virtio_ring.c 19.5 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
/* Virtio ring implementation.
 *
 *  Copyright 2007 Rusty Russell IBM Corporation
 *
 *  This program is free software; you can redistribute it and/or modify
 *  it under the terms of the GNU General Public License as published by
 *  the Free Software Foundation; either version 2 of the License, or
 *  (at your option) any later version.
 *
 *  This program is distributed in the hope that it will be useful,
 *  but WITHOUT ANY WARRANTY; without even the implied warranty of
 *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 *  GNU General Public License for more details.
 *
 *  You should have received a copy of the GNU General Public License
 *  along with this program; if not, write to the Free Software
 *  Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
 */
#include <linux/virtio.h>
#include <linux/virtio_ring.h>
21
#include <linux/virtio_config.h>
22
#include <linux/device.h>
23
#include <linux/slab.h>
24
#include <linux/module.h>
25
#include <linux/hrtimer.h>
26

27 28 29 30 31
/* virtio guest is communicating with a virtual "device" that actually runs on
 * a host processor.  Memory barriers are used to control SMP effects. */
#ifdef CONFIG_SMP
/* Where possible, use SMP barriers which are more lightweight than mandatory
 * barriers, because mandatory barriers control MMIO effects on accesses
32 33 34 35 36 37
 * through relaxed memory I/O windows (which virtio-pci does not use). */
#define virtio_mb(vq) \
	do { if ((vq)->weak_barriers) smp_mb(); else mb(); } while(0)
#define virtio_rmb(vq) \
	do { if ((vq)->weak_barriers) smp_rmb(); else rmb(); } while(0)
#define virtio_wmb(vq) \
38
	do { if ((vq)->weak_barriers) smp_wmb(); else wmb(); } while(0)
39 40 41 42
#else
/* We must force memory ordering even if guest is UP since host could be
 * running on another CPU, but SMP barriers are defined to barrier() in that
 * configuration. So fall back to mandatory barriers instead. */
43 44 45
#define virtio_mb(vq) mb()
#define virtio_rmb(vq) rmb()
#define virtio_wmb(vq) wmb()
46 47
#endif

48 49
#ifdef DEBUG
/* For development, we want to crash whenever the ring is screwed. */
50 51 52 53 54 55
#define BAD_RING(_vq, fmt, args...)				\
	do {							\
		dev_err(&(_vq)->vq.vdev->dev,			\
			"%s:"fmt, (_vq)->vq.name, ##args);	\
		BUG();						\
	} while (0)
56 57 58 59
/* Caller is supposed to guarantee no reentry. */
#define START_USE(_vq)						\
	do {							\
		if ((_vq)->in_use)				\
60 61
			panic("%s:in_use = %i\n",		\
			      (_vq)->vq.name, (_vq)->in_use);	\
62
		(_vq)->in_use = __LINE__;			\
63
	} while (0)
64
#define END_USE(_vq) \
65
	do { BUG_ON(!(_vq)->in_use); (_vq)->in_use = 0; } while(0)
66
#else
67 68 69 70 71 72
#define BAD_RING(_vq, fmt, args...)				\
	do {							\
		dev_err(&_vq->vq.vdev->dev,			\
			"%s:"fmt, (_vq)->vq.name, ##args);	\
		(_vq)->broken = true;				\
	} while (0)
73 74 75 76 77 78 79 80 81 82 83
#define START_USE(vq)
#define END_USE(vq)
#endif

struct vring_virtqueue
{
	struct virtqueue vq;

	/* Actual memory layout for this queue */
	struct vring vring;

84 85 86
	/* Can we use weak barriers? */
	bool weak_barriers;

87 88 89
	/* Other side has made a mess, don't try any more. */
	bool broken;

90 91 92
	/* Host supports indirect buffers */
	bool indirect;

93 94 95
	/* Host publishes avail event idx */
	bool event;

96 97 98 99 100 101 102 103
	/* Number of free buffers */
	unsigned int num_free;
	/* Head of free buffer list. */
	unsigned int free_head;
	/* Number we've added since last sync. */
	unsigned int num_added;

	/* Last used index we've seen. */
A
Anthony Liguori 已提交
104
	u16 last_used_idx;
105 106 107 108

	/* How to notify other side. FIXME: commonalize hcalls! */
	void (*notify)(struct virtqueue *vq);

109 110 111
	/* Index of the queue */
	int queue_index;

112 113 114
#ifdef DEBUG
	/* They're supposed to lock for us. */
	unsigned int in_use;
115 116 117 118

	/* Figure out if their kicks are too delayed. */
	bool last_add_time_valid;
	ktime_t last_add_time;
119 120 121 122 123 124 125 126
#endif

	/* Tokens for callbacks. */
	void *data[];
};

#define to_vvq(_vq) container_of(_vq, struct vring_virtqueue, vq)

127 128 129 130
/* Set up an indirect table of descriptors and add it to the queue. */
static int vring_add_indirect(struct vring_virtqueue *vq,
			      struct scatterlist sg[],
			      unsigned int out,
M
Michael S. Tsirkin 已提交
131 132
			      unsigned int in,
			      gfp_t gfp)
133 134 135 136 137
{
	struct vring_desc *desc;
	unsigned head;
	int i;

138 139 140 141 142 143 144
	/*
	 * We require lowmem mappings for the descriptors because
	 * otherwise virt_to_phys will give us bogus addresses in the
	 * virtqueue.
	 */
	gfp &= ~(__GFP_HIGHMEM | __GFP_HIGH);

M
Michael S. Tsirkin 已提交
145
	desc = kmalloc((out + in) * sizeof(struct vring_desc), gfp);
146
	if (!desc)
147
		return -ENOMEM;
148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183

	/* Transfer entries from the sg list into the indirect page */
	for (i = 0; i < out; i++) {
		desc[i].flags = VRING_DESC_F_NEXT;
		desc[i].addr = sg_phys(sg);
		desc[i].len = sg->length;
		desc[i].next = i+1;
		sg++;
	}
	for (; i < (out + in); i++) {
		desc[i].flags = VRING_DESC_F_NEXT|VRING_DESC_F_WRITE;
		desc[i].addr = sg_phys(sg);
		desc[i].len = sg->length;
		desc[i].next = i+1;
		sg++;
	}

	/* Last one doesn't continue. */
	desc[i-1].flags &= ~VRING_DESC_F_NEXT;
	desc[i-1].next = 0;

	/* We're about to use a buffer */
	vq->num_free--;

	/* Use a single buffer which doesn't continue */
	head = vq->free_head;
	vq->vring.desc[head].flags = VRING_DESC_F_INDIRECT;
	vq->vring.desc[head].addr = virt_to_phys(desc);
	vq->vring.desc[head].len = i * sizeof(struct vring_desc);

	/* Update free pointer */
	vq->free_head = vq->vring.desc[head].next;

	return head;
}

184 185 186 187 188 189 190
int virtqueue_get_queue_index(struct virtqueue *_vq)
{
	struct vring_virtqueue *vq = to_vvq(_vq);
	return vq->queue_index;
}
EXPORT_SYMBOL_GPL(virtqueue_get_queue_index);

191
/**
192
 * virtqueue_add_buf - expose buffer to other end
193 194 195 196 197 198 199 200 201 202 203 204 205 206 207
 * @vq: the struct virtqueue we're talking about.
 * @sg: the description of the buffer(s).
 * @out_num: the number of sg readable by other side
 * @in_num: the number of sg which are writable (after readable ones)
 * @data: the token identifying the buffer.
 * @gfp: how to do memory allocations (if necessary).
 *
 * Caller must ensure we don't call this with other virtqueue operations
 * at the same time (except where noted).
 *
 * Returns remaining capacity of queue or a negative error
 * (ie. ENOSPC).  Note that it only really makes sense to treat all
 * positive return values as "available": indirect buffers mean that
 * we can put an entire sg[] array inside a single queue entry.
 */
208 209 210 211 212 213
int virtqueue_add_buf(struct virtqueue *_vq,
		      struct scatterlist sg[],
		      unsigned int out,
		      unsigned int in,
		      void *data,
		      gfp_t gfp)
214 215
{
	struct vring_virtqueue *vq = to_vvq(_vq);
M
Michael S. Tsirkin 已提交
216 217
	unsigned int i, avail, uninitialized_var(prev);
	int head;
218

219 220
	START_USE(vq);

221
	BUG_ON(data == NULL);
222

223 224 225 226 227 228 229 230 231 232 233 234 235
#ifdef DEBUG
	{
		ktime_t now = ktime_get();

		/* No kick or get, with .1 second between?  Warn. */
		if (vq->last_add_time_valid)
			WARN_ON(ktime_to_ms(ktime_sub(now, vq->last_add_time))
					    > 100);
		vq->last_add_time = now;
		vq->last_add_time_valid = true;
	}
#endif

236 237 238
	/* If the host supports indirect descriptor tables, and we have multiple
	 * buffers, then go indirect. FIXME: tune this threshold */
	if (vq->indirect && (out + in) > 1 && vq->num_free) {
M
Michael S. Tsirkin 已提交
239
		head = vring_add_indirect(vq, sg, out, in, gfp);
M
Michael S. Tsirkin 已提交
240
		if (likely(head >= 0))
241 242 243
			goto add_head;
	}

244 245 246 247 248 249
	BUG_ON(out + in > vq->vring.num);
	BUG_ON(out + in == 0);

	if (vq->num_free < out + in) {
		pr_debug("Can't add buf len %i - avail = %i\n",
			 out + in, vq->num_free);
250 251 252 253 254
		/* FIXME: for historical reasons, we force a notify here if
		 * there are outgoing parts to the buffer.  Presumably the
		 * host should service the ring ASAP. */
		if (out)
			vq->notify(&vq->vq);
255 256 257 258 259 260 261 262 263 264
		END_USE(vq);
		return -ENOSPC;
	}

	/* We're about to use some buffers from the free list. */
	vq->num_free -= out + in;

	head = vq->free_head;
	for (i = vq->free_head; out; i = vq->vring.desc[i].next, out--) {
		vq->vring.desc[i].flags = VRING_DESC_F_NEXT;
265
		vq->vring.desc[i].addr = sg_phys(sg);
266 267 268 269 270 271
		vq->vring.desc[i].len = sg->length;
		prev = i;
		sg++;
	}
	for (; in; i = vq->vring.desc[i].next, in--) {
		vq->vring.desc[i].flags = VRING_DESC_F_NEXT|VRING_DESC_F_WRITE;
272
		vq->vring.desc[i].addr = sg_phys(sg);
273 274 275 276 277 278 279 280 281 282
		vq->vring.desc[i].len = sg->length;
		prev = i;
		sg++;
	}
	/* Last one doesn't continue. */
	vq->vring.desc[prev].flags &= ~VRING_DESC_F_NEXT;

	/* Update free pointer */
	vq->free_head = i;

283
add_head:
284 285 286 287
	/* Set token. */
	vq->data[head] = data;

	/* Put entry in available array (but don't update avail->idx until they
R
Rusty Russell 已提交
288
	 * do sync). */
289
	avail = (vq->vring.avail->idx & (vq->vring.num-1));
290 291
	vq->vring.avail->ring[avail] = head;

292 293 294 295 296 297 298 299 300 301 302
	/* Descriptors and available array need to be set before we expose the
	 * new available array entries. */
	virtio_wmb(vq);
	vq->vring.avail->idx++;
	vq->num_added++;

	/* This is very unlikely, but theoretically possible.  Kick
	 * just in case. */
	if (unlikely(vq->num_added == (1 << 16) - 1))
		virtqueue_kick(_vq);

303 304
	pr_debug("Added buffer head %i to %p\n", head, vq);
	END_USE(vq);
305 306

	return vq->num_free;
307
}
308
EXPORT_SYMBOL_GPL(virtqueue_add_buf);
309

310
/**
311
 * virtqueue_kick_prepare - first half of split virtqueue_kick call.
312 313
 * @vq: the struct virtqueue
 *
314 315 316
 * Instead of virtqueue_kick(), you can do:
 *	if (virtqueue_kick_prepare(vq))
 *		virtqueue_notify(vq);
317
 *
318 319
 * This is sometimes useful because the virtqueue_kick_prepare() needs
 * to be serialized, but the actual virtqueue_notify() call does not.
320
 */
321
bool virtqueue_kick_prepare(struct virtqueue *_vq)
322 323
{
	struct vring_virtqueue *vq = to_vvq(_vq);
324
	u16 new, old;
325 326
	bool needs_kick;

327
	START_USE(vq);
328 329 330
	/* We need to expose available array entries before checking avail
	 * event. */
	virtio_mb(vq);
331

332 333
	old = vq->vring.avail->idx - vq->num_added;
	new = vq->vring.avail->idx;
334 335
	vq->num_added = 0;

336 337 338 339 340 341 342 343
#ifdef DEBUG
	if (vq->last_add_time_valid) {
		WARN_ON(ktime_to_ms(ktime_sub(ktime_get(),
					      vq->last_add_time)) > 100);
	}
	vq->last_add_time_valid = false;
#endif

344 345 346 347 348 349
	if (vq->event) {
		needs_kick = vring_need_event(vring_avail_event(&vq->vring),
					      new, old);
	} else {
		needs_kick = !(vq->vring.used->flags & VRING_USED_F_NO_NOTIFY);
	}
350
	END_USE(vq);
351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383
	return needs_kick;
}
EXPORT_SYMBOL_GPL(virtqueue_kick_prepare);

/**
 * virtqueue_notify - second half of split virtqueue_kick call.
 * @vq: the struct virtqueue
 *
 * This does not need to be serialized.
 */
void virtqueue_notify(struct virtqueue *_vq)
{
	struct vring_virtqueue *vq = to_vvq(_vq);

	/* Prod other side to tell it about changes. */
	vq->notify(_vq);
}
EXPORT_SYMBOL_GPL(virtqueue_notify);

/**
 * virtqueue_kick - update after add_buf
 * @vq: the struct virtqueue
 *
 * After one or more virtqueue_add_buf calls, invoke this to kick
 * the other side.
 *
 * Caller must ensure we don't call this with other virtqueue
 * operations at the same time (except where noted).
 */
void virtqueue_kick(struct virtqueue *vq)
{
	if (virtqueue_kick_prepare(vq))
		virtqueue_notify(vq);
384
}
385
EXPORT_SYMBOL_GPL(virtqueue_kick);
386 387 388 389 390 391 392 393 394 395

static void detach_buf(struct vring_virtqueue *vq, unsigned int head)
{
	unsigned int i;

	/* Clear data ptr. */
	vq->data[head] = NULL;

	/* Put back on free list: find end */
	i = head;
396 397 398 399 400

	/* Free the indirect table */
	if (vq->vring.desc[i].flags & VRING_DESC_F_INDIRECT)
		kfree(phys_to_virt(vq->vring.desc[i].addr));

401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416
	while (vq->vring.desc[i].flags & VRING_DESC_F_NEXT) {
		i = vq->vring.desc[i].next;
		vq->num_free++;
	}

	vq->vring.desc[i].next = vq->free_head;
	vq->free_head = head;
	/* Plus final descriptor */
	vq->num_free++;
}

static inline bool more_used(const struct vring_virtqueue *vq)
{
	return vq->last_used_idx != vq->vring.used->idx;
}

417 418 419 420 421 422 423 424 425 426 427 428 429 430
/**
 * virtqueue_get_buf - get the next used buffer
 * @vq: the struct virtqueue we're talking about.
 * @len: the length written into the buffer
 *
 * If the driver wrote data into the buffer, @len will be set to the
 * amount written.  This means you don't need to clear the buffer
 * beforehand to ensure there's no data leakage in the case of short
 * writes.
 *
 * Caller must ensure we don't call this with other virtqueue
 * operations at the same time (except where noted).
 *
 * Returns NULL if there are no used buffers, or the "data" token
431
 * handed to virtqueue_add_buf().
432
 */
433
void *virtqueue_get_buf(struct virtqueue *_vq, unsigned int *len)
434 435 436 437
{
	struct vring_virtqueue *vq = to_vvq(_vq);
	void *ret;
	unsigned int i;
R
Rusty Russell 已提交
438
	u16 last_used;
439 440 441

	START_USE(vq);

442 443 444 445 446
	if (unlikely(vq->broken)) {
		END_USE(vq);
		return NULL;
	}

447 448 449 450 451 452
	if (!more_used(vq)) {
		pr_debug("No more buffers in queue\n");
		END_USE(vq);
		return NULL;
	}

453
	/* Only get used array entries after they have been exposed by host. */
454
	virtio_rmb(vq);
455

R
Rusty Russell 已提交
456 457 458
	last_used = (vq->last_used_idx & (vq->vring.num - 1));
	i = vq->vring.used->ring[last_used].id;
	*len = vq->vring.used->ring[last_used].len;
459 460 461 462 463 464 465 466 467 468 469 470 471 472

	if (unlikely(i >= vq->vring.num)) {
		BAD_RING(vq, "id %u out of range\n", i);
		return NULL;
	}
	if (unlikely(!vq->data[i])) {
		BAD_RING(vq, "id %u is not a head!\n", i);
		return NULL;
	}

	/* detach_buf clears data, so grab it now. */
	ret = vq->data[i];
	detach_buf(vq, i);
	vq->last_used_idx++;
473 474 475 476 477
	/* If we expect an interrupt for the next entry, tell host
	 * by writing event index and flush out the write before
	 * the read in the next get_buf call. */
	if (!(vq->vring.avail->flags & VRING_AVAIL_F_NO_INTERRUPT)) {
		vring_used_event(&vq->vring) = vq->last_used_idx;
478
		virtio_mb(vq);
479 480
	}

481 482 483 484
#ifdef DEBUG
	vq->last_add_time_valid = false;
#endif

485 486 487
	END_USE(vq);
	return ret;
}
488
EXPORT_SYMBOL_GPL(virtqueue_get_buf);
489

490 491 492 493 494 495 496 497 498
/**
 * virtqueue_disable_cb - disable callbacks
 * @vq: the struct virtqueue we're talking about.
 *
 * Note that this is not necessarily synchronous, hence unreliable and only
 * useful as an optimization.
 *
 * Unlike other operations, this need not be serialized.
 */
499
void virtqueue_disable_cb(struct virtqueue *_vq)
500 501 502 503 504
{
	struct vring_virtqueue *vq = to_vvq(_vq);

	vq->vring.avail->flags |= VRING_AVAIL_F_NO_INTERRUPT;
}
505
EXPORT_SYMBOL_GPL(virtqueue_disable_cb);
506

507 508 509 510 511 512 513 514 515 516 517
/**
 * virtqueue_enable_cb - restart callbacks after disable_cb.
 * @vq: the struct virtqueue we're talking about.
 *
 * This re-enables callbacks; it returns "false" if there are pending
 * buffers in the queue, to detect a possible race between the driver
 * checking for more work, and enabling callbacks.
 *
 * Caller must ensure we don't call this with other virtqueue
 * operations at the same time (except where noted).
 */
518
bool virtqueue_enable_cb(struct virtqueue *_vq)
519 520 521 522 523 524 525
{
	struct vring_virtqueue *vq = to_vvq(_vq);

	START_USE(vq);

	/* We optimistically turn back on interrupts, then check if there was
	 * more to do. */
526 527 528
	/* Depending on the VIRTIO_RING_F_EVENT_IDX feature, we need to
	 * either clear the flags bit or point the event index at the next
	 * entry. Always do both to keep code simple. */
529
	vq->vring.avail->flags &= ~VRING_AVAIL_F_NO_INTERRUPT;
530
	vring_used_event(&vq->vring) = vq->last_used_idx;
531
	virtio_mb(vq);
532 533 534 535 536 537 538 539
	if (unlikely(more_used(vq))) {
		END_USE(vq);
		return false;
	}

	END_USE(vq);
	return true;
}
540
EXPORT_SYMBOL_GPL(virtqueue_enable_cb);
541

542 543 544 545 546 547 548 549 550 551 552 553 554
/**
 * virtqueue_enable_cb_delayed - restart callbacks after disable_cb.
 * @vq: the struct virtqueue we're talking about.
 *
 * This re-enables callbacks but hints to the other side to delay
 * interrupts until most of the available buffers have been processed;
 * it returns "false" if there are many pending buffers in the queue,
 * to detect a possible race between the driver checking for more work,
 * and enabling callbacks.
 *
 * Caller must ensure we don't call this with other virtqueue
 * operations at the same time (except where noted).
 */
555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570
bool virtqueue_enable_cb_delayed(struct virtqueue *_vq)
{
	struct vring_virtqueue *vq = to_vvq(_vq);
	u16 bufs;

	START_USE(vq);

	/* We optimistically turn back on interrupts, then check if there was
	 * more to do. */
	/* Depending on the VIRTIO_RING_F_USED_EVENT_IDX feature, we need to
	 * either clear the flags bit or point the event index at the next
	 * entry. Always do both to keep code simple. */
	vq->vring.avail->flags &= ~VRING_AVAIL_F_NO_INTERRUPT;
	/* TODO: tune this threshold */
	bufs = (u16)(vq->vring.avail->idx - vq->last_used_idx) * 3 / 4;
	vring_used_event(&vq->vring) = vq->last_used_idx + bufs;
571
	virtio_mb(vq);
572 573 574 575 576 577 578 579 580 581
	if (unlikely((u16)(vq->vring.used->idx - vq->last_used_idx) > bufs)) {
		END_USE(vq);
		return false;
	}

	END_USE(vq);
	return true;
}
EXPORT_SYMBOL_GPL(virtqueue_enable_cb_delayed);

582 583 584 585
/**
 * virtqueue_detach_unused_buf - detach first unused buffer
 * @vq: the struct virtqueue we're talking about.
 *
586
 * Returns NULL or the "data" token handed to virtqueue_add_buf().
587 588 589
 * This is not valid on an active queue; it is useful only for device
 * shutdown.
 */
590
void *virtqueue_detach_unused_buf(struct virtqueue *_vq)
591 592 593 594 595 596 597 598 599 600 601 602 603
{
	struct vring_virtqueue *vq = to_vvq(_vq);
	unsigned int i;
	void *buf;

	START_USE(vq);

	for (i = 0; i < vq->vring.num; i++) {
		if (!vq->data[i])
			continue;
		/* detach_buf clears data, so grab it now. */
		buf = vq->data[i];
		detach_buf(vq, i);
604
		vq->vring.avail->idx--;
605 606 607 608 609 610 611 612 613
		END_USE(vq);
		return buf;
	}
	/* That should have freed everything. */
	BUG_ON(vq->num_free != vq->vring.num);

	END_USE(vq);
	return NULL;
}
614
EXPORT_SYMBOL_GPL(virtqueue_detach_unused_buf);
615

616 617 618 619 620 621 622 623 624 625 626 627 628
irqreturn_t vring_interrupt(int irq, void *_vq)
{
	struct vring_virtqueue *vq = to_vvq(_vq);

	if (!more_used(vq)) {
		pr_debug("virtqueue interrupt with no work for %p\n", vq);
		return IRQ_NONE;
	}

	if (unlikely(vq->broken))
		return IRQ_HANDLED;

	pr_debug("virtqueue callback for %p (%p)\n", vq, vq->vq.callback);
629 630
	if (vq->vq.callback)
		vq->vq.callback(&vq->vq);
631 632 633

	return IRQ_HANDLED;
}
634
EXPORT_SYMBOL_GPL(vring_interrupt);
635

636 637
struct virtqueue *vring_new_virtqueue(unsigned int index,
				      unsigned int num,
638
				      unsigned int vring_align,
639
				      struct virtio_device *vdev,
640
				      bool weak_barriers,
641 642
				      void *pages,
				      void (*notify)(struct virtqueue *),
643 644
				      void (*callback)(struct virtqueue *),
				      const char *name)
645 646 647 648
{
	struct vring_virtqueue *vq;
	unsigned int i;

649 650 651 652 653 654
	/* We assume num is a power of 2. */
	if (num & (num - 1)) {
		dev_warn(&vdev->dev, "Bad virtqueue length %u\n", num);
		return NULL;
	}

655 656 657 658
	vq = kmalloc(sizeof(*vq) + sizeof(void *)*num, GFP_KERNEL);
	if (!vq)
		return NULL;

659
	vring_init(&vq->vring, num, pages, vring_align);
660 661
	vq->vq.callback = callback;
	vq->vq.vdev = vdev;
662
	vq->vq.name = name;
663
	vq->notify = notify;
664
	vq->weak_barriers = weak_barriers;
665 666 667
	vq->broken = false;
	vq->last_used_idx = 0;
	vq->num_added = 0;
668
	vq->queue_index = index;
669
	list_add_tail(&vq->vq.list, &vdev->vqs);
670 671
#ifdef DEBUG
	vq->in_use = false;
672
	vq->last_add_time_valid = false;
673 674
#endif

675
	vq->indirect = virtio_has_feature(vdev, VIRTIO_RING_F_INDIRECT_DESC);
676
	vq->event = virtio_has_feature(vdev, VIRTIO_RING_F_EVENT_IDX);
677

678 679 680 681 682 683 684
	/* No callback?  Tell other side not to bother us. */
	if (!callback)
		vq->vring.avail->flags |= VRING_AVAIL_F_NO_INTERRUPT;

	/* Put everything in free lists. */
	vq->num_free = num;
	vq->free_head = 0;
685
	for (i = 0; i < num-1; i++) {
686
		vq->vring.desc[i].next = i+1;
687 688 689
		vq->data[i] = NULL;
	}
	vq->data[i] = NULL;
690 691 692

	return &vq->vq;
}
693
EXPORT_SYMBOL_GPL(vring_new_virtqueue);
694 695 696

void vring_del_virtqueue(struct virtqueue *vq)
{
697
	list_del(&vq->list);
698 699
	kfree(to_vvq(vq));
}
700
EXPORT_SYMBOL_GPL(vring_del_virtqueue);
701

702 703 704 705 706 707 708
/* Manipulates transport-specific feature bits. */
void vring_transport_features(struct virtio_device *vdev)
{
	unsigned int i;

	for (i = VIRTIO_TRANSPORT_F_START; i < VIRTIO_TRANSPORT_F_END; i++) {
		switch (i) {
709 710
		case VIRTIO_RING_F_INDIRECT_DESC:
			break;
711 712
		case VIRTIO_RING_F_EVENT_IDX:
			break;
713 714 715 716 717 718 719 720
		default:
			/* We don't understand this bit. */
			clear_bit(i, vdev->features);
		}
	}
}
EXPORT_SYMBOL_GPL(vring_transport_features);

721 722 723 724 725 726 727
/**
 * virtqueue_get_vring_size - return the size of the virtqueue's vring
 * @vq: the struct virtqueue containing the vring of interest.
 *
 * Returns the size of the vring.  This is mainly used for boasting to
 * userspace.  Unlike other operations, this need not be serialized.
 */
R
Rick Jones 已提交
728 729 730 731 732 733 734 735 736
unsigned int virtqueue_get_vring_size(struct virtqueue *_vq)
{

	struct vring_virtqueue *vq = to_vvq(_vq);

	return vq->vring.num;
}
EXPORT_SYMBOL_GPL(virtqueue_get_vring_size);

737
MODULE_LICENSE("GPL");