr)
{
	unsigned int end = minor + nr;
	int rc;

	if (end > nr_minors) {
		unsigned long *bitmap, *old;

160
		bitmap = kcalloc(BITS_TO_LONGS(end), sizeof(*bitmap),
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
				 GFP_KERNEL);
		if (bitmap == NULL)
			return -ENOMEM;

		spin_lock(&minor_lock);
		if (end > nr_minors) {
			old = minors;
			memcpy(bitmap, minors,
			       BITS_TO_LONGS(nr_minors) * sizeof(*bitmap));
			minors = bitmap;
			nr_minors = BITS_TO_LONGS(end) * BITS_PER_LONG;
		} else
			old = bitmap;
		spin_unlock(&minor_lock);
		kfree(old);
	}

	spin_lock(&minor_lock);
	if (find_next_bit(minors, end, minor) >= end) {
		for (; minor < end; ++minor)
			__set_bit(minor, minors);
		rc = 0;
	} else
		rc = -EBUSY;
	spin_unlock(&minor_lock);

	return rc;
}

static void xlbd_release_minors(unsigned int minor, unsigned int nr)
{
	unsigned int end = minor + nr;

	BUG_ON(end > nr_minors);
	spin_lock(&minor_lock);
	for (; minor < end; ++minor)
		__clear_bit(minor, minors);
	spin_unlock(&minor_lock);
}

201
202
203
204
205
206
static void blkif_restart_queue_callback(void *arg)
{
	struct blkfront_info *info = (struct blkfront_info *)arg;
	schedule_work(&info->work);
}

207
static int blkif_getgeo(struct block_device *bd, struct hd_geometry *hg)
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
{
	/* We don't have real geometry info, but let's at least return
	   values consistent with the size of the device */
	sector_t nsect = get_capacity(bd->bd_disk);
	sector_t cylinders = nsect;

	hg->heads = 0xff;
	hg->sectors = 0x3f;
	sector_div(cylinders, hg->heads * hg->sectors);
	hg->cylinders = cylinders;
	if ((sector_t)(hg->cylinders + 1) * hg->heads * hg->sectors < nsect)
		hg->cylinders = 0xffff;
	return 0;
}

Al Viro's avatar
Al Viro committed
223
static int blkif_ioctl(struct block_device *bdev, fmode_t mode,
224
		       unsigned command, unsigned long argument)
225
{
Al Viro's avatar
Al Viro committed
226
	struct blkfront_info *info = bdev->bd_disk->private_data;
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
	int i;

	dev_dbg(&info->xbdev->dev, "command: 0x%x, argument: 0x%lx\n",
		command, (long)argument);

	switch (command) {
	case CDROMMULTISESSION:
		dev_dbg(&info->xbdev->dev, "FIXME: support multisession CDs later\n");
		for (i = 0; i < sizeof(struct cdrom_multisession); i++)
			if (put_user(0, (char __user *)(argument + i)))
				return -EFAULT;
		return 0;

	case CDROM_GET_CAPABILITY: {
		struct gendisk *gd = info->gd;
		if (gd->flags & GENHD_FL_CD)
			return 0;
		return -EINVAL;
	}

	default:
		/*printk(KERN_ALERT "ioctl %08x not supported by Xen blkdev\n",
		  command);*/
		return -EINVAL; /* same return as native Linux */
	}

	return 0;
}

256
/*
257
 * Generate a Xen blkfront IO request from a blk layer request.  Reads
258
 * and writes are handled as expected.
259
 *
260
 * @req: a request struct
261
262
263
264
265
266
267
268
 */
static int blkif_queue_request(struct request *req)
{
	struct blkfront_info *info = req->rq_disk->private_data;
	unsigned long buffer_mfn;
	struct blkif_request *ring_req;
	unsigned long id;
	unsigned int fsect, lsect;
269
	int i, ref;
270
	grant_ref_t gref_head;
271
	struct scatterlist *sg;
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288

	if (unlikely(info->connected != BLKIF_STATE_CONNECTED))
		return 1;

	if (gnttab_alloc_grant_references(
		BLKIF_MAX_SEGMENTS_PER_REQUEST, &gref_head) < 0) {
		gnttab_request_free_callback(
			&info->callback,
			blkif_restart_queue_callback,
			info,
			BLKIF_MAX_SEGMENTS_PER_REQUEST);
		return 1;
	}

	/* Fill out a communications ring structure. */
	ring_req = RING_GET_REQUEST(&info->ring, info->ring.req_prod_pvt);
	id = get_id_from_freelist(info);
289
	info->shadow[id].request = req;
290

291
	ring_req->u.rw.id = id;
292
	ring_req->u.rw.sector_number = (blkif_sector_t)blk_rq_pos(req);
293
	ring_req->u.rw.handle = info->handle;
294
295
296

	ring_req->operation = rq_data_dir(req) ?
		BLKIF_OP_WRITE : BLKIF_OP_READ;
297
298
299

	if (req->cmd_flags & (REQ_FLUSH | REQ_FUA)) {
		/*
300
301
		 * Ideally we can do an unordered flush-to-disk. In case the
		 * backend onlysupports barriers, use that. A barrier request
302
303
304
305
		 * a superset of FUA, so we can implement it the same
		 * way.  (It's also a FLUSH+FUA, since it is
		 * guaranteed ordered WRT previous writes.)
		 */
306
		ring_req->operation = info->flush_op;
307
	}
308

309
	if (unlikely(req->cmd_flags & (REQ_DISCARD | REQ_SECURE))) {
310
311
312
		/* id, sector_number and handle are set above. */
		ring_req->operation = BLKIF_OP_DISCARD;
		ring_req->u.discard.nr_sectors = blk_rq_sectors(req);
313
314
315
316
		if ((req->cmd_flags & REQ_SECURE) && info->feature_secdiscard)
			ring_req->u.discard.flag = BLKIF_DISCARD_SECURE;
		else
			ring_req->u.discard.flag = 0;
317
	} else {
318
319
320
321
		ring_req->u.rw.nr_segments = blk_rq_map_sg(req->q, req,
							   info->sg);
		BUG_ON(ring_req->u.rw.nr_segments >
		       BLKIF_MAX_SEGMENTS_PER_REQUEST);
322

323
		for_each_sg(info->sg, sg, ring_req->u.rw.nr_segments, i) {
324
325
326
327
328
329
			buffer_mfn = pfn_to_mfn(page_to_pfn(sg_page(sg)));
			fsect = sg->offset >> 9;
			lsect = fsect + (sg->length >> 9) - 1;
			/* install a grant reference. */
			ref = gnttab_claim_grant_reference(&gref_head);
			BUG_ON(ref == -ENOSPC);
330

331
332
333
334
335
336
337
338
339
340
341
342
343
			gnttab_grant_foreign_access_ref(
					ref,
					info->xbdev->otherend_id,
					buffer_mfn,
					rq_data_dir(req));

			info->shadow[id].frame[i] = mfn_to_pfn(buffer_mfn);
			ring_req->u.rw.seg[i] =
					(struct blkif_request_segment) {
						.gref       = ref,
						.first_sect = fsect,
						.last_sect  = lsect };
		}
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
	}

	info->ring.req_prod_pvt++;

	/* Keep a private copy so we can reissue requests when recovering. */
	info->shadow[id].req = *ring_req;

	gnttab_free_grant_references(gref_head);

	return 0;
}


static inline void flush_requests(struct blkfront_info *info)
{
	int notify;

	RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(&info->ring, notify);

	if (notify)
		notify_remote_via_irq(info->irq);
}

/*
 * do_blkif_request
 *  read a block; request is in a request queue
 */
371
static void do_blkif_request(struct request_queue *rq)
372
373
374
375
376
377
378
379
380
{
	struct blkfront_info *info = NULL;
	struct request *req;
	int queued;

	pr_debug("Entered do_blkif_request\n");

	queued = 0;

381
	while ((req = blk_peek_request(rq)) != NULL) {
382
383
384
385
386
		info = req->rq_disk->private_data;

		if (RING_FULL(&info->ring))
			goto wait;

387
		blk_start_request(req);
388

389
390
391
		if ((req->cmd_type != REQ_TYPE_FS) ||
		    ((req->cmd_flags & (REQ_FLUSH | REQ_FUA)) &&
		    !info->flush_op)) {
392
393
394
395
			__blk_end_request_all(req, -EIO);
			continue;
		}

396
		pr_debug("do_blk_req %p: cmd %p, sec %lx, "
397
398
399
400
			 "(%u/%u) buffer:%p [%s]\n",
			 req, req->cmd, (unsigned long)blk_rq_pos(req),
			 blk_rq_cur_sectors(req), blk_rq_sectors(req),
			 req->buffer, rq_data_dir(req) ? "write" : "read");
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418

		if (blkif_queue_request(req)) {
			blk_requeue_request(rq, req);
wait:
			/* Avoid pointless unplugs. */
			blk_stop_queue(rq);
			break;
		}

		queued++;
	}

	if (queued != 0)
		flush_requests(info);
}

static int xlvbd_init_blk_queue(struct gendisk *gd, u16 sector_size)
{
419
	struct request_queue *rq;
420
	struct blkfront_info *info = gd->private_data;
421
422
423
424
425

	rq = blk_init_queue(do_blkif_request, &blkif_io_lock);
	if (rq == NULL)
		return -1;

426
	queue_flag_set_unlocked(QUEUE_FLAG_VIRT, rq);
427

428
429
430
431
432
	if (info->feature_discard) {
		queue_flag_set_unlocked(QUEUE_FLAG_DISCARD, rq);
		blk_queue_max_discard_sectors(rq, get_capacity(gd));
		rq->limits.discard_granularity = info->discard_granularity;
		rq->limits.discard_alignment = info->discard_alignment;
433
434
		if (info->feature_secdiscard)
			queue_flag_set_unlocked(QUEUE_FLAG_SECDISCARD, rq);
435
436
	}

437
	/* Hard sector size and max sectors impersonate the equiv. hardware. */
438
	blk_queue_logical_block_size(rq, sector_size);
439
	blk_queue_max_hw_sectors(rq, 512);
440
441
442
443
444
445

	/* Each segment in a request is up to an aligned page in size. */
	blk_queue_segment_boundary(rq, PAGE_SIZE - 1);
	blk_queue_max_segment_size(rq, PAGE_SIZE);

	/* Ensure a merged request will fit in a single I/O ring slot. */
446
	blk_queue_max_segments(rq, BLKIF_MAX_SEGMENTS_PER_REQUEST);
447
448
449
450

	/* Make sure buffer addresses are sector-aligned. */
	blk_queue_dma_alignment(rq, 511);

451
452
453
	/* Make sure we don't use bounce buffers. */
	blk_queue_bounce_limit(rq, BLK_BOUNCE_ANY);

454
455
456
457
458
459
	gd->queue = rq;

	return 0;
}


460
static void xlvbd_flush(struct blkfront_info *info)
461
{
462
	blk_queue_flush(info->rq, info->feature_flush);
463
	printk(KERN_INFO "blkfront: %s: %s: %s\n",
464
	       info->gd->disk_name,
465
466
467
	       info->flush_op == BLKIF_OP_WRITE_BARRIER ?
		"barrier" : (info->flush_op == BLKIF_OP_FLUSH_DISKCACHE ?
		"flush diskcache" : "barrier or flush"),
468
	       info->feature_flush ? "enabled" : "disabled");
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
static int xen_translate_vdev(int vdevice, int *minor, unsigned int *offset)
{
	int major;
	major = BLKIF_MAJOR(vdevice);
	*minor = BLKIF_MINOR(vdevice);
	switch (major) {
		case XEN_IDE0_MAJOR:
			*offset = (*minor / 64) + EMULATED_HD_DISK_NAME_OFFSET;
			*minor = ((*minor / 64) * PARTS_PER_DISK) +
				EMULATED_HD_DISK_MINOR_OFFSET;
			break;
		case XEN_IDE1_MAJOR:
			*offset = (*minor / 64) + 2 + EMULATED_HD_DISK_NAME_OFFSET;
			*minor = (((*minor / 64) + 2) * PARTS_PER_DISK) +
				EMULATED_HD_DISK_MINOR_OFFSET;
			break;
		case XEN_SCSI_DISK0_MAJOR:
			*offset = (*minor / PARTS_PER_DISK) + EMULATED_SD_DISK_NAME_OFFSET;
			*minor = *minor + EMULATED_SD_DISK_MINOR_OFFSET;
			break;
		case XEN_SCSI_DISK1_MAJOR:
		case XEN_SCSI_DISK2_MAJOR:
		case XEN_SCSI_DISK3_MAJOR:
		case XEN_SCSI_DISK4_MAJOR:
		case XEN_SCSI_DISK5_MAJOR:
		case XEN_SCSI_DISK6_MAJOR:
		case XEN_SCSI_DISK7_MAJOR:
			*offset = (*minor / PARTS_PER_DISK) + 
				((major - XEN_SCSI_DISK1_MAJOR + 1) * 16) +
				EMULATED_SD_DISK_NAME_OFFSET;
			*minor = *minor +
				((major - XEN_SCSI_DISK1_MAJOR + 1) * 16 * PARTS_PER_DISK) +
				EMULATED_SD_DISK_MINOR_OFFSET;
			break;
		case XEN_SCSI_DISK8_MAJOR:
		case XEN_SCSI_DISK9_MAJOR:
		case XEN_SCSI_DISK10_MAJOR:
		case XEN_SCSI_DISK11_MAJOR:
		case XEN_SCSI_DISK12_MAJOR:
		case XEN_SCSI_DISK13_MAJOR:
		case XEN_SCSI_DISK14_MAJOR:
		case XEN_SCSI_DISK15_MAJOR:
			*offset = (*minor / PARTS_PER_DISK) + 
				((major - XEN_SCSI_DISK8_MAJOR + 8) * 16) +
				EMULATED_SD_DISK_NAME_OFFSET;
			*minor = *minor +
				((major - XEN_SCSI_DISK8_MAJOR + 8) * 16 * PARTS_PER_DISK) +
				EMULATED_SD_DISK_MINOR_OFFSET;
			break;
		case XENVBD_MAJOR:
			*offset = *minor / PARTS_PER_DISK;
			break;
		default:
			printk(KERN_WARNING "blkfront: your disk configuration is "
					"incorrect, please use an xvd device instead\n");
			return -ENODEV;
	}
	return 0;
}
530

531
532
533
static int xlvbd_alloc_gendisk(blkif_sector_t capacity,
			       struct blkfront_info *info,
			       u16 vdisk_info, u16 sector_size)
534
535
536
{
	struct gendisk *gd;
	int nr_minors = 1;
537
	int err;
538
539
540
	unsigned int offset;
	int minor;
	int nr_parts;
541
542
543
544

	BUG_ON(info->gd != NULL);
	BUG_ON(info->rq != NULL);

545
546
547
548
549
550
551
	if ((info->vdevice>>EXT_SHIFT) > 1) {
		/* this is above the extended range; something is wrong */
		printk(KERN_WARNING "blkfront: vdevice 0x%x is above the extended range; ignoring\n", info->vdevice);
		return -ENODEV;
	}

	if (!VDEV_IS_EXTENDED(info->vdevice)) {
552
553
554
555
		err = xen_translate_vdev(info->vdevice, &minor, &offset);
		if (err)
			return err;		
 		nr_parts = PARTS_PER_DISK;
556
557
558
	} else {
		minor = BLKIF_MINOR_EXT(info->vdevice);
		nr_parts = PARTS_PER_EXT_DISK;
559
		offset = minor / nr_parts;
560
		if (xen_hvm_domain() && offset < EMULATED_HD_DISK_NAME_OFFSET + 4)
561
562
563
			printk(KERN_WARNING "blkfront: vdevice 0x%x might conflict with "
					"emulated IDE disks,\n\t choose an xvd device name"
					"from xvde on\n", info->vdevice);
564
	}
565
	err = -ENODEV;
566
567
568

	if ((minor % nr_parts) == 0)
		nr_minors = nr_parts;
569

570
571
572
573
574
	err = xlbd_reserve_minors(minor, nr_minors);
	if (err)
		goto out;
	err = -ENODEV;

575
576
	gd = alloc_disk(nr_minors);
	if (gd == NULL)
577
		goto release;
578

579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
	if (nr_minors > 1) {
		if (offset < 26)
			sprintf(gd->disk_name, "%s%c", DEV_NAME, 'a' + offset);
		else
			sprintf(gd->disk_name, "%s%c%c", DEV_NAME,
				'a' + ((offset / 26)-1), 'a' + (offset % 26));
	} else {
		if (offset < 26)
			sprintf(gd->disk_name, "%s%c%d", DEV_NAME,
				'a' + offset,
				minor & (nr_parts - 1));
		else
			sprintf(gd->disk_name, "%s%c%c%d", DEV_NAME,
				'a' + ((offset / 26) - 1),
				'a' + (offset % 26),
				minor & (nr_parts - 1));
	}
596
597
598
599
600
601
602
603
604
605

	gd->major = XENVBD_MAJOR;
	gd->first_minor = minor;
	gd->fops = &xlvbd_block_fops;
	gd->private_data = info;
	gd->driverfs_dev = &(info->xbdev->dev);
	set_capacity(gd, capacity);

	if (xlvbd_init_blk_queue(gd, sector_size)) {
		del_gendisk(gd);
606
		goto release;
607
608
609
610
611
	}

	info->rq = gd->queue;
	info->gd = gd;

612
	xlvbd_flush(info);
613
614
615
616
617
618
619
620
621
622
623
624

	if (vdisk_info & VDISK_READONLY)
		set_disk_ro(gd, 1);

	if (vdisk_info & VDISK_REMOVABLE)
		gd->flags |= GENHD_FL_REMOVABLE;

	if (vdisk_info & VDISK_CDROM)
		gd->flags |= GENHD_FL_CD;

	return 0;

625
626
 release:
	xlbd_release_minors(minor, nr_minors);
627
628
629
630
 out:
	return err;
}

631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
static void xlvbd_release_gendisk(struct blkfront_info *info)
{
	unsigned int minor, nr_minors;
	unsigned long flags;

	if (info->rq == NULL)
		return;

	spin_lock_irqsave(&blkif_io_lock, flags);

	/* No more blkif_request(). */
	blk_stop_queue(info->rq);

	/* No more gnttab callback work. */
	gnttab_cancel_free_callback(&info->callback);
	spin_unlock_irqrestore(&blkif_io_lock, flags);

	/* Flush gnttab callback work. Must be done with no locks held. */
649
	flush_work_sync(&info->work);
650
651
652
653
654
655
656
657
658
659
660
661
662
663

	del_gendisk(info->gd);

	minor = info->gd->first_minor;
	nr_minors = info->gd->minors;
	xlbd_release_minors(minor, nr_minors);

	blk_cleanup_queue(info->rq);
	info->rq = NULL;

	put_disk(info->gd);
	info->gd = NULL;
}

664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
static void kick_pending_request_queues(struct blkfront_info *info)
{
	if (!RING_FULL(&info->ring)) {
		/* Re-enable calldowns. */
		blk_start_queue(info->rq);
		/* Kick things off immediately. */
		do_blkif_request(info->rq);
	}
}

static void blkif_restart_queue(struct work_struct *work)
{
	struct blkfront_info *info = container_of(work, struct blkfront_info, work);

	spin_lock_irq(&blkif_io_lock);
	if (info->connected == BLKIF_STATE_CONNECTED)
		kick_pending_request_queues(info);
	spin_unlock_irq(&blkif_io_lock);
}

static void blkif_free(struct blkfront_info *info, int suspend)
{
	/* Prevent new requests being issued until we fix things up. */
	spin_lock_irq(&blkif_io_lock);
	info->connected = suspend ?
		BLKIF_STATE_SUSPENDED : BLKIF_STATE_DISCONNECTED;
	/* No more blkif_request(). */
	if (info->rq)
		blk_stop_queue(info->rq);
	/* No more gnttab callback work. */
	gnttab_cancel_free_callback(&info->callback);
	spin_unlock_irq(&blkif_io_lock);

	/* Flush gnttab callback work. Must be done with no locks held. */
698
	flush_work_sync(&info->work);
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715

	/* Free resources associated with old device channel. */
	if (info->ring_ref != GRANT_INVALID_REF) {
		gnttab_end_foreign_access(info->ring_ref, 0,
					  (unsigned long)info->ring.sring);
		info->ring_ref = GRANT_INVALID_REF;
		info->ring.sring = NULL;
	}
	if (info->irq)
		unbind_from_irqhandler(info->irq, info);
	info->evtchn = info->irq = 0;

}

static void blkif_completion(struct blk_shadow *s)
{
	int i;
716
717
	/* Do not let BLKIF_OP_DISCARD as nr_segment is in the same place
	 * flag. */
718
	for (i = 0; i < s->req.u.rw.nr_segments; i++)
719
		gnttab_end_foreign_access(s->req.u.rw.seg[i].gref, 0, 0UL);
720
721
722
723
724
725
726
727
728
}

static irqreturn_t blkif_interrupt(int irq, void *dev_id)
{
	struct request *req;
	struct blkif_response *bret;
	RING_IDX i, rp;
	unsigned long flags;
	struct blkfront_info *info = (struct blkfront_info *)dev_id;
729
	int error;
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746

	spin_lock_irqsave(&blkif_io_lock, flags);

	if (unlikely(info->connected != BLKIF_STATE_CONNECTED)) {
		spin_unlock_irqrestore(&blkif_io_lock, flags);
		return IRQ_HANDLED;
	}

 again:
	rp = info->ring.sring->rsp_prod;
	rmb(); /* Ensure we see queued responses up to 'rp'. */

	for (i = info->ring.rsp_cons; i != rp; i++) {
		unsigned long id;

		bret = RING_GET_RESPONSE(&info->ring, i);
		id   = bret->id;
747
		req  = info->shadow[id].request;
748

749
750
		if (bret->operation != BLKIF_OP_DISCARD)
			blkif_completion(&info->shadow[id]);
751
752
753

		add_id_to_freelist(info, id);

754
		error = (bret->status == BLKIF_RSP_OKAY) ? 0 : -EIO;
755
		switch (bret->operation) {
756
757
758
759
760
761
762
		case BLKIF_OP_DISCARD:
			if (unlikely(bret->status == BLKIF_RSP_EOPNOTSUPP)) {
				struct request_queue *rq = info->rq;
				printk(KERN_WARNING "blkfront: %s: discard op failed\n",
					   info->gd->disk_name);
				error = -EOPNOTSUPP;
				info->feature_discard = 0;
763
				info->feature_secdiscard = 0;
764
				queue_flag_clear(QUEUE_FLAG_DISCARD, rq);
765
				queue_flag_clear(QUEUE_FLAG_SECDISCARD, rq);
766
767
768
			}
			__blk_end_request_all(req, error);
			break;
769
		case BLKIF_OP_FLUSH_DISKCACHE:
770
771
		case BLKIF_OP_WRITE_BARRIER:
			if (unlikely(bret->status == BLKIF_RSP_EOPNOTSUPP)) {
772
773
774
				printk(KERN_WARNING "blkfront: %s: write %s op failed\n",
				       info->flush_op == BLKIF_OP_WRITE_BARRIER ?
				       "barrier" :  "flush disk cache",
775
				       info->gd->disk_name);
776
				error = -EOPNOTSUPP;
777
778
			}
			if (unlikely(bret->status == BLKIF_RSP_ERROR &&
779
				     info->shadow[id].req.u.rw.nr_segments == 0)) {
780
781
782
				printk(KERN_WARNING "blkfront: %s: empty write %s op failed\n",
				       info->flush_op == BLKIF_OP_WRITE_BARRIER ?
				       "barrier" :  "flush disk cache",
783
784
785
786
787
788
				       info->gd->disk_name);
				error = -EOPNOTSUPP;
			}
			if (unlikely(error)) {
				if (error == -EOPNOTSUPP)
					error = 0;
789
				info->feature_flush = 0;
790
				info->flush_op = 0;
791
				xlvbd_flush(info);
792
793
794
795
796
797
798
799
			}
			/* fall through */
		case BLKIF_OP_READ:
		case BLKIF_OP_WRITE:
			if (unlikely(bret->status != BLKIF_RSP_OKAY))
				dev_dbg(&info->xbdev->dev, "Bad return from blkdev data "
					"request: %x\n", bret->status);

800
			__blk_end_request_all(req, error);
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
			break;
		default:
			BUG();
		}
	}

	info->ring.rsp_cons = i;

	if (i != info->ring.req_prod_pvt) {
		int more_to_do;
		RING_FINAL_CHECK_FOR_RESPONSES(&info->ring, more_to_do);
		if (more_to_do)
			goto again;
	} else
		info->ring.sring->rsp_event = i + 1;

	kick_pending_request_queues(info);

	spin_unlock_irqrestore(&blkif_io_lock, flags);

	return IRQ_HANDLED;
}


static int setup_blkring(struct xenbus_device *dev,
			 struct blkfront_info *info)
{
	struct blkif_sring *sring;
	int err;

	info->ring_ref = GRANT_INVALID_REF;

833
	sring = (struct blkif_sring *)__get_free_page(GFP_NOIO | __GFP_HIGH);
834
835
836
837
838
839
	if (!sring) {
		xenbus_dev_fatal(dev, -ENOMEM, "allocating shared ring");
		return -ENOMEM;
	}
	SHARED_RING_INIT(sring);
	FRONT_RING_INIT(&info->ring, sring, PAGE_SIZE);
840
841

	sg_init_table(info->sg, BLKIF_MAX_SEGMENTS_PER_REQUEST);
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872

	err = xenbus_grant_ring(dev, virt_to_mfn(info->ring.sring));
	if (err < 0) {
		free_page((unsigned long)sring);
		info->ring.sring = NULL;
		goto fail;
	}
	info->ring_ref = err;

	err = xenbus_alloc_evtchn(dev, &info->evtchn);
	if (err)
		goto fail;

	err = bind_evtchn_to_irqhandler(info->evtchn,
					blkif_interrupt,
					IRQF_SAMPLE_RANDOM, "blkif", info);
	if (err <= 0) {
		xenbus_dev_fatal(dev, err,
				 "bind_evtchn_to_irqhandler failed");
		goto fail;
	}
	info->irq = err;

	return 0;
fail:
	blkif_free(info, 0);
	return err;
}


/* Common code used when first setting up, and when resuming. */
873
static int talk_to_blkback(struct xenbus_device *dev,
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
			   struct blkfront_info *info)
{
	const char *message = NULL;
	struct xenbus_transaction xbt;
	int err;

	/* Create shared ring, alloc event channel. */
	err = setup_blkring(dev, info);
	if (err)
		goto out;

again:
	err = xenbus_transaction_start(&xbt);
	if (err) {
		xenbus_dev_fatal(dev, err, "starting transaction");
		goto destroy_blkring;
	}

	err = xenbus_printf(xbt, dev->nodename,
			    "ring-ref", "%u", info->ring_ref);
	if (err) {
		message = "writing ring-ref";
		goto abort_transaction;
	}
	err = xenbus_printf(xbt, dev->nodename,
			    "event-channel", "%u", info->evtchn);
	if (err) {
		message = "writing event-channel";
		goto abort_transaction;
	}
904
905
906
907
908
909
	err = xenbus_printf(xbt, dev->nodename, "protocol", "%s",
			    XEN_IO_PROTO_ABI_NATIVE);
	if (err) {
		message = "writing protocol";
		goto abort_transaction;
	}
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948

	err = xenbus_transaction_end(xbt, 0);
	if (err) {
		if (err == -EAGAIN)
			goto again;
		xenbus_dev_fatal(dev, err, "completing transaction");
		goto destroy_blkring;
	}

	xenbus_switch_state(dev, XenbusStateInitialised);

	return 0;

 abort_transaction:
	xenbus_transaction_end(xbt, 1);
	if (message)
		xenbus_dev_fatal(dev, err, "%s", message);
 destroy_blkring:
	blkif_free(info, 0);
 out:
	return err;
}

/**
 * Entry point to this code when a new device is created.  Allocate the basic
 * structures and the ring buffer for communication with the backend, and
 * inform the backend of the appropriate details for those.  Switch to
 * Initialised state.
 */
static int blkfront_probe(struct xenbus_device *dev,
			  const struct xenbus_device_id *id)
{
	int err, vdevice, i;
	struct blkfront_info *info;

	/* FIXME: Use dynamic device id if this is not set. */
	err = xenbus_scanf(XBT_NIL, dev->nodename,
			   "virtual-device", "%i", &vdevice);
	if (err != 1) {
949
950
951
952
953
954
955
		/* go looking in the extended area instead */
		err = xenbus_scanf(XBT_NIL, dev->nodename, "virtual-device-ext",
				   "%i", &vdevice);
		if (err != 1) {
			xenbus_dev_fatal(dev, err, "reading virtual-device");
			return err;
		}
956
957
	}

958
959
960
961
	if (xen_hvm_domain()) {
		char *type;
		int len;
		/* no unplug has been done: do not hook devices != xen vbds */
962
		if (xen_platform_pci_unplug & XEN_UNPLUG_UNNECESSARY) {
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
			int major;

			if (!VDEV_IS_EXTENDED(vdevice))
				major = BLKIF_MAJOR(vdevice);
			else
				major = XENVBD_MAJOR;

			if (major != XENVBD_MAJOR) {
				printk(KERN_INFO
						"%s: HVM does not support vbd %d as xen block device\n",
						__FUNCTION__, vdevice);
				return -ENODEV;
			}
		}
		/* do not create a PV cdrom device if we are an HVM guest */
		type = xenbus_read(XBT_NIL, dev->nodename, "device-type", &len);
		if (IS_ERR(type))
			return -ENODEV;
		if (strncmp(type, "cdrom", 5) == 0) {
			kfree(type);
983
984
			return -ENODEV;
		}
985
		kfree(type);
986
	}
987
988
989
990
991
992
	info = kzalloc(sizeof(*info), GFP_KERNEL);
	if (!info) {
		xenbus_dev_fatal(dev, -ENOMEM, "allocating info structure");
		return -ENOMEM;
	}

993
	mutex_init(&info->mutex);
994
995
996
997
998
999
	info->xbdev = dev;
	info->vdevice = vdevice;
	info->connected = BLKIF_STATE_DISCONNECTED;
	INIT_WORK(&info->work, blkif_restart_queue);

	for (i = 0; i < BLK_RING_SIZE; i++)
1000
		info->shadow[i].req.u.rw.id = i+1;