virtio_ring.c 11.9 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 23
#include <linux/device.h>

24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41
/* 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
 * through relaxed memory I/O windows (which virtio does not use). */
#define virtio_mb() smp_mb()
#define virtio_rmb() smp_rmb()
#define virtio_wmb() smp_wmb()
#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. */
#define virtio_mb() mb()
#define virtio_rmb() rmb()
#define virtio_wmb() wmb()
#endif

42 43
#ifdef DEBUG
/* For development, we want to crash whenever the ring is screwed. */
44 45 46 47 48 49
#define BAD_RING(_vq, fmt, args...)				\
	do {							\
		dev_err(&(_vq)->vq.vdev->dev,			\
			"%s:"fmt, (_vq)->vq.name, ##args);	\
		BUG();						\
	} while (0)
50 51 52 53
/* Caller is supposed to guarantee no reentry. */
#define START_USE(_vq)						\
	do {							\
		if ((_vq)->in_use)				\
54 55
			panic("%s:in_use = %i\n",		\
			      (_vq)->vq.name, (_vq)->in_use);	\
56
		(_vq)->in_use = __LINE__;			\
57
	} while (0)
58
#define END_USE(_vq) \
59
	do { BUG_ON(!(_vq)->in_use); (_vq)->in_use = 0; } while(0)
60
#else
61 62 63 64 65 66
#define BAD_RING(_vq, fmt, args...)				\
	do {							\
		dev_err(&_vq->vq.vdev->dev,			\
			"%s:"fmt, (_vq)->vq.name, ##args);	\
		(_vq)->broken = true;				\
	} while (0)
67 68 69 70 71 72 73 74 75 76 77 78 79 80
#define START_USE(vq)
#define END_USE(vq)
#endif

struct vring_virtqueue
{
	struct virtqueue vq;

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

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

81 82 83
	/* Host supports indirect buffers */
	bool indirect;

84 85 86 87 88 89 90 91
	/* 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
92
	u16 last_used_idx;
93 94 95 96 97 98 99 100 101 102 103 104 105 106 107

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

#ifdef DEBUG
	/* They're supposed to lock for us. */
	unsigned int in_use;
#endif

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

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

108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156
/* 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,
			      unsigned int in)
{
	struct vring_desc *desc;
	unsigned head;
	int i;

	desc = kmalloc((out + in) * sizeof(struct vring_desc), GFP_ATOMIC);
	if (!desc)
		return vq->vring.num;

	/* 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;
}

157 158 159 160 161 162 163 164 165
static int vring_add_buf(struct virtqueue *_vq,
			 struct scatterlist sg[],
			 unsigned int out,
			 unsigned int in,
			 void *data)
{
	struct vring_virtqueue *vq = to_vvq(_vq);
	unsigned int i, avail, head, uninitialized_var(prev);

166 167
	START_USE(vq);

168
	BUG_ON(data == NULL);
169 170 171 172 173 174 175 176 177

	/* 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) {
		head = vring_add_indirect(vq, sg, out, in);
		if (head != vq->vring.num)
			goto add_head;
	}

178 179 180 181 182 183
	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);
184 185 186 187 188
		/* 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);
189 190 191 192 193 194 195 196 197 198
		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;
199
		vq->vring.desc[i].addr = sg_phys(sg);
200 201 202 203 204 205
		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;
206
		vq->vring.desc[i].addr = sg_phys(sg);
207 208 209 210 211 212 213 214 215 216
		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;

217
add_head:
218 219 220 221 222 223 224 225 226 227
	/* Set token. */
	vq->data[head] = data;

	/* Put entry in available array (but don't update avail->idx until they
	 * do sync).  FIXME: avoid modulus here? */
	avail = (vq->vring.avail->idx + vq->num_added++) % vq->vring.num;
	vq->vring.avail->ring[avail] = head;

	pr_debug("Added buffer head %i to %p\n", head, vq);
	END_USE(vq);
228 229 230 231 232

	/* If we're indirect, we can fit many (assuming not OOM). */
	if (vq->indirect)
		return vq->num_free ? vq->vring.num : 0;
	return vq->num_free;
233 234 235 236 237 238 239 240
}

static void vring_kick(struct virtqueue *_vq)
{
	struct vring_virtqueue *vq = to_vvq(_vq);
	START_USE(vq);
	/* Descriptors and available array need to be set before we expose the
	 * new available array entries. */
241
	virtio_wmb();
242 243 244 245 246

	vq->vring.avail->idx += vq->num_added;
	vq->num_added = 0;

	/* Need to update avail index before checking if we should notify */
247
	virtio_mb();
248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264

	if (!(vq->vring.used->flags & VRING_USED_F_NO_NOTIFY))
		/* Prod other side to tell it about changes. */
		vq->notify(&vq->vq);

	END_USE(vq);
}

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;
265 266 267 268 269

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

270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293
	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;
}

static void *vring_get_buf(struct virtqueue *_vq, unsigned int *len)
{
	struct vring_virtqueue *vq = to_vvq(_vq);
	void *ret;
	unsigned int i;

	START_USE(vq);

294 295 296 297 298
	if (unlikely(vq->broken)) {
		END_USE(vq);
		return NULL;
	}

299 300 301 302 303 304
	if (!more_used(vq)) {
		pr_debug("No more buffers in queue\n");
		END_USE(vq);
		return NULL;
	}

305
	/* Only get used array entries after they have been exposed by host. */
306
	virtio_rmb();
307

308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327
	i = vq->vring.used->ring[vq->last_used_idx%vq->vring.num].id;
	*len = vq->vring.used->ring[vq->last_used_idx%vq->vring.num].len;

	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++;
	END_USE(vq);
	return ret;
}

328 329 330 331 332 333 334 335
static void vring_disable_cb(struct virtqueue *_vq)
{
	struct vring_virtqueue *vq = to_vvq(_vq);

	vq->vring.avail->flags |= VRING_AVAIL_F_NO_INTERRUPT;
}

static bool vring_enable_cb(struct virtqueue *_vq)
336 337 338 339 340 341 342 343
{
	struct vring_virtqueue *vq = to_vvq(_vq);

	START_USE(vq);

	/* We optimistically turn back on interrupts, then check if there was
	 * more to do. */
	vq->vring.avail->flags &= ~VRING_AVAIL_F_NO_INTERRUPT;
344
	virtio_mb();
345 346 347 348 349 350 351 352 353
	if (unlikely(more_used(vq))) {
		END_USE(vq);
		return false;
	}

	END_USE(vq);
	return true;
}

354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377
static void *vring_detach_unused_buf(struct virtqueue *_vq)
{
	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);
		END_USE(vq);
		return buf;
	}
	/* That should have freed everything. */
	BUG_ON(vq->num_free != vq->vring.num);

	END_USE(vq);
	return NULL;
}

378 379 380 381 382 383 384 385 386 387 388 389 390
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);
391 392
	if (vq->vq.callback)
		vq->vq.callback(&vq->vq);
393 394 395

	return IRQ_HANDLED;
}
396
EXPORT_SYMBOL_GPL(vring_interrupt);
397 398 399 400 401

static struct virtqueue_ops vring_vq_ops = {
	.add_buf = vring_add_buf,
	.get_buf = vring_get_buf,
	.kick = vring_kick,
402 403
	.disable_cb = vring_disable_cb,
	.enable_cb = vring_enable_cb,
404
	.detach_unused_buf = vring_detach_unused_buf,
405 406 407
};

struct virtqueue *vring_new_virtqueue(unsigned int num,
408
				      unsigned int vring_align,
409 410 411
				      struct virtio_device *vdev,
				      void *pages,
				      void (*notify)(struct virtqueue *),
412 413
				      void (*callback)(struct virtqueue *),
				      const char *name)
414 415 416 417
{
	struct vring_virtqueue *vq;
	unsigned int i;

418 419 420 421 422 423
	/* We assume num is a power of 2. */
	if (num & (num - 1)) {
		dev_warn(&vdev->dev, "Bad virtqueue length %u\n", num);
		return NULL;
	}

424 425 426 427
	vq = kmalloc(sizeof(*vq) + sizeof(void *)*num, GFP_KERNEL);
	if (!vq)
		return NULL;

428
	vring_init(&vq->vring, num, pages, vring_align);
429 430 431
	vq->vq.callback = callback;
	vq->vq.vdev = vdev;
	vq->vq.vq_ops = &vring_vq_ops;
432
	vq->vq.name = name;
433 434 435 436
	vq->notify = notify;
	vq->broken = false;
	vq->last_used_idx = 0;
	vq->num_added = 0;
437
	list_add_tail(&vq->vq.list, &vdev->vqs);
438 439 440 441
#ifdef DEBUG
	vq->in_use = false;
#endif

442 443
	vq->indirect = virtio_has_feature(vdev, VIRTIO_RING_F_INDIRECT_DESC);

444 445 446 447 448 449 450 451 452 453 454 455
	/* 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;
	for (i = 0; i < num-1; i++)
		vq->vring.desc[i].next = i+1;

	return &vq->vq;
}
456
EXPORT_SYMBOL_GPL(vring_new_virtqueue);
457 458 459

void vring_del_virtqueue(struct virtqueue *vq)
{
460
	list_del(&vq->list);
461 462
	kfree(to_vvq(vq));
}
463
EXPORT_SYMBOL_GPL(vring_del_virtqueue);
464

465 466 467 468 469 470 471
/* 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) {
472 473
		case VIRTIO_RING_F_INDIRECT_DESC:
			break;
474 475 476 477 478 479 480 481
		default:
			/* We don't understand this bit. */
			clear_bit(i, vdev->features);
		}
	}
}
EXPORT_SYMBOL_GPL(vring_transport_features);

482
MODULE_LICENSE("GPL");