Merge git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-virtio

* git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-virtio:
  virtio: enhance id_matching for virtio drivers
  virtio: fix id_matching for virtio drivers
  virtio: handle short buffers in virtio_rng.
  virtio_blk: add missing __dev{init,exit} markings
  virtio: indirect ring entries (VIRTIO_RING_F_INDIRECT_DESC)
  virtio: teach virtio_has_feature() about transport features
  virtio: expose features in sysfs
  virtio_pci: optional MSI-X support
  virtio_pci: split up vp_interrupt
  virtio: find_vqs/del_vqs virtio operations
  virtio: add names to virtqueue struct, mapping from devices to queues.
  virtio: meet virtio spec by finalizing features before using device
  virtio: fix obsolete documentation on probe function

drivers/block/virtio_blk.c

@@ -254,7 +254,7 @@ static int index_to_minor(int index)
 	return index << PART_BITS;
 }
 
-static int virtblk_probe(struct virtio_device *vdev)
+static int __devinit virtblk_probe(struct virtio_device *vdev)
 {
 	struct virtio_blk *vblk;
 	int err;
@@ -288,7 +288,7 @@ static int virtblk_probe(struct virtio_device *vdev)
 	sg_init_table(vblk->sg, vblk->sg_elems);
 
 	/* We expect one virtqueue, for output. */
-	vblk->vq = vdev->config->find_vq(vdev, 0, blk_done);
+	vblk->vq = virtio_find_single_vq(vdev, blk_done, "requests");
 	if (IS_ERR(vblk->vq)) {
 		err = PTR_ERR(vblk->vq);
 		goto out_free_vblk;
@@ -388,14 +388,14 @@ out_put_disk:
 out_mempool:
 	mempool_destroy(vblk->pool);
 out_free_vq:
-	vdev->config->del_vq(vblk->vq);
+	vdev->config->del_vqs(vdev);
 out_free_vblk:
 	kfree(vblk);
 out:
 	return err;
 }
 
-static void virtblk_remove(struct virtio_device *vdev)
+static void __devexit virtblk_remove(struct virtio_device *vdev)
 {
 	struct virtio_blk *vblk = vdev->priv;
 
@@ -409,7 +409,7 @@ static void virtblk_remove(struct virtio_device *vdev)
 	blk_cleanup_queue(vblk->disk->queue);
 	put_disk(vblk->disk);
 	mempool_destroy(vblk->pool);
-	vdev->config->del_vq(vblk->vq);
+	vdev->config->del_vqs(vdev);
 	kfree(vblk);
 }
 

drivers/char/hw_random/virtio-rng.c

@@ -35,13 +35,13 @@ static DECLARE_COMPLETION(have_data);
 
 static void random_recv_done(struct virtqueue *vq)
 {
-	int len;
+	unsigned int len;
 
 	/* We can get spurious callbacks, e.g. shared IRQs + virtio_pci. */
 	if (!vq->vq_ops->get_buf(vq, &len))
 		return;
 
-	data_left = len / sizeof(random_data[0]);
+	data_left += len;
 	complete(&have_data);
 }
 
@@ -49,7 +49,7 @@ static void register_buffer(void)
 {
 	struct scatterlist sg;
 
-	sg_init_one(&sg, random_data, RANDOM_DATA_SIZE);
+	sg_init_one(&sg, random_data+data_left, RANDOM_DATA_SIZE-data_left);
 	/* There should always be room for one buffer. */
 	if (vq->vq_ops->add_buf(vq, &sg, 0, 1, random_data) != 0)
 		BUG();
@@ -59,24 +59,32 @@ static void register_buffer(void)
 /* At least we don't udelay() in a loop like some other drivers. */
 static int virtio_data_present(struct hwrng *rng, int wait)
 {
-	if (data_left)
+	if (data_left >= sizeof(u32))
 		return 1;
 
+again:
 	if (!wait)
 		return 0;
 
 	wait_for_completion(&have_data);
+
+	/* Not enough?  Re-register. */
+	if (unlikely(data_left < sizeof(u32))) {
+		register_buffer();
+		goto again;
+	}
+
 	return 1;
 }
 
 /* virtio_data_present() must have succeeded before this is called. */
 static int virtio_data_read(struct hwrng *rng, u32 *data)
 {
-	BUG_ON(!data_left);
-
-	*data = random_data[--data_left];
+	BUG_ON(data_left < sizeof(u32));
+	data_left -= sizeof(u32);
+	*data = random_data[data_left / 4];
 
-	if (!data_left) {
+	if (data_left < sizeof(u32)) {
 		init_completion(&have_data);
 		register_buffer();
 	}
@@ -94,13 +102,13 @@ static int virtrng_probe(struct virtio_device *vdev)
 	int err;
 
 	/* We expect a single virtqueue. */
-	vq = vdev->config->find_vq(vdev, 0, random_recv_done);
+	vq = virtio_find_single_vq(vdev, random_recv_done, "input");
 	if (IS_ERR(vq))
 		return PTR_ERR(vq);
 
 	err = hwrng_register(&virtio_hwrng);
 	if (err) {
-		vdev->config->del_vq(vq);
+		vdev->config->del_vqs(vdev);
 		return err;
 	}
 
@@ -112,7 +120,7 @@ static void virtrng_remove(struct virtio_device *vdev)
 {
 	vdev->config->reset(vdev);
 	hwrng_unregister(&virtio_hwrng);
-	vdev->config->del_vq(vq);
+	vdev->config->del_vqs(vdev);
 }
 
 static struct virtio_device_id id_table[] = {

drivers/char/virtio_console.c

@@ -188,6 +188,9 @@ static void hvc_handle_input(struct virtqueue *vq)
  * Finally we put our input buffer in the input queue, ready to receive. */
 static int __devinit virtcons_probe(struct virtio_device *dev)
 {
+	vq_callback_t *callbacks[] = { hvc_handle_input, NULL};
+	const char *names[] = { "input", "output" };
+	struct virtqueue *vqs[2];
 	int err;
 
 	vdev = dev;
@@ -199,20 +202,15 @@ static int __devinit virtcons_probe(struct virtio_device *dev)
 		goto fail;
 	}
 
-	/* Find the input queue. */
+	/* Find the queues. */
 	/* FIXME: This is why we want to wean off hvc: we do nothing
 	 * when input comes in. */
-	in_vq = vdev->config->find_vq(vdev, 0, hvc_handle_input);
-	if (IS_ERR(in_vq)) {
-		err = PTR_ERR(in_vq);
+	err = vdev->config->find_vqs(vdev, 2, vqs, callbacks, names);
+	if (err)
 		goto free;
-	}
 
-	out_vq = vdev->config->find_vq(vdev, 1, NULL);
-	if (IS_ERR(out_vq)) {
-		err = PTR_ERR(out_vq);
-		goto free_in_vq;
-	}
+	in_vq = vqs[0];
+	out_vq = vqs[1];
 
 	/* Start using the new console output. */
 	virtio_cons.get_chars = get_chars;
@@ -233,17 +231,15 @@ static int __devinit virtcons_probe(struct virtio_device *dev)
 	hvc = hvc_alloc(0, 0, &virtio_cons, PAGE_SIZE);
 	if (IS_ERR(hvc)) {
 		err = PTR_ERR(hvc);
-		goto free_out_vq;
+		goto free_vqs;
 	}
 
 	/* Register the input buffer the first time. */
 	add_inbuf();
 	return 0;
 
-free_out_vq:
-	vdev->config->del_vq(out_vq);
-free_in_vq:
-	vdev->config->del_vq(in_vq);
+free_vqs:
+	vdev->config->del_vqs(vdev);
 free:
 	kfree(inbuf);
 fail:

drivers/lguest/lguest_device.c

@@ -228,7 +228,8 @@ extern void lguest_setup_irq(unsigned int irq);
  * function. */
 static struct virtqueue *lg_find_vq(struct virtio_device *vdev,
 				    unsigned index,
-				    void (*callback)(struct virtqueue *vq))
+				    void (*callback)(struct virtqueue *vq),
+				    const char *name)
 {
 	struct lguest_device *ldev = to_lgdev(vdev);
 	struct lguest_vq_info *lvq;
@@ -263,7 +264,7 @@ static struct virtqueue *lg_find_vq(struct virtio_device *vdev,
 	/* OK, tell virtio_ring.c to set up a virtqueue now we know its size
 	 * and we've got a pointer to its pages. */
 	vq = vring_new_virtqueue(lvq->config.num, LGUEST_VRING_ALIGN,
-				 vdev, lvq->pages, lg_notify, callback);
+				 vdev, lvq->pages, lg_notify, callback, name);
 	if (!vq) {
 		err = -ENOMEM;
 		goto unmap;
@@ -312,6 +313,38 @@ static void lg_del_vq(struct virtqueue *vq)
 	kfree(lvq);
 }
 
+static void lg_del_vqs(struct virtio_device *vdev)
+{
+	struct virtqueue *vq, *n;
+
+	list_for_each_entry_safe(vq, n, &vdev->vqs, list)
+		lg_del_vq(vq);
+}
+
+static int lg_find_vqs(struct virtio_device *vdev, unsigned nvqs,
+		       struct virtqueue *vqs[],
+		       vq_callback_t *callbacks[],
+		       const char *names[])
+{
+	struct lguest_device *ldev = to_lgdev(vdev);
+	int i;
+
+	/* We must have this many virtqueues. */
+	if (nvqs > ldev->desc->num_vq)
+		return -ENOENT;
+
+	for (i = 0; i < nvqs; ++i) {
+		vqs[i] = lg_find_vq(vdev, i, callbacks[i], names[i]);
+		if (IS_ERR(vqs[i]))
+			goto error;
+	}
+	return 0;
+
+error:
+	lg_del_vqs(vdev);
+	return PTR_ERR(vqs[i]);
+}
+
 /* The ops structure which hooks everything together. */
 static struct virtio_config_ops lguest_config_ops = {
 	.get_features = lg_get_features,
@@ -321,8 +354,8 @@ static struct virtio_config_ops lguest_config_ops = {
 	.get_status = lg_get_status,
 	.set_status = lg_set_status,
 	.reset = lg_reset,
-	.find_vq = lg_find_vq,
-	.del_vq = lg_del_vq,
+	.find_vqs = lg_find_vqs,
+	.del_vqs = lg_del_vqs,
 };
 
 /* The root device for the lguest virtio devices.  This makes them appear as

drivers/net/virtio_net.c

@@ -845,6 +845,10 @@ static int virtnet_probe(struct virtio_device *vdev)
 	int err;
 	struct net_device *dev;
 	struct virtnet_info *vi;
+	struct virtqueue *vqs[3];
+	vq_callback_t *callbacks[] = { skb_recv_done, skb_xmit_done, NULL};
+	const char *names[] = { "input", "output", "control" };
+	int nvqs;
 
 	/* Allocate ourselves a network device with room for our info */
 	dev = alloc_etherdev(sizeof(struct virtnet_info));
@@ -905,25 +909,19 @@ static int virtnet_probe(struct virtio_device *vdev)
 	if (virtio_has_feature(vdev, VIRTIO_NET_F_MRG_RXBUF))
 		vi->mergeable_rx_bufs = true;
 
-	/* We expect two virtqueues, receive then send. */
-	vi->rvq = vdev->config->find_vq(vdev, 0, skb_recv_done);
-	if (IS_ERR(vi->rvq)) {
-		err = PTR_ERR(vi->rvq);
+	/* We expect two virtqueues, receive then send,
+	 * and optionally control. */
+	nvqs = virtio_has_feature(vi->vdev, VIRTIO_NET_F_CTRL_VQ) ? 3 : 2;
+
+	err = vdev->config->find_vqs(vdev, nvqs, vqs, callbacks, names);
+	if (err)
 		goto free;
-	}
 
-	vi->svq = vdev->config->find_vq(vdev, 1, skb_xmit_done);
-	if (IS_ERR(vi->svq)) {
-		err = PTR_ERR(vi->svq);
-		goto free_recv;
-	}
+	vi->rvq = vqs[0];
+	vi->svq = vqs[1];
 
 	if (virtio_has_feature(vi->vdev, VIRTIO_NET_F_CTRL_VQ)) {
-		vi->cvq = vdev->config->find_vq(vdev, 2, NULL);
-		if (IS_ERR(vi->cvq)) {
-			err = PTR_ERR(vi->svq);
-			goto free_send;
-		}
+		vi->cvq = vqs[2];
 
 		if (virtio_has_feature(vi->vdev, VIRTIO_NET_F_CTRL_VLAN))
 			dev->features |= NETIF_F_HW_VLAN_FILTER;
@@ -941,7 +939,7 @@ static int virtnet_probe(struct virtio_device *vdev)
 	err = register_netdev(dev);
 	if (err) {
 		pr_debug("virtio_net: registering device failed\n");
-		goto free_ctrl;
+		goto free_vqs;
 	}
 
 	/* Last of all, set up some receive buffers. */
@@ -962,13 +960,8 @@ static int virtnet_probe(struct virtio_device *vdev)
 
 unregister:
 	unregister_netdev(dev);
-free_ctrl:
-	if (virtio_has_feature(vi->vdev, VIRTIO_NET_F_CTRL_VQ))
-		vdev->config->del_vq(vi->cvq);
-free_send:
-	vdev->config->del_vq(vi->svq);
-free_recv:
-	vdev->config->del_vq(vi->rvq);
+free_vqs:
+	vdev->config->del_vqs(vdev);
 free:
 	free_netdev(dev);
 	return err;
@@ -994,12 +987,10 @@ static void virtnet_remove(struct virtio_device *vdev)
 
 	BUG_ON(vi->num != 0);
 
-	vdev->config->del_vq(vi->svq);
-	vdev->config->del_vq(vi->rvq);
-	if (virtio_has_feature(vi->vdev, VIRTIO_NET_F_CTRL_VQ))
-		vdev->config->del_vq(vi->cvq);
 	unregister_netdev(vi->dev);
 
+	vdev->config->del_vqs(vi->vdev);
+
 	while (vi->pages)
 		__free_pages(get_a_page(vi, GFP_KERNEL), 0);
 

drivers/s390/kvm/kvm_virtio.c

@@ -174,7 +174,8 @@ static void kvm_notify(struct virtqueue *vq)
  */
 static struct virtqueue *kvm_find_vq(struct virtio_device *vdev,
 				     unsigned index,
-				    void (*callback)(struct virtqueue *vq))
+				     void (*callback)(struct virtqueue *vq),
+				     const char *name)
 {
 	struct kvm_device *kdev = to_kvmdev(vdev);
 	struct kvm_vqconfig *config;
@@ -194,7 +195,7 @@ static struct virtqueue *kvm_find_vq(struct virtio_device *vdev,
 
 	vq = vring_new_virtqueue(config->num, KVM_S390_VIRTIO_RING_ALIGN,
 				 vdev, (void *) config->address,
-				 kvm_notify, callback);
+				 kvm_notify, callback, name);
 	if (!vq) {
 		err = -ENOMEM;
 		goto unmap;
@@ -226,6 +227,38 @@ static void kvm_del_vq(struct virtqueue *vq)
 				       KVM_S390_VIRTIO_RING_ALIGN));
 }
 
+static void kvm_del_vqs(struct virtio_device *vdev)
+{
+	struct virtqueue *vq, *n;
+
+	list_for_each_entry_safe(vq, n, &vdev->vqs, list)
+		kvm_del_vq(vq);
+}
+
+static int kvm_find_vqs(struct virtio_device *vdev, unsigned nvqs,
+			struct virtqueue *vqs[],
+			vq_callback_t *callbacks[],
+			const char *names[])
+{
+	struct kvm_device *kdev = to_kvmdev(vdev);
+	int i;
+
+	/* We must have this many virtqueues. */
+	if (nvqs > kdev->desc->num_vq)
+		return -ENOENT;
+
+	for (i = 0; i < nvqs; ++i) {
+		vqs[i] = kvm_find_vq(vdev, i, callbacks[i], names[i]);
+		if (IS_ERR(vqs[i]))
+			goto error;
+	}
+	return 0;
+
+error:
+	kvm_del_vqs(vdev);
+	return PTR_ERR(vqs[i]);
+}
+
 /*
  * The config ops structure as defined by virtio config
  */
@@ -237,8 +270,8 @@ static struct virtio_config_ops kvm_vq_configspace_ops = {
 	.get_status = kvm_get_status,
 	.set_status = kvm_set_status,
 	.reset = kvm_reset,
-	.find_vq = kvm_find_vq,
-	.del_vq = kvm_del_vq,
+	.find_vqs = kvm_find_vqs,
+	.del_vqs = kvm_del_vqs,
 };
 
 /*

drivers/virtio/virtio.c

@@ -31,21 +31,37 @@ static ssize_t modalias_show(struct device *_d,
 	return sprintf(buf, "virtio:d%08Xv%08X\n",
 		       dev->id.device, dev->id.vendor);
 }
+static ssize_t features_show(struct device *_d,
+			     struct device_attribute *attr, char *buf)
+{
+	struct virtio_device *dev = container_of(_d, struct virtio_device, dev);
+	unsigned int i;
+	ssize_t len = 0;
+
+	/* We actually represent this as a bitstring, as it could be
+	 * arbitrary length in future. */
+	for (i = 0; i < ARRAY_SIZE(dev->features)*BITS_PER_LONG; i++)
+		len += sprintf(buf+len, "%c",
+			       test_bit(i, dev->features) ? '1' : '0');
+	len += sprintf(buf+len, "\n");
+	return len;
+}
 static struct device_attribute virtio_dev_attrs[] = {
 	__ATTR_RO(device),
 	__ATTR_RO(vendor),
 	__ATTR_RO(status),
 	__ATTR_RO(modalias),
+	__ATTR_RO(features),
 	__ATTR_NULL
 };
 
 static inline int virtio_id_match(const struct virtio_device *dev,
 				  const struct virtio_device_id *id)
 {
-	if (id->device != dev->id.device)
+	if (id->device != dev->id.device && id->device != VIRTIO_DEV_ANY_ID)
 		return 0;
 
-	return id->vendor == VIRTIO_DEV_ANY_ID || id->vendor != dev->id.vendor;
+	return id->vendor == VIRTIO_DEV_ANY_ID || id->vendor == dev->id.vendor;
 }
 
 /* This looks through all the IDs a driver claims to support.  If any of them
@@ -118,13 +134,14 @@ static int virtio_dev_probe(struct device *_d)
 		if (device_features & (1 << i))
 			set_bit(i, dev->features);
 
+	dev->config->finalize_features(dev);
+
 	err = drv->probe(dev);
 	if (err)
 		add_status(dev, VIRTIO_CONFIG_S_FAILED);
-	else {
-		dev->config->finalize_features(dev);
+	else
 		add_status(dev, VIRTIO_CONFIG_S_DRIVER_OK);
-	}
+
 	return err;
 }
 
@@ -185,6 +202,8 @@ int register_virtio_device(struct virtio_device *dev)
 	/* Acknowledge that we've seen the device. */
 	add_status(dev, VIRTIO_CONFIG_S_ACKNOWLEDGE);
 
+	INIT_LIST_HEAD(&dev->vqs);
+
 	/* device_register() causes the bus infrastructure to look for a
 	 * matching driver. */
 	err = device_register(&dev->dev);

drivers/virtio/virtio_balloon.c

@@ -204,6 +204,9 @@ static int balloon(void *_vballoon)
 static int virtballoon_probe(struct virtio_device *vdev)
 {
 	struct virtio_balloon *vb;
+	struct virtqueue *vqs[2];
+	vq_callback_t *callbacks[] = { balloon_ack, balloon_ack };
+	const char *names[] = { "inflate", "deflate" };
 	int err;
 
 	vdev->priv = vb = kmalloc(sizeof(*vb), GFP_KERNEL);
@@ -218,22 +221,17 @@ static int virtballoon_probe(struct virtio_device *vdev)
 	vb->vdev = vdev;
 
 	/* We expect two virtqueues. */
-	vb->inflate_vq = vdev->config->find_vq(vdev, 0, balloon_ack);
-	if (IS_ERR(vb->inflate_vq)) {
-		err = PTR_ERR(vb->inflate_vq);
+	err = vdev->config->find_vqs(vdev, 2, vqs, callbacks, names);
+	if (err)
 		goto out_free_vb;
-	}
 
-	vb->deflate_vq = vdev->config->find_vq(vdev, 1, balloon_ack);
-	if (IS_ERR(vb->deflate_vq)) {
-		err = PTR_ERR(vb->deflate_vq);
-		goto out_del_inflate_vq;
-	}
+	vb->inflate_vq = vqs[0];
+	vb->deflate_vq = vqs[1];
 
 	vb->thread = kthread_run(balloon, vb, "vballoon");
 	if (IS_ERR(vb->thread)) {
 		err = PTR_ERR(vb->thread);
-		goto out_del_deflate_vq;
+		goto out_del_vqs;
 	}
 
 	vb->tell_host_first
@@ -241,10 +239,8 @@ static int virtballoon_probe(struct virtio_device *vdev)
 
 	return 0;
 
-out_del_deflate_vq:
-	vdev->config->del_vq(vb->deflate_vq);
-out_del_inflate_vq:
-	vdev->config->del_vq(vb->inflate_vq);
+out_del_vqs:
+	vdev->config->del_vqs(vdev);
 out_free_vb:
 	kfree(vb);
 out:
@@ -264,8 +260,7 @@ static void virtballoon_remove(struct virtio_device *vdev)
 	/* Now we reset the device so we can clean up the queues. */
 	vdev->config->reset(vdev);
 
-	vdev->config->del_vq(vb->deflate_vq);
-	vdev->config->del_vq(vb->inflate_vq);
+	vdev->config->del_vqs(vdev);
 	kfree(vb);
 }
 

drivers/virtio/virtio_pci.c

@@ -42,6 +42,26 @@ struct virtio_pci_device
 	/* a list of queues so we can dispatch IRQs */
 	spinlock_t lock;
 	struct list_head virtqueues;
+
+	/* MSI-X support */
+	int msix_enabled;
+	int intx_enabled;
+	struct msix_entry *msix_entries;
+	/* Name strings for interrupts. This size should be enough,
+	 * and I'm too lazy to allocate each name separately. */
+	char (*msix_names)[256];
+	/* Number of available vectors */
+	unsigned msix_vectors;
+	/* Vectors allocated */
+	unsigned msix_used_vectors;
+};
+
+/* Constants for MSI-X */
+/* Use first vector for configuration changes, second and the rest for
+ * virtqueues Thus, we need at least 2 vectors for MSI. */
+enum {
+	VP_MSIX_CONFIG_VECTOR = 0,
+	VP_MSIX_VQ_VECTOR = 1,
 };
 
 struct virtio_pci_vq_info
@@ -60,6 +80,9 @@ struct virtio_pci_vq_info
 
 	/* the list node for the virtqueues list */
 	struct list_head node;
+
+	/* MSI-X vector (or none) */
+	unsigned vector;
 };
 
 /* Qumranet donated their vendor ID for devices 0x1000 thru 0x10FF. */
@@ -109,7 +132,8 @@ static void vp_get(struct virtio_device *vdev, unsigned offset,
 		   void *buf, unsigned len)
 {
 	struct virtio_pci_device *vp_dev = to_vp_device(vdev);
-	void __iomem *ioaddr = vp_dev->ioaddr + VIRTIO_PCI_CONFIG + offset;
+	void __iomem *ioaddr = vp_dev->ioaddr +
+				VIRTIO_PCI_CONFIG(vp_dev) + offset;
 	u8 *ptr = buf;
 	int i;
 
@@ -123,7 +147,8 @@ static void vp_set(struct virtio_device *vdev, unsigned offset,
 		   const void *buf, unsigned len)
 {
 	struct virtio_pci_device *vp_dev = to_vp_device(vdev);
-	void __iomem *ioaddr = vp_dev->ioaddr + VIRTIO_PCI_CONFIG + offset;
+	void __iomem *ioaddr = vp_dev->ioaddr +
+				VIRTIO_PCI_CONFIG(vp_dev) + offset;
 	const u8 *ptr = buf;
 	int i;
 
@@ -164,6 +189,37 @@ static void vp_notify(struct virtqueue *vq)
 	iowrite16(info->queue_index, vp_dev->ioaddr + VIRTIO_PCI_QUEUE_NOTIFY);
 }
 
+/* Handle a configuration change: Tell driver if it wants to know. */
+static irqreturn_t vp_config_changed(int irq, void *opaque)
+{
+	struct virtio_pci_device *vp_dev = opaque;
+	struct virtio_driver *drv;
+	drv = container_of(vp_dev->vdev.dev.driver,
+			   struct virtio_driver, driver);
+
+	if (drv && drv->config_changed)
+		drv->config_changed(&vp_dev->vdev);
+	return IRQ_HANDLED;
+}
+
+/* Notify all virtqueues on an interrupt. */
+static irqreturn_t vp_vring_interrupt(int irq, void *opaque)
+{
+	struct virtio_pci_device *vp_dev = opaque;
+	struct virtio_pci_vq_info *info;
+	irqreturn_t ret = IRQ_NONE;
+	unsigned long flags;
+
+	spin_lock_irqsave(&vp_dev->lock, flags);
+	list_for_each_entry(info, &vp_dev->virtqueues, node) {
+		if (vring_interrupt(irq, info->vq) == IRQ_HANDLED)
+			ret = IRQ_HANDLED;
+	}
+	spin_unlock_irqrestore(&vp_dev->lock, flags);
+
+	return ret;
+}
+
 /* A small wrapper to also acknowledge the interrupt when it's handled.
  * I really need an EIO hook for the vring so I can ack the interrupt once we
  * know that we'll be handling the IRQ but before we invoke the callback since
@@ -173,9 +229,6 @@ static void vp_notify(struct virtqueue *vq)
 static irqreturn_t vp_interrupt(int irq, void *opaque)
 {
 	struct virtio_pci_device *vp_dev = opaque;
-	struct virtio_pci_vq_info *info;
-	irqreturn_t ret = IRQ_NONE;
-	unsigned long flags;
 	u8 isr;
 
 	/* reading the ISR has the effect of also clearing it so it's very
@@ -187,34 +240,137 @@ static irqreturn_t vp_interrupt(int irq, void *opaque)
 		return IRQ_NONE;
 
 	/* Configuration change?  Tell driver if it wants to know. */
-	if (isr & VIRTIO_PCI_ISR_CONFIG) {
-		struct virtio_driver *drv;
-		drv = container_of(vp_dev->vdev.dev.driver,
-				   struct virtio_driver, driver);
+	if (isr & VIRTIO_PCI_ISR_CONFIG)
+		vp_config_changed(irq, opaque);
 
-		if (drv && drv->config_changed)
-			drv->config_changed(&vp_dev->vdev);
+	return vp_vring_interrupt(irq, opaque);
+}
+
+static void vp_free_vectors(struct virtio_device *vdev)
+{
+	struct virtio_pci_device *vp_dev = to_vp_device(vdev);
+	int i;
+
+	if (vp_dev->intx_enabled) {
+		free_irq(vp_dev->pci_dev->irq, vp_dev);
+		vp_dev->intx_enabled = 0;
 	}
 
-	spin_lock_irqsave(&vp_dev->lock, flags);
-	list_for_each_entry(info, &vp_dev->virtqueues, node) {
-		if (vring_interrupt(irq, info->vq) == IRQ_HANDLED)
-			ret = IRQ_HANDLED;
+	for (i = 0; i < vp_dev->msix_used_vectors; ++i)
+		free_irq(vp_dev->msix_entries[i].vector, vp_dev);
+	vp_dev->msix_used_vectors = 0;
+
+	if (vp_dev->msix_enabled) {
+		/* Disable the vector used for configuration */
+		iowrite16(VIRTIO_MSI_NO_VECTOR,
+			  vp_dev->ioaddr + VIRTIO_MSI_CONFIG_VECTOR);
+		/* Flush the write out to device */
+		ioread16(vp_dev->ioaddr + VIRTIO_MSI_CONFIG_VECTOR);
+
+		vp_dev->msix_enabled = 0;
+		pci_disable_msix(vp_dev->pci_dev);
 	}
-	spin_unlock_irqrestore(&vp_dev->lock, flags);
+}
 
-	return ret;
+static int vp_enable_msix(struct pci_dev *dev, struct msix_entry *entries,
+			  int *options, int noptions)
+{
+	int i;
+	for (i = 0; i < noptions; ++i)
+		if (!pci_enable_msix(dev, entries, options[i]))
+			return options[i];
+	return -EBUSY;
+}
+
+static int vp_request_vectors(struct virtio_device *vdev, unsigned max_vqs)
+{
+	struct virtio_pci_device *vp_dev = to_vp_device(vdev);
+	const char *name = dev_name(&vp_dev->vdev.dev);
+	unsigned i, v;
+	int err = -ENOMEM;
+	/* We want at most one vector per queue and one for config changes.
+	 * Fallback to separate vectors for config and a shared for queues.
+	 * Finally fall back to regular interrupts. */
+	int options[] = { max_vqs + 1, 2 };
+	int nvectors = max(options[0], options[1]);
+
+	vp_dev->msix_entries = kmalloc(nvectors * sizeof *vp_dev->msix_entries,
+				       GFP_KERNEL);
+	if (!vp_dev->msix_entries)
+		goto error_entries;
+	vp_dev->msix_names = kmalloc(nvectors * sizeof *vp_dev->msix_names,
+				     GFP_KERNEL);
+	if (!vp_dev->msix_names)
+		goto error_names;
+
+	for (i = 0; i < nvectors; ++i)
+		vp_dev->msix_entries[i].entry = i;
+
+	err = vp_enable_msix(vp_dev->pci_dev, vp_dev->msix_entries,
+			     options, ARRAY_SIZE(options));
+	if (err < 0) {
+		/* Can't allocate enough MSI-X vectors, use regular interrupt */
+		vp_dev->msix_vectors = 0;
+		err = request_irq(vp_dev->pci_dev->irq, vp_interrupt,
+				  IRQF_SHARED, name, vp_dev);
+		if (err)
+			goto error_irq;
+		vp_dev->intx_enabled = 1;
+	} else {
+		vp_dev->msix_vectors = err;
+		vp_dev->msix_enabled = 1;
+
+		/* Set the vector used for configuration */
+		v = vp_dev->msix_used_vectors;
+		snprintf(vp_dev->msix_names[v], sizeof *vp_dev->msix_names,
+			 "%s-config", name);
+		err = request_irq(vp_dev->msix_entries[v].vector,
+				  vp_config_changed, 0, vp_dev->msix_names[v],
+				  vp_dev);
+		if (err)
+			goto error_irq;
+		++vp_dev->msix_used_vectors;
+
+		iowrite16(v, vp_dev->ioaddr + VIRTIO_MSI_CONFIG_VECTOR);
+		/* Verify we had enough resources to assign the vector */
+		v = ioread16(vp_dev->ioaddr + VIRTIO_MSI_CONFIG_VECTOR);
+		if (v == VIRTIO_MSI_NO_VECTOR) {
+			err = -EBUSY;
+			goto error_irq;
+		}
+	}
+
+	if (vp_dev->msix_vectors && vp_dev->msix_vectors != max_vqs + 1) {
+		/* Shared vector for all VQs */
+		v = vp_dev->msix_used_vectors;
+		snprintf(vp_dev->msix_names[v], sizeof *vp_dev->msix_names,
+			 "%s-virtqueues", name);
+		err = request_irq(vp_dev->msix_entries[v].vector,
+				  vp_vring_interrupt, 0, vp_dev->msix_names[v],
+				  vp_dev);
+		if (err)
+			goto error_irq;
+		++vp_dev->msix_used_vectors;
+	}
+	return 0;
+error_irq:
+	vp_free_vectors(vdev);
+	kfree(vp_dev->msix_names);
+error_names:
+	kfree(vp_dev->msix_entries);
+error_entries:
+	return err;
 }
 
-/* the config->find_vq() implementation */
 static struct virtqueue *vp_find_vq(struct virtio_device *vdev, unsigned index,
-				    void (*callback)(struct virtqueue *vq))
+				    void (*callback)(struct virtqueue *vq),
+				    const char *name)
 {
 	struct virtio_pci_device *vp_dev = to_vp_device(vdev);
 	struct virtio_pci_vq_info *info;
 	struct virtqueue *vq;
 	unsigned long flags, size;
-	u16 num;
+	u16 num, vector;
 	int err;
 
 	/* Select the queue we're interested in */
@@ -233,6 +389,7 @@ static struct virtqueue *vp_find_vq(struct virtio_device *vdev, unsigned index,
 
 	info->queue_index = index;
 	info->num = num;
+	info->vector = VIRTIO_MSI_NO_VECTOR;
 
 	size = PAGE_ALIGN(vring_size(num, VIRTIO_PCI_VRING_ALIGN));
 	info->queue = alloc_pages_exact(size, GFP_KERNEL|__GFP_ZERO);
@@ -247,7 +404,7 @@ static struct virtqueue *vp_find_vq(struct virtio_device *vdev, unsigned index,
 
 	/* create the vring */
 	vq = vring_new_virtqueue(info->num, VIRTIO_PCI_VRING_ALIGN,
-				 vdev, info->queue, vp_notify, callback);
+				 vdev, info->queue, vp_notify, callback, name);
 	if (!vq) {
 		err = -ENOMEM;
 		goto out_activate_queue;
@@ -256,12 +413,43 @@ static struct virtqueue *vp_find_vq(struct virtio_device *vdev, unsigned index,
 	vq->priv = info;
 	info->vq = vq;
 
+	/* allocate per-vq vector if available and necessary */
+	if (callback && vp_dev->msix_used_vectors < vp_dev->msix_vectors) {
+		vector = vp_dev->msix_used_vectors;
+		snprintf(vp_dev->msix_names[vector], sizeof *vp_dev->msix_names,
+			 "%s-%s", dev_name(&vp_dev->vdev.dev), name);
+		err = request_irq(vp_dev->msix_entries[vector].vector,
+				  vring_interrupt, 0,
+				  vp_dev->msix_names[vector], vq);
+		if (err)
+			goto out_request_irq;
+		info->vector = vector;
+		++vp_dev->msix_used_vectors;
+	} else
+		vector = VP_MSIX_VQ_VECTOR;
+
+	 if (callback && vp_dev->msix_enabled) {
+		iowrite16(vector, vp_dev->ioaddr + VIRTIO_MSI_QUEUE_VECTOR);
+		vector = ioread16(vp_dev->ioaddr + VIRTIO_MSI_QUEUE_VECTOR);
+		if (vector == VIRTIO_MSI_NO_VECTOR) {
+			err = -EBUSY;
+			goto out_assign;
+		}
+	}
+
 	spin_lock_irqsave(&vp_dev->lock, flags);
 	list_add(&info->node, &vp_dev->virtqueues);
 	spin_unlock_irqrestore(&vp_dev->lock, flags);
 
 	return vq;
 
+out_assign:
+	if (info->vector != VIRTIO_MSI_NO_VECTOR) {
+		free_irq(vp_dev->msix_entries[info->vector].vector, vq);
+		--vp_dev->msix_used_vectors;
+	}
+out_request_irq:
+	vring_del_virtqueue(vq);
 out_activate_queue:
 	iowrite32(0, vp_dev->ioaddr + VIRTIO_PCI_QUEUE_PFN);
 	free_pages_exact(info->queue, size);
@@ -270,21 +458,27 @@ out_info:
 	return ERR_PTR(err);
 }
 
-/* the config->del_vq() implementation */
 static void vp_del_vq(struct virtqueue *vq)
 {
 	struct virtio_pci_device *vp_dev = to_vp_device(vq->vdev);
 	struct virtio_pci_vq_info *info = vq->priv;
-	unsigned long flags, size;
+	unsigned long size;
 
-	spin_lock_irqsave(&vp_dev->lock, flags);
-	list_del(&info->node);
-	spin_unlock_irqrestore(&vp_dev->lock, flags);
+	iowrite16(info->queue_index, vp_dev->ioaddr + VIRTIO_PCI_QUEUE_SEL);
+
+	if (info->vector != VIRTIO_MSI_NO_VECTOR)
+		free_irq(vp_dev->msix_entries[info->vector].vector, vq);
+
+	if (vp_dev->msix_enabled) {
+		iowrite16(VIRTIO_MSI_NO_VECTOR,
+			  vp_dev->ioaddr + VIRTIO_MSI_QUEUE_VECTOR);
+		/* Flush the write out to device */
+		ioread8(vp_dev->ioaddr + VIRTIO_PCI_ISR);
+	}
 
 	vring_del_virtqueue(vq);
 
 	/* Select and deactivate the queue */
-	iowrite16(info->queue_index, vp_dev->ioaddr + VIRTIO_PCI_QUEUE_SEL);
 	iowrite32(0, vp_dev->ioaddr + VIRTIO_PCI_QUEUE_PFN);
 
 	size = PAGE_ALIGN(vring_size(info->num, VIRTIO_PCI_VRING_ALIGN));
@@ -292,14 +486,57 @@ static void vp_del_vq(struct virtqueue *vq)
 	kfree(info);
 }
 
+/* the config->del_vqs() implementation */
+static void vp_del_vqs(struct virtio_device *vdev)
+{
+	struct virtqueue *vq, *n;
+
+	list_for_each_entry_safe(vq, n, &vdev->vqs, list)
+		vp_del_vq(vq);
+
+	vp_free_vectors(vdev);
+}
+
+/* the config->find_vqs() implementation */
+static int vp_find_vqs(struct virtio_device *vdev, unsigned nvqs,
+		       struct virtqueue *vqs[],
+		       vq_callback_t *callbacks[],
+		       const char *names[])
+{
+	int vectors = 0;
+	int i, err;
+
+	/* How many vectors would we like? */
+	for (i = 0; i < nvqs; ++i)
+		if (callbacks[i])
+			++vectors;
+
+	err = vp_request_vectors(vdev, vectors);
+	if (err)
+		goto error_request;
+
+	for (i = 0; i < nvqs; ++i) {
+		vqs[i] = vp_find_vq(vdev, i, callbacks[i], names[i]);
+		if (IS_ERR(vqs[i]))
+			goto error_find;
+	}
+	return 0;
+
+error_find:
+	vp_del_vqs(vdev);
+
+error_request:
+	return PTR_ERR(vqs[i]);
+}
+
 static struct virtio_config_ops virtio_pci_config_ops = {
 	.get		= vp_get,
 	.set		= vp_set,
 	.get_status	= vp_get_status,
 	.set_status	= vp_set_status,
 	.reset		= vp_reset,
-	.find_vq	= vp_find_vq,
-	.del_vq		= vp_del_vq,
+	.find_vqs	= vp_find_vqs,
+	.del_vqs	= vp_del_vqs,
 	.get_features	= vp_get_features,
 	.finalize_features = vp_finalize_features,
 };
@@ -310,7 +547,7 @@ static void virtio_pci_release_dev(struct device *_d)
 	struct virtio_pci_device *vp_dev = to_vp_device(dev);
 	struct pci_dev *pci_dev = vp_dev->pci_dev;
 
-	free_irq(pci_dev->irq, vp_dev);
+	vp_del_vqs(dev);
 	pci_set_drvdata(pci_dev, NULL);
 	pci_iounmap(pci_dev, vp_dev->ioaddr);
 	pci_release_regions(pci_dev);
@@ -369,21 +606,13 @@ static int __devinit virtio_pci_probe(struct pci_dev *pci_dev,
 	vp_dev->vdev.id.vendor = pci_dev->subsystem_vendor;
 	vp_dev->vdev.id.device = pci_dev->subsystem_device;
 
-	/* register a handler for the queue with the PCI device's interrupt */
-	err = request_irq(vp_dev->pci_dev->irq, vp_interrupt, IRQF_SHARED,
-			  dev_name(&vp_dev->vdev.dev), vp_dev);
-	if (err)
-		goto out_set_drvdata;
-
 	/* finally register the virtio device */
 	err = register_virtio_device(&vp_dev->vdev);
 	if (err)
-		goto out_req_irq;
+		goto out_set_drvdata;
 
 	return 0;
 
-out_req_irq:
-	free_irq(pci_dev->irq, vp_dev);
 out_set_drvdata:
 	pci_set_drvdata(pci_dev, NULL);
 	pci_iounmap(pci_dev, vp_dev->ioaddr);

drivers/virtio/virtio_ring.c

@@ -23,21 +23,30 @@
 
 #ifdef DEBUG
 /* For development, we want to crash whenever the ring is screwed. */
-#define BAD_RING(_vq, fmt...)			\
-	do { dev_err(&(_vq)->vq.vdev->dev, fmt); BUG(); } while(0)
+#define BAD_RING(_vq, fmt, args...)				\
+	do {							\
+		dev_err(&(_vq)->vq.vdev->dev,			\
+			"%s:"fmt, (_vq)->vq.name, ##args);	\
+		BUG();						\
+	} while (0)
 /* Caller is supposed to guarantee no reentry. */
 #define START_USE(_vq)						\
 	do {							\
 		if ((_vq)->in_use)				\
-			panic("in_use = %i\n", (_vq)->in_use);	\
+			panic("%s:in_use = %i\n",		\
+			      (_vq)->vq.name, (_vq)->in_use);	\
 		(_vq)->in_use = __LINE__;			\
 		mb();						\
-	} while(0)
+	} while (0)
 #define END_USE(_vq) \
 	do { BUG_ON(!(_vq)->in_use); (_vq)->in_use = 0; mb(); } while(0)
 #else
-#define BAD_RING(_vq, fmt...)			\
-	do { dev_err(&_vq->vq.vdev->dev, fmt); (_vq)->broken = true; } while(0)
+#define BAD_RING(_vq, fmt, args...)				\
+	do {							\
+		dev_err(&_vq->vq.vdev->dev,			\
+			"%s:"fmt, (_vq)->vq.name, ##args);	\
+		(_vq)->broken = true;				\
+	} while (0)
 #define START_USE(vq)
 #define END_USE(vq)
 #endif
@@ -52,6 +61,9 @@ struct vring_virtqueue
 	/* Other side has made a mess, don't try any more. */
 	bool broken;
 
+	/* Host supports indirect buffers */
+	bool indirect;
+
 	/* Number of free buffers */
 	unsigned int num_free;
 	/* Head of free buffer list. */
@@ -76,6 +88,55 @@ struct vring_virtqueue
 
 #define to_vvq(_vq) container_of(_vq, struct vring_virtqueue, vq)
 
+/* 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;
+}
+
 static int vring_add_buf(struct virtqueue *_vq,
 			 struct scatterlist sg[],
 			 unsigned int out,
@@ -85,12 +146,21 @@ static int vring_add_buf(struct virtqueue *_vq,
 	struct vring_virtqueue *vq = to_vvq(_vq);
 	unsigned int i, avail, head, uninitialized_var(prev);
 
+	START_USE(vq);
+
 	BUG_ON(data == NULL);
+
+	/* 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;
+	}
+
 	BUG_ON(out + in > vq->vring.num);
 	BUG_ON(out + in == 0);
 
-	START_USE(vq);
-
 	if (vq->num_free < out + in) {
 		pr_debug("Can't add buf len %i - avail = %i\n",
 			 out + in, vq->num_free);
@@ -127,6 +197,7 @@ static int vring_add_buf(struct virtqueue *_vq,
 	/* Update free pointer */
 	vq->free_head = i;
 
+add_head:
 	/* Set token. */
 	vq->data[head] = data;
 
@@ -170,6 +241,11 @@ static void detach_buf(struct vring_virtqueue *vq, unsigned int head)
 
 	/* Put back on free list: find end */
 	i = head;
+
+	/* Free the indirect table */
+	if (vq->vring.desc[i].flags & VRING_DESC_F_INDIRECT)
+		kfree(phys_to_virt(vq->vring.desc[i].addr));
+
 	while (vq->vring.desc[i].flags & VRING_DESC_F_NEXT) {
 		i = vq->vring.desc[i].next;
 		vq->num_free++;
@@ -284,7 +360,8 @@ struct virtqueue *vring_new_virtqueue(unsigned int num,
 				      struct virtio_device *vdev,
 				      void *pages,
 				      void (*notify)(struct virtqueue *),
-				      void (*callback)(struct virtqueue *))
+				      void (*callback)(struct virtqueue *),
+				      const char *name)
 {
 	struct vring_virtqueue *vq;
 	unsigned int i;
@@ -303,14 +380,18 @@ struct virtqueue *vring_new_virtqueue(unsigned int num,
 	vq->vq.callback = callback;
 	vq->vq.vdev = vdev;
 	vq->vq.vq_ops = &vring_vq_ops;
+	vq->vq.name = name;
 	vq->notify = notify;
 	vq->broken = false;
 	vq->last_used_idx = 0;
 	vq->num_added = 0;
+	list_add_tail(&vq->vq.list, &vdev->vqs);
 #ifdef DEBUG
 	vq->in_use = false;
 #endif
 
+	vq->indirect = virtio_has_feature(vdev, VIRTIO_RING_F_INDIRECT_DESC);
+
 	/* No callback?  Tell other side not to bother us. */
 	if (!callback)
 		vq->vring.avail->flags |= VRING_AVAIL_F_NO_INTERRUPT;
@@ -327,6 +408,7 @@ EXPORT_SYMBOL_GPL(vring_new_virtqueue);
 
 void vring_del_virtqueue(struct virtqueue *vq)
 {
+	list_del(&vq->list);
 	kfree(to_vvq(vq));
 }
 EXPORT_SYMBOL_GPL(vring_del_virtqueue);
@@ -338,6 +420,8 @@ void vring_transport_features(struct virtio_device *vdev)
 
 	for (i = VIRTIO_TRANSPORT_F_START; i < VIRTIO_TRANSPORT_F_END; i++) {
 		switch (i) {
+		case VIRTIO_RING_F_INDIRECT_DESC:
+			break;
 		default:
 			/* We don't understand this bit. */
 			clear_bit(i, vdev->features);

include/linux/virtio.h

@@ -10,14 +10,17 @@
 
 /**
  * virtqueue - a queue to register buffers for sending or receiving.
+ * @list: the chain of virtqueues for this device
  * @callback: the function to call when buffers are consumed (can be NULL).
+ * @name: the name of this virtqueue (mainly for debugging)
  * @vdev: the virtio device this queue was created for.
  * @vq_ops: the operations for this virtqueue (see below).
  * @priv: a pointer for the virtqueue implementation to use.
  */
-struct virtqueue
-{
+struct virtqueue {
+	struct list_head list;
 	void (*callback)(struct virtqueue *vq);
+	const char *name;
 	struct virtio_device *vdev;
 	struct virtqueue_ops *vq_ops;
 	void *priv;
@@ -76,15 +79,16 @@ struct virtqueue_ops {
  * @dev: underlying device.
  * @id: the device type identification (used to match it with a driver).
  * @config: the configuration ops for this device.
+ * @vqs: the list of virtqueues for this device.
  * @features: the features supported by both driver and device.
  * @priv: private pointer for the driver's use.
  */
-struct virtio_device
-{
+struct virtio_device {
 	int index;
 	struct device dev;
 	struct virtio_device_id id;
 	struct virtio_config_ops *config;
+	struct list_head vqs;
 	/* Note that this is a Linux set_bit-style bitmap. */
 	unsigned long features[1];
 	void *priv;
@@ -99,8 +103,7 @@ void unregister_virtio_device(struct virtio_device *dev);
  * @id_table: the ids serviced by this driver.
  * @feature_table: an array of feature numbers supported by this device.
  * @feature_table_size: number of entries in the feature table array.
- * @probe: the function to call when a device is found.  Returns a token for
- *    remove, or PTR_ERR().
+ * @probe: the function to call when a device is found.  Returns 0 or -errno.
  * @remove: the function when a device is removed.
  * @config_changed: optional function to call when the device configuration
  *    changes; may be called in interrupt context.

include/linux/virtio_config.h

@@ -29,6 +29,7 @@
 #define VIRTIO_F_NOTIFY_ON_EMPTY	24
 
 #ifdef __KERNEL__
+#include <linux/err.h>
 #include <linux/virtio.h>
 
 /**
@@ -49,15 +50,26 @@
  * @set_status: write the status byte
  *	vdev: the virtio_device
  *	status: the new status byte
+ * @request_vqs: request the specified number of virtqueues
+ *	vdev: the virtio_device
+ *	max_vqs: the max number of virtqueues we want
+ *      If supplied, must call before any virtqueues are instantiated.
+ *      To modify the max number of virtqueues after request_vqs has been
+ *      called, call free_vqs and then request_vqs with a new value.
+ * @free_vqs: cleanup resources allocated by request_vqs
+ *	vdev: the virtio_device
+ *      If supplied, must call after all virtqueues have been deleted.
  * @reset: reset the device
  *	vdev: the virtio device
  *	After this, status and feature negotiation must be done again
- * @find_vq: find a virtqueue and instantiate it.
+ * @find_vqs: find virtqueues and instantiate them.
  *	vdev: the virtio_device
- *	index: the 0-based virtqueue number in case there's more than one.
- *	callback: the virqtueue callback
- *	Returns the new virtqueue or ERR_PTR() (eg. -ENOENT).
- * @del_vq: free a virtqueue found by find_vq().
+ *	nvqs: the number of virtqueues to find
+ *	vqs: on success, includes new virtqueues
+ *	callbacks: array of callbacks, for each virtqueue
+ *	names: array of virtqueue names (mainly for debugging)
+ *	Returns 0 on success or error status
+ * @del_vqs: free virtqueues found by find_vqs().
  * @get_features: get the array of feature bits for this device.
  *	vdev: the virtio_device
  *	Returns the first 32 feature bits (all we currently need).
@@ -66,6 +78,7 @@
  *	This gives the final feature bits for the device: it can change
  *	the dev->feature bits if it wants.
  */
+typedef void vq_callback_t(struct virtqueue *);
 struct virtio_config_ops
 {
 	void (*get)(struct virtio_device *vdev, unsigned offset,
@@ -75,10 +88,11 @@ struct virtio_config_ops
 	u8 (*get_status)(struct virtio_device *vdev);
 	void (*set_status)(struct virtio_device *vdev, u8 status);
 	void (*reset)(struct virtio_device *vdev);
-	struct virtqueue *(*find_vq)(struct virtio_device *vdev,
-				     unsigned index,
-				     void (*callback)(struct virtqueue *));
-	void (*del_vq)(struct virtqueue *vq);
+	int (*find_vqs)(struct virtio_device *, unsigned nvqs,
+			struct virtqueue *vqs[],
+			vq_callback_t *callbacks[],
+			const char *names[]);
+	void (*del_vqs)(struct virtio_device *);
 	u32 (*get_features)(struct virtio_device *vdev);
 	void (*finalize_features)(struct virtio_device *vdev);
 };
@@ -99,7 +113,9 @@ static inline bool virtio_has_feature(const struct virtio_device *vdev,
 	if (__builtin_constant_p(fbit))
 		BUILD_BUG_ON(fbit >= 32);
 
+	if (fbit < VIRTIO_TRANSPORT_F_START)
 		virtio_check_driver_offered_feature(vdev, fbit);
+
 	return test_bit(fbit, vdev->features);
 }
 
@@ -126,5 +142,18 @@ static inline int virtio_config_buf(struct virtio_device *vdev,
 	vdev->config->get(vdev, offset, buf, len);
 	return 0;
 }
+
+static inline
+struct virtqueue *virtio_find_single_vq(struct virtio_device *vdev,
+					vq_callback_t *c, const char *n)
+{
+	vq_callback_t *callbacks[] = { c };
+	const char *names[] = { n };
+	struct virtqueue *vq;
+	int err = vdev->config->find_vqs(vdev, 1, &vq, callbacks, names);
+	if (err < 0)
+		return ERR_PTR(err);
+	return vq;
+}
 #endif /* __KERNEL__ */
 #endif /* _LINUX_VIRTIO_CONFIG_H */

include/linux/virtio_pci.h

@@ -47,9 +47,17 @@
 /* The bit of the ISR which indicates a device configuration change. */
 #define VIRTIO_PCI_ISR_CONFIG		0x2
 
+/* MSI-X registers: only enabled if MSI-X is enabled. */
+/* A 16-bit vector for configuration changes. */
+#define VIRTIO_MSI_CONFIG_VECTOR        20
+/* A 16-bit vector for selected queue notifications. */
+#define VIRTIO_MSI_QUEUE_VECTOR         22
+/* Vector value used to disable MSI for queue */
+#define VIRTIO_MSI_NO_VECTOR            0xffff
+
 /* The remaining space is defined by each driver as the per-driver
  * configuration space */
-#define VIRTIO_PCI_CONFIG		20
+#define VIRTIO_PCI_CONFIG(dev)		((dev)->msix_enabled ? 24 : 20)
 
 /* Virtio ABI version, this must match exactly */
 #define VIRTIO_PCI_ABI_VERSION		0

include/linux/virtio_ring.h

@@ -14,6 +14,8 @@
 #define VRING_DESC_F_NEXT	1
 /* This marks a buffer as write-only (otherwise read-only). */
 #define VRING_DESC_F_WRITE	2
+/* This means the buffer contains a list of buffer descriptors. */
+#define VRING_DESC_F_INDIRECT	4
 
 /* The Host uses this in used->flags to advise the Guest: don't kick me when
  * you add a buffer.  It's unreliable, so it's simply an optimization.  Guest
@@ -24,6 +26,9 @@
  * optimization.  */
 #define VRING_AVAIL_F_NO_INTERRUPT	1
 
+/* We support indirect buffer descriptors */
+#define VIRTIO_RING_F_INDIRECT_DESC	28
+
 /* Virtio ring descriptors: 16 bytes.  These can chain together via "next". */
 struct vring_desc
 {
@@ -119,7 +124,8 @@ struct virtqueue *vring_new_virtqueue(unsigned int num,
 				      struct virtio_device *vdev,
 				      void *pages,
 				      void (*notify)(struct virtqueue *vq),
-				      void (*callback)(struct virtqueue *vq));
+				      void (*callback)(struct virtqueue *vq),
+				      const char *name);
 void vring_del_virtqueue(struct virtqueue *vq);
 /* Filter out transport-specific feature bits. */
 void vring_transport_features(struct virtio_device *vdev);

net/9p/trans_virtio.c

@@ -246,7 +246,7 @@ static int p9_virtio_probe(struct virtio_device *vdev)
 	chan->vdev = vdev;
 
 	/* We expect one virtqueue, for requests. */
-	chan->vq = vdev->config->find_vq(vdev, 0, req_done);
+	chan->vq = virtio_find_single_vq(vdev, req_done, "requests");
 	if (IS_ERR(chan->vq)) {
 		err = PTR_ERR(chan->vq);
 		goto out_free_vq;
@@ -261,7 +261,7 @@ static int p9_virtio_probe(struct virtio_device *vdev)
 	return 0;
 
 out_free_vq:
-	vdev->config->del_vq(chan->vq);
+	vdev->config->del_vqs(vdev);
 fail:
 	mutex_lock(&virtio_9p_lock);
 	chan_index--;
@@ -332,7 +332,7 @@ static void p9_virtio_remove(struct virtio_device *vdev)
 	BUG_ON(chan->inuse);
 
 	if (chan->initialized) {
-		vdev->config->del_vq(chan->vq);
+		vdev->config->del_vqs(vdev);
 		chan->initialized = false;
 	}
 }

scripts/mod/file2alias.c

@@ -641,7 +641,7 @@ static int do_virtio_entry(const char *filename, struct virtio_device_id *id,
 	id->vendor = TO_NATIVE(id->vendor);
 
 	strcpy(alias, "virtio:");
-	ADD(alias, "d", 1, id->device);
+	ADD(alias, "d", id->device != VIRTIO_DEV_ANY_ID, id->device);
 	ADD(alias, "v", id->vendor != VIRTIO_DEV_ANY_ID, id->vendor);
 
 	add_wildcard(alias);