[PATCH v5 00/22] virtio pci support VIRTIO_F_RING_RESET

@@ -166,6 +166,13 @@ struct virtio_pci_common_cfg {
 	__le32 queue_used_hi;		/* read-write */
 };
 
+struct virtio_pci_common_cfg_notify {
+	struct virtio_pci_common_cfg cfg;
+
+	__le16 queue_notify_data;	/* read-write */
+	__le16 padding;
+};
+
 /* Fields in VIRTIO_PCI_CAP_PCI_CFG: */
 struct virtio_pci_cfg_cap {
 	struct virtio_pci_cap cap;

@@ -2355,12 +2355,10 @@ struct virtqueue *vring_new_virtqueue(unsigned int index,
 }
 EXPORT_SYMBOL_GPL(vring_new_virtqueue);
 
-void vring_del_virtqueue(struct virtqueue *_vq)
+static void __vring_del_virtqueue(struct vring_virtqueue *vq)
 {
-	struct vring_virtqueue *vq = to_vvq(_vq);
-
 	spin_lock(&vq->vq.vdev->vqs_list_lock);
-	list_del(&_vq->list);
+	list_del(&vq->vq.list);
 	spin_unlock(&vq->vq.vdev->vqs_list_lock);
 
 	if (vq->we_own_ring) {

@@ -2393,6 +2391,13 @@ void vring_del_virtqueue(struct virtqueue *_vq)
 		kfree(vq->split.desc_state);
 		kfree(vq->split.desc_extra);
 	}
+}
+
+void vring_del_virtqueue(struct virtqueue *_vq)
+{
+	struct vring_virtqueue *vq = to_vvq(_vq);
+
+	__vring_del_virtqueue(vq);
 	kfree(vq);
 }
 EXPORT_SYMBOL_GPL(vring_del_virtqueue);

@@ -1730,6 +1730,7 @@ static struct virtqueue *vring_create_virtqueue_packed(
 	vq->vq.vdev = vdev;
 	vq->vq.num_free = num;
 	vq->vq.index = index;
+	vq->vq.reset = VIRTQUEUE_RESET_STAGE_NONE;
 	vq->we_own_ring = true;
 	vq->notify = notify;
 	vq->weak_barriers = weak_barriers;

@@ -2218,6 +2219,7 @@ static int __vring_init_virtqueue(struct virtqueue *_vq,
 	vq->vq.vdev = vdev;
 	vq->vq.num_free = vring.num;
 	vq->vq.index = index;
+	vq->vq.reset = VIRTQUEUE_RESET_STAGE_NONE;
 	vq->we_own_ring = false;
 	vq->notify = notify;
 	vq->weak_barriers = weak_barriers;

@@ -2397,11 +2399,25 @@ void vring_del_virtqueue(struct virtqueue *_vq)
 {
 	struct vring_virtqueue *vq = to_vvq(_vq);
 
-	__vring_del_virtqueue(vq);
+	if (_vq->reset != VIRTQUEUE_RESET_STAGE_RELEASE)
+		__vring_del_virtqueue(vq);
 	kfree(vq);
 }
 EXPORT_SYMBOL_GPL(vring_del_virtqueue);
 
+void vring_release_virtqueue(struct virtqueue *_vq)
+{
+	struct vring_virtqueue *vq = to_vvq(_vq);
+
+	if (_vq->reset != VIRTQUEUE_RESET_STAGE_DEVICE)
+		return;
+
+	__vring_del_virtqueue(vq);
+
+	_vq->reset = VIRTQUEUE_RESET_STAGE_RELEASE;
+}
+EXPORT_SYMBOL_GPL(vring_release_virtqueue);
+
 /* Manipulates transport-specific feature bits. */
 void vring_transport_features(struct virtio_device *vdev)
 {

@@ -10,6 +10,12 @@
 #include <linux/mod_devicetable.h>
 #include <linux/gfp.h>
 
+enum virtqueue_reset_stage {
+	VIRTQUEUE_RESET_STAGE_NONE,
+	VIRTQUEUE_RESET_STAGE_DEVICE,
+	VIRTQUEUE_RESET_STAGE_RELEASE,
+};
+
 /**
  * virtqueue - a queue to register buffers for sending or receiving.
  * @list: the chain of virtqueues for this device

@@ -32,6 +38,7 @@ struct virtqueue {
 	unsigned int index;
 	unsigned int num_free;
 	void *priv;
+	enum virtqueue_reset_stage reset;
 };
 
 int virtqueue_add_outbuf(struct virtqueue *vq,

@@ -196,4 +203,9 @@ void unregister_virtio_driver(struct virtio_driver *drv);
 #define module_virtio_driver(__virtio_driver) \
 	module_driver(__virtio_driver, register_virtio_driver, \
 			unregister_virtio_driver)
+/*
+ * Resets a virtqueue. Just frees the ring, not free vq.
+ * This function must be called after reset_vq().
+ */
+void vring_release_virtqueue(struct virtqueue *vq);
 #endif /* _LINUX_VIRTIO_H */

@@ -74,6 +74,18 @@ struct virtio_shm_region {
  * @set_vq_affinity: set the affinity for a virtqueue (optional).
  * @get_vq_affinity: get the affinity for a virtqueue (optional).
  * @get_shm_region: get a shared memory region based on the index.
+ * @reset_vq: reset a queue individually (optional).
+ *	vq: the virtqueue
+ *	Returns 0 on success or error status
+ *	After successfully calling this, be sure to call
+ *	virtqueue_detach_unused_buf() to recycle the buffer in the ring, and
+ *	then call vring_release_virtqueue() to release the vq ring.
+ *	Caller should guarantee that the vring is not accessed by any functions
+ *	of virtqueue.
+ * @enable_reset_vq: enable a reset queue
+ *	vq: the virtqueue
+ *	Returns 0 on success or error status
+ *	If reset_vq is set, then enable_reset_vq must also be set.
  */
 typedef void vq_callback_t(struct virtqueue *);
 struct virtio_config_ops {

@@ -100,6 +112,8 @@ struct virtio_config_ops {
 			int index);
 	bool (*get_shm_region)(struct virtio_device *vdev,
 			       struct virtio_shm_region *region, u8 id);
+	int (*reset_vq)(struct virtqueue *vq);
+	int (*enable_reset_vq)(struct virtqueue *vq);
 };
 
 /* If driver didn't advertise the feature, it will never appear. */

@@ -52,7 +52,7 @@
  * rest are per-device feature bits.
  */
 #define VIRTIO_TRANSPORT_F_START	28
-#define VIRTIO_TRANSPORT_F_END		38
+#define VIRTIO_TRANSPORT_F_END		41
 
 #ifndef VIRTIO_CONFIG_NO_LEGACY
 /* Do we get callbacks when the ring is completely used, even if we've

@@ -92,4 +92,9 @@
  * Does the device support Single Root I/O Virtualization?
  */
 #define VIRTIO_F_SR_IOV			37
+
+/*
+ * This feature indicates that the driver can reset a queue individually.
+ */
+#define VIRTIO_F_RING_RESET		40
 #endif /* _UAPI_LINUX_VIRTIO_CONFIG_H */

@@ -2169,23 +2169,17 @@ irqreturn_t vring_interrupt(int irq, void *_vq)
 EXPORT_SYMBOL_GPL(vring_interrupt);
 
 /* Only available for split ring */
-struct virtqueue *__vring_new_virtqueue(unsigned int index,
-					struct vring vring,
-					struct virtio_device *vdev,
-					bool weak_barriers,
-					bool context,
-					bool (*notify)(struct virtqueue *),
-					void (*callback)(struct virtqueue *),
-					const char *name)
+static int __vring_init_virtqueue(struct virtqueue *_vq,
+				  unsigned int index,
+				  struct vring vring,
+				  struct virtio_device *vdev,
+				  bool weak_barriers,
+				  bool context,
+				  bool (*notify)(struct virtqueue *),
+				  void (*callback)(struct virtqueue *),
+				  const char *name)
 {
-	struct vring_virtqueue *vq;
-
-	if (virtio_has_feature(vdev, VIRTIO_F_RING_PACKED))
-		return NULL;
-
-	vq = kmalloc(sizeof(*vq), GFP_KERNEL);
-	if (!vq)
-		return NULL;
+	struct vring_virtqueue *vq = to_vvq(_vq);
 
 	vq->packed_ring = false;
 	vq->vq.callback = callback;

@@ -2245,13 +2239,42 @@ struct virtqueue *__vring_new_virtqueue(unsigned int index,
 	spin_lock(&vdev->vqs_list_lock);
 	list_add_tail(&vq->vq.list, &vdev->vqs);
 	spin_unlock(&vdev->vqs_list_lock);
-	return &vq->vq;
+	return 0;
 
 err_extra:
 	kfree(vq->split.desc_state);
 err_state:
-	kfree(vq);
-	return NULL;
+	return -ENOMEM;
+}
+
+struct virtqueue *__vring_new_virtqueue(unsigned int index,
+					struct vring vring,
+					struct virtio_device *vdev,
+					bool weak_barriers,
+					bool context,
+					bool (*notify)(struct virtqueue *),
+					void (*callback)(struct virtqueue *),
+					const char *name)
+{
+	struct vring_virtqueue *vq;
+	int err;
+
+	if (virtio_has_feature(vdev, VIRTIO_F_RING_PACKED))
+		return NULL;
+
+	vq = kmalloc(sizeof(*vq), GFP_KERNEL);
+	if (!vq)
+		return NULL;
+
+	err = __vring_init_virtqueue(&vq->vq, index, vring, vdev, weak_barriers,
+				     context, notify, callback, name);
+
+	if (err) {
+		kfree(vq);
+		return NULL;
+	}
+
+	return &vq->vq;
 }
 EXPORT_SYMBOL_GPL(__vring_new_virtqueue);
 

@@ -2255,7 +2255,7 @@ struct virtqueue *__vring_new_virtqueue(unsigned int index,
 }
 EXPORT_SYMBOL_GPL(__vring_new_virtqueue);
 
-struct virtqueue *vring_create_virtqueue(
+struct virtqueue *vring_setup_virtqueue(
 	unsigned int index,
 	unsigned int num,
 	unsigned int vring_align,

@@ -2265,7 +2265,8 @@ struct virtqueue *vring_create_virtqueue(
 	bool context,
 	bool (*notify)(struct virtqueue *),
 	void (*callback)(struct virtqueue *),
-	const char *name)
+	const char *name,
+	struct virtqueue *vq)
 {
 
 	if (virtio_has_feature(vdev, VIRTIO_F_RING_PACKED))

@@ -2277,7 +2278,7 @@ struct virtqueue *vring_create_virtqueue(
 			vdev, weak_barriers, may_reduce_num,
 			context, notify, callback, name);
 }
-EXPORT_SYMBOL_GPL(vring_create_virtqueue);
+EXPORT_SYMBOL_GPL(vring_setup_virtqueue);
 
 /* Only available for split ring */
 struct virtqueue *vring_new_virtqueue(unsigned int index,

@@ -65,16 +65,33 @@ struct virtqueue;
  * expected.  The caller should query virtqueue_get_vring_size to learn
  * the actual size of the ring.
  */
-struct virtqueue *vring_create_virtqueue(unsigned int index,
-					 unsigned int num,
-					 unsigned int vring_align,
-					 struct virtio_device *vdev,
-					 bool weak_barriers,
-					 bool may_reduce_num,
-					 bool ctx,
-					 bool (*notify)(struct virtqueue *vq),
-					 void (*callback)(struct virtqueue *vq),
-					 const char *name);
+struct virtqueue *vring_setup_virtqueue(unsigned int index,
+					unsigned int num,
+					unsigned int vring_align,
+					struct virtio_device *vdev,
+					bool weak_barriers,
+					bool may_reduce_num,
+					bool ctx,
+					bool (*notify)(struct virtqueue *vq),
+					void (*callback)(struct virtqueue *vq),
+					const char *name,
+					struct virtqueue *vq);
+
+static inline struct virtqueue *vring_create_virtqueue(unsigned int index,
+						       unsigned int num,
+						       unsigned int vring_align,
+						       struct virtio_device *vdev,
+						       bool weak_barriers,
+						       bool may_reduce_num,
+						       bool ctx,
+						       bool (*notify)(struct virtqueue *vq),
+						       void (*callback)(struct virtqueue *vq),
+						       const char *name)
+{
+	return vring_setup_virtqueue(index, num, vring_align, vdev,
+				     weak_barriers, may_reduce_num, ctx,
+				     notify, callback, name, NULL);
+}
 
 /* Creates a virtqueue with a custom layout. */
 struct virtqueue *__vring_new_virtqueue(unsigned int index,

@@ -224,10 +224,12 @@ static struct virtqueue *setup_vq(struct virtio_pci_device *vp_dev,
 				virtqueue_get_avail_addr(vq),
 				virtqueue_get_used_addr(vq));
 
-	vq->priv = (void __force *)vp_modern_map_vq_notify(mdev, index, NULL);
 	if (!vq->priv) {
-		err = -ENOMEM;
-		goto err_map_notify;
+		vq->priv = (void __force *)vp_modern_map_vq_notify(mdev, index, NULL);
+		if (!vq->priv) {
+			err = -ENOMEM;
+			goto err_map_notify;
+		}
 	}
 
 	if (msix_vec != VIRTIO_MSI_NO_VECTOR) {

@@ -1723,6 +1723,7 @@ static struct virtqueue *vring_create_virtqueue_packed(
 
 		vq->vq.callback = callback;
 		vq->vq.name = name;
+		vq->vq.priv = NULL;
 		vq->indirect = virtio_has_feature(vdev, VIRTIO_RING_F_INDIRECT_DESC) &&
 			!context;
 	}

@@ -2211,6 +2212,7 @@ static int __vring_init_virtqueue(struct virtqueue *_vq,
 	if (!reset) {
 		vq->vq.callback = callback;
 		vq->vq.name = name;
+		vq->vq.priv = NULL;
 		vq->indirect = virtio_has_feature(vdev, VIRTIO_RING_F_INDIRECT_DESC) &&
 			!context;
 	}

@@ -255,9 +255,11 @@ static void vp_del_vq(struct virtqueue *vq)
 	struct virtio_pci_vq_info *info = vp_dev->vqs[vq->index];
 	unsigned long flags;
 
-	spin_lock_irqsave(&vp_dev->lock, flags);
-	list_del(&info->node);
-	spin_unlock_irqrestore(&vp_dev->lock, flags);
+	if (!vq->reset) {
+		spin_lock_irqsave(&vp_dev->lock, flags);
+		list_del(&info->node);
+		spin_unlock_irqrestore(&vp_dev->lock, flags);
+	}
 
 	vp_dev->del_vq(info);
 	kfree(info);

@@ -34,6 +34,9 @@ static void vp_transport_features(struct virtio_device *vdev, u64 features)
 	if ((features & BIT_ULL(VIRTIO_F_SR_IOV)) &&
 			pci_find_ext_capability(pci_dev, PCI_EXT_CAP_ID_SRIOV))
 		__virtio_set_bit(vdev, VIRTIO_F_SR_IOV);
+
+	if (features & BIT_ULL(VIRTIO_F_RING_RESET))
+		__virtio_set_bit(vdev, VIRTIO_F_RING_RESET);
 }
 
 /* virtio config->finalize_features() implementation */

@@ -176,6 +179,59 @@ static void vp_reset(struct virtio_device *vdev)
 	vp_disable_cbs(vdev);
 }
 
+static int vp_modern_reset_vq(struct virtqueue *vq)
+{
+	struct virtio_pci_device *vp_dev = to_vp_device(vq->vdev);
+	struct virtio_pci_modern_device *mdev = &vp_dev->mdev;
+	struct virtio_pci_vq_info *info;
+	unsigned long flags;
+
+	if (!virtio_has_feature(vq->vdev, VIRTIO_F_RING_RESET))
+		return -ENOENT;
+
+	vp_modern_set_queue_reset(mdev, vq->index);
+
+	info = vp_dev->vqs[vq->index];
+
+	/* delete vq from irq handler */
+	spin_lock_irqsave(&vp_dev->lock, flags);
+	list_del(&info->node);
+	spin_unlock_irqrestore(&vp_dev->lock, flags);
+
+	INIT_LIST_HEAD(&info->node);
+
+	vq->reset = VIRTQUEUE_RESET_STAGE_DEVICE;
+
+	return 0;
+}
+
+static int vp_modern_enable_reset_vq(struct virtqueue *vq)
+{
+	struct virtio_pci_device *vp_dev = to_vp_device(vq->vdev);
+	struct virtio_pci_modern_device *mdev = &vp_dev->mdev;
+	struct virtio_pci_vq_info *info;
+	struct virtqueue *_vq;
+
+	if (vq->reset != VIRTQUEUE_RESET_STAGE_RELEASE)
+		return -EBUSY;
+
+	/* check queue reset status */
+	if (vp_modern_get_queue_reset(mdev, vq->index) != 1)
+		return -EBUSY;
+
+	info = vp_dev->vqs[vq->index];
+	_vq = vp_setup_vq(vq->vdev, vq->index, NULL, NULL, NULL,
+			 info->msix_vector);
+	if (IS_ERR(_vq)) {
+		vq->reset = VIRTQUEUE_RESET_STAGE_RELEASE;
+		return PTR_ERR(_vq);
+	}
+
+	vp_modern_set_queue_enable(&vp_dev->mdev, vq->index, true);
+
+	return 0;
+}
+
 static u16 vp_config_vector(struct virtio_pci_device *vp_dev, u16 vector)
 {
 	return vp_modern_config_vector(&vp_dev->mdev, vector);

@@ -397,6 +453,8 @@ static const struct virtio_config_ops virtio_pci_config_nodev_ops = {
 	.set_vq_affinity = vp_set_vq_affinity,
 	.get_vq_affinity = vp_get_vq_affinity,
 	.get_shm_region  = vp_get_shm_region,
+	.reset_vq	 = vp_modern_reset_vq,
+	.enable_reset_vq = vp_modern_enable_reset_vq,
 };
 
 static const struct virtio_config_ops virtio_pci_config_ops = {

@@ -415,6 +473,8 @@ static const struct virtio_config_ops virtio_pci_config_ops = {
 	.set_vq_affinity = vp_set_vq_affinity,
 	.get_vq_affinity = vp_get_vq_affinity,
 	.get_shm_region  = vp_get_shm_region,
+	.reset_vq	 = vp_modern_reset_vq,
+	.enable_reset_vq = vp_modern_enable_reset_vq,
 };
 
 /* the PCI probing function */

@@ -198,6 +198,16 @@ struct vring_virtqueue {
 #endif
 };
 
+static int __vring_init_virtqueue(struct virtqueue *_vq,
+				  unsigned int index,
+				  struct vring vring,
+				  struct virtio_device *vdev,
+				  bool weak_barriers,
+				  bool context,
+				  bool (*notify)(struct virtqueue *),
+				  void (*callback)(struct virtqueue *),
+				  const char *name,
+				  bool reset);
 
 /*
  * Helpers.

@@ -925,9 +935,9 @@ static struct virtqueue *vring_create_virtqueue_split(
 	bool context,
 	bool (*notify)(struct virtqueue *),
 	void (*callback)(struct virtqueue *),
-	const char *name)
+	const char *name,
+	struct virtqueue *vq)
 {
-	struct virtqueue *vq;
 	void *queue = NULL;
 	dma_addr_t dma_addr;
 	size_t queue_size_in_bytes;

@@ -964,12 +974,17 @@ static struct virtqueue *vring_create_virtqueue_split(
 	queue_size_in_bytes = vring_size(num, vring_align);
 	vring_init(&vring, num, queue, vring_align);
 
-	vq = __vring_new_virtqueue(index, vring, vdev, weak_barriers, context,
-				   notify, callback, name);
 	if (!vq) {
-		vring_free_queue(vdev, queue_size_in_bytes, queue,
-				 dma_addr);
-		return NULL;
+		vq = __vring_new_virtqueue(index, vring, vdev, weak_barriers,
+					   context, notify, callback, name);
+		if (!vq)
+			goto err;
+
+	} else {
+		if (__vring_init_virtqueue(vq, index, vring, vdev,
+					   weak_barriers, context, notify,
+					   callback, name, true))
+			goto err;
 	}
 
 	to_vvq(vq)->split.queue_dma_addr = dma_addr;

@@ -977,6 +992,9 @@ static struct virtqueue *vring_create_virtqueue_split(
 	to_vvq(vq)->we_own_ring = true;
 
 	return vq;
+err:
+	vring_free_queue(vdev, queue_size_in_bytes, queue, dma_addr);
+	return NULL;
 }
 
 

@@ -2177,14 +2195,20 @@ static int __vring_init_virtqueue(struct virtqueue *_vq,
 				  bool context,
 				  bool (*notify)(struct virtqueue *),
 				  void (*callback)(struct virtqueue *),
-				  const char *name)
+				  const char *name,
+				  bool reset)
 {
 	struct vring_virtqueue *vq = to_vvq(_vq);
 
+	if (!reset) {
+		vq->vq.callback = callback;
+		vq->vq.name = name;
+		vq->indirect = virtio_has_feature(vdev, VIRTIO_RING_F_INDIRECT_DESC) &&
+			!context;
+	}
+
 	vq->packed_ring = false;
-	vq->vq.callback = callback;
 	vq->vq.vdev = vdev;
-	vq->vq.name = name;
 	vq->vq.num_free = vring.num;
 	vq->vq.index = index;
 	vq->we_own_ring = false;

@@ -2200,8 +2224,6 @@ static int __vring_init_virtqueue(struct virtqueue *_vq,
 	vq->last_add_time_valid = false;
 #endif
 
-	vq->indirect = virtio_has_feature(vdev, VIRTIO_RING_F_INDIRECT_DESC) &&
-		!context;
 	vq->event = virtio_has_feature(vdev, VIRTIO_RING_F_EVENT_IDX);
 
 	if (virtio_has_feature(vdev, VIRTIO_F_ORDER_PLATFORM))

@@ -2215,7 +2237,7 @@ static int __vring_init_virtqueue(struct virtqueue *_vq,
 	vq->split.avail_idx_shadow = 0;
 
 	/* No callback?  Tell other side not to bother us. */
-	if (!callback) {
+	if (!vq->vq.callback) {
 		vq->split.avail_flags_shadow |= VRING_AVAIL_F_NO_INTERRUPT;
 		if (!vq->event)
 			vq->split.vring.avail->flags = cpu_to_virtio16(vdev,

@@ -2267,7 +2289,7 @@ struct virtqueue *__vring_new_virtqueue(unsigned int index,
 		return NULL;
 
 	err = __vring_init_virtqueue(&vq->vq, index, vring, vdev, weak_barriers,
-				     context, notify, callback, name);
+				     context, notify, callback, name, false);
 
 	if (err) {
 		kfree(vq);

@@ -2299,7 +2321,7 @@ struct virtqueue *vring_setup_virtqueue(
 
 	return vring_create_virtqueue_split(index, num, vring_align,
 			vdev, weak_barriers, may_reduce_num,
-			context, notify, callback, name);
+			context, notify, callback, name, vq);
 }
 EXPORT_SYMBOL_GPL(vring_setup_virtqueue);
 

@@ -233,6 +233,44 @@ int virtio_find_vqs_ctx(struct virtio_device *vdev, unsigned nvqs,
 				      desc);
 }
 
+/**
+ * virtio_reset_vq - reset a queue individually
+ * @vq: the virtqueue
+ *
+ * returns 0 on success or error status
+ *
+ * After successfully calling this, be sure to call
+ * virtqueue_detach_unused_buf() to recycle the buffer in the ring, and
+ * then call vring_release_virtqueue() to release the vq ring.
+ *
+ * Caller should guarantee that the vring is not accessed by any functions
+ * of virtqueue.
+ */
+static inline
+int virtio_reset_vq(struct virtqueue *vq)
+{
+	if (!vq->vdev->config->reset_vq)
+		return -ENOENT;
+
+	return vq->vdev->config->reset_vq(vq);
+}
+
+/**
+ * virtio_enable_resetq - enable a reset queue
+ * @vq: the virtqueue
+ *
+ * returns 0 on success or error status
+ *
+ */
+static inline
+int virtio_enable_resetq(struct virtqueue *vq)
+{
+	if (!vq->vdev->config->enable_reset_vq)
+		return -ENOENT;
+
+	return vq->vdev->config->enable_reset_vq(vq);
+}
+
 /**
  * virtio_device_ready - enable vq use in probe function
  * @vdev: the device

@@ -1683,7 +1683,8 @@ static struct virtqueue *vring_create_virtqueue_packed(
 	bool context,
 	bool (*notify)(struct virtqueue *),
 	void (*callback)(struct virtqueue *),
-	const char *name)
+	const char *name,
+	struct virtqueue *_vq)
 {
 	struct vring_virtqueue *vq;
 	struct vring_packed_desc *ring;

@@ -1713,13 +1714,20 @@ static struct virtqueue *vring_create_virtqueue_packed(
 	if (!device)
 		goto err_device;
 
-	vq = kmalloc(sizeof(*vq), GFP_KERNEL);
-	if (!vq)
-		goto err_vq;
+	if (_vq) {
+		vq = to_vvq(_vq);
+	} else {
+		vq = kmalloc(sizeof(*vq), GFP_KERNEL);
+		if (!vq)
+			goto err_vq;
+
+		vq->vq.callback = callback;
+		vq->vq.name = name;
+		vq->indirect = virtio_has_feature(vdev, VIRTIO_RING_F_INDIRECT_DESC) &&
+			!context;
+	}
 
-	vq->vq.callback = callback;
 	vq->vq.vdev = vdev;
-	vq->vq.name = name;
 	vq->vq.num_free = num;
 	vq->vq.index = index;
 	vq->we_own_ring = true;

@@ -1736,8 +1744,6 @@ static struct virtqueue *vring_create_virtqueue_packed(
 	vq->last_add_time_valid = false;
 #endif
 
-	vq->indirect = virtio_has_feature(vdev, VIRTIO_RING_F_INDIRECT_DESC) &&
-		!context;
 	vq->event = virtio_has_feature(vdev, VIRTIO_RING_F_EVENT_IDX);
 
 	if (virtio_has_feature(vdev, VIRTIO_F_ORDER_PLATFORM))

@@ -1778,7 +1784,7 @@ static struct virtqueue *vring_create_virtqueue_packed(
 		goto err_desc_extra;
 
 	/* No callback?  Tell other side not to bother us. */
-	if (!callback) {
+	if (!vq->vq.callback) {
 		vq->packed.event_flags_shadow = VRING_PACKED_EVENT_FLAG_DISABLE;
 		vq->packed.vring.driver->flags =
 			cpu_to_le16(vq->packed.event_flags_shadow);

@@ -1792,7 +1798,8 @@ static struct virtqueue *vring_create_virtqueue_packed(
 err_desc_extra:
 	kfree(vq->packed.desc_state);
 err_desc_state:
-	kfree(vq);
+	if (!_vq)
+		kfree(vq);
 err_vq:
 	vring_free_queue(vdev, event_size_in_bytes, device, device_event_dma_addr);
 err_device:

@@ -2317,7 +2324,7 @@ struct virtqueue *vring_setup_virtqueue(
 	if (virtio_has_feature(vdev, VIRTIO_F_RING_PACKED))
 		return vring_create_virtqueue_packed(index, num, vring_align,
 				vdev, weak_barriers, may_reduce_num,
-				context, notify, callback, name);
+				context, notify, callback, name, vq);
 
 	return vring_create_virtqueue_split(index, num, vring_align,
 			vdev, weak_barriers, may_reduce_num,

@@ -2804,36 +2804,45 @@ static void free_receive_page_frags(struct virtnet_info *vi)
 			put_page(vi->rq[i].alloc_frag.page);
 }
 
-static void free_unused_bufs(struct virtnet_info *vi)
+static void virtnet_sq_free_unused_bufs(struct virtnet_info *vi,
+					struct send_queue *sq)
 {
 	void *buf;
-	int i;
 
-	for (i = 0; i < vi->max_queue_pairs; i++) {
-		struct virtqueue *vq = vi->sq[i].vq;
-		while ((buf = virtqueue_detach_unused_buf(vq)) != NULL) {
-			if (!is_xdp_frame(buf))
-				dev_kfree_skb(buf);
-			else
-				xdp_return_frame(ptr_to_xdp(buf));
-		}
+	while ((buf = virtqueue_detach_unused_buf(sq->vq)) != NULL) {
+		if (!is_xdp_frame(buf))
+			dev_kfree_skb(buf);
+		else
+			xdp_return_frame(ptr_to_xdp(buf));
 	}
+}
 
-	for (i = 0; i < vi->max_queue_pairs; i++) {
-		struct virtqueue *vq = vi->rq[i].vq;
-
-		while ((buf = virtqueue_detach_unused_buf(vq)) != NULL) {
-			if (vi->mergeable_rx_bufs) {
-				put_page(virt_to_head_page(buf));
-			} else if (vi->big_packets) {
-				give_pages(&vi->rq[i], buf);
-			} else {
-				put_page(virt_to_head_page(buf));
-			}
-		}
+static void virtnet_rq_free_unused_bufs(struct virtnet_info *vi,
+					struct receive_queue *rq)
+{
+	void *buf;
+
+	while ((buf = virtqueue_detach_unused_buf(rq->vq)) != NULL) {
+		if (vi->mergeable_rx_bufs)
+			put_page(virt_to_head_page(buf));
+		else if (vi->big_packets)
+			give_pages(rq, buf);
+		else
+			put_page(virt_to_head_page(buf));
 	}
 }
 
+static void free_unused_bufs(struct virtnet_info *vi)
+{
+	int i;
+
+	for (i = 0; i < vi->max_queue_pairs; i++)
+		virtnet_sq_free_unused_bufs(vi, vi->sq + i);
+
+	for (i = 0; i < vi->max_queue_pairs; i++)
+		virtnet_rq_free_unused_bufs(vi, vi->rq + i);
+}
+
 static void virtnet_del_vqs(struct virtnet_info *vi)
 {
 	struct virtio_device *vdev = vi->vdev;

@@ -3,6 +3,7 @@
 #include <linux/virtio_pci_modern.h>
 #include <linux/module.h>
 #include <linux/pci.h>
+#include <linux/delay.h>
 
 /*
  * vp_modern_map_capability - map a part of virtio pci capability

@@ -463,6 +464,41 @@ void vp_modern_set_status(struct virtio_pci_modern_device *mdev,
 }
 EXPORT_SYMBOL_GPL(vp_modern_set_status);
 
+/*
+ * vp_modern_get_queue_reset - get the queue reset status
+ * @mdev: the modern virtio-pci device
+ * @index: queue index
+ */
+int vp_modern_get_queue_reset(struct virtio_pci_modern_device *mdev, u16 index)
+{
+	struct virtio_pci_common_cfg_reset __iomem *cfg;
+
+	cfg = (struct virtio_pci_common_cfg_reset *)mdev->common;
+
+	vp_iowrite16(index, &cfg->cfg.queue_select);
+	return vp_ioread16(&cfg->queue_reset);
+}
+EXPORT_SYMBOL_GPL(vp_modern_get_queue_reset);
+
+/*
+ * vp_modern_set_queue_reset - reset the queue
+ * @mdev: the modern virtio-pci device
+ * @index: queue index
+ */
+void vp_modern_set_queue_reset(struct virtio_pci_modern_device *mdev, u16 index)
+{
+	struct virtio_pci_common_cfg_reset __iomem *cfg;
+
+	cfg = (struct virtio_pci_common_cfg_reset *)mdev->common;
+
+	vp_iowrite16(index, &cfg->cfg.queue_select);
+	vp_iowrite16(1, &cfg->queue_reset);
+
+	while (vp_ioread16(&cfg->queue_reset) != 1)
+		msleep(1);
+}
+EXPORT_SYMBOL_GPL(vp_modern_set_queue_reset);
+
 /*
  * vp_modern_queue_vector - set the MSIX vector for a specific virtqueue
  * @mdev: the modern virtio-pci device

@@ -106,4 +106,6 @@ void __iomem * vp_modern_map_vq_notify(struct virtio_pci_modern_device *mdev,
 				       u16 index, resource_size_t *pa);
 int vp_modern_probe(struct virtio_pci_modern_device *mdev);
 void vp_modern_remove(struct virtio_pci_modern_device *mdev);
+int vp_modern_get_queue_reset(struct virtio_pci_modern_device *mdev, u16 index);
+void vp_modern_set_queue_reset(struct virtio_pci_modern_device *mdev, u16 index);
 #endif

@@ -173,6 +173,13 @@ struct virtio_pci_common_cfg_notify {
 	__le16 padding;
 };
 
+struct virtio_pci_common_cfg_reset {
+	struct virtio_pci_common_cfg cfg;
+
+	__le16 queue_notify_data;	/* read-write */
+	__le16 queue_reset;		/* read-write */
+};
+
 /* Fields in VIRTIO_PCI_CAP_PCI_CFG: */
 struct virtio_pci_cfg_cap {
 	struct virtio_pci_cap cap;

@@ -943,6 +943,9 @@ static struct virtqueue *vring_create_virtqueue_split(
 	size_t queue_size_in_bytes;
 	struct vring vring;
 
+	if (vdev->max_ring_num && num > vdev->max_ring_num)
+		num = vdev->max_ring_num;
+
 	/* We assume num is a power of 2. */
 	if (num & (num - 1)) {
 		dev_warn(&vdev->dev, "Bad virtqueue length %u\n", num);

@@ -1692,6 +1695,9 @@ static struct virtqueue *vring_create_virtqueue_packed(
 	dma_addr_t ring_dma_addr, driver_event_dma_addr, device_event_dma_addr;
 	size_t ring_size_in_bytes, event_size_in_bytes;
 
+	if (vdev->max_ring_num && num > vdev->max_ring_num)
+		num = vdev->max_ring_num;
+
 	ring_size_in_bytes = num * sizeof(struct vring_packed_desc);
 
 	ring = vring_alloc_queue(vdev, ring_size_in_bytes,

@@ -127,6 +127,7 @@ struct virtio_device {
 	struct list_head vqs;
 	u64 features;
 	void *priv;
+	u16 max_ring_num;
 };
 
 static inline struct virtio_device *dev_to_virtio(struct device *_dev)

@@ -200,6 +200,36 @@ static inline bool virtio_has_dma_quirk(const struct virtio_device *vdev)
 	return !virtio_has_feature(vdev, VIRTIO_F_ACCESS_PLATFORM);
 }
 
+/**
+ * virtio_set_max_ring_num - set max ring num
+ * @vdev: the device
+ * @num: max ring num. Zero clear the limit.
+ *
+ * When creating a virtqueue, use this value as the upper limit of ring num.
+ *
+ * Returns 0 on success or error status
+ */
+static inline
+int virtio_set_max_ring_num(struct virtio_device *vdev, u16 num)
+{
+	if (!num) {
+		vdev->max_ring_num = num;
+		return 0;
+	}
+
+	if (!virtio_has_feature(vdev, VIRTIO_F_RING_PACKED)) {
+		if (!is_power_of_2(num)) {
+			num = __rounddown_pow_of_two(num);
+
+			if (!num)
+				return -EINVAL;
+		}
+	}
+
+	vdev->max_ring_num = num;
+	return 0;
+}
+
 static inline
 struct virtqueue *virtio_find_single_vq(struct virtio_device *vdev,
 					vq_callback_t *c, const char *n)

@@ -35,6 +35,8 @@ module_param(napi_tx, bool, 0644);
 #define GOOD_PACKET_LEN (ETH_HLEN + VLAN_HLEN + ETH_DATA_LEN)
 #define GOOD_COPY_LEN	128
 
+#define VIRTNET_DEFAULT_MAX_RING_NUM 1024
+
 #define VIRTNET_RX_PAD (NET_IP_ALIGN + NET_SKB_PAD)
 
 /* Amount of XDP headroom to prepend to packets for use by xdp_adjust_head */

@@ -3045,6 +3047,8 @@ static int virtnet_find_vqs(struct virtnet_info *vi)
 			ctx[rxq2vq(i)] = true;
 	}
 
+	virtio_set_max_ring_num(vi->vdev, VIRTNET_DEFAULT_MAX_RING_NUM);
+
 	ret = virtio_find_vqs_ctx(vi->vdev, total_vqs, vqs, callbacks,
 				  names, ctx, NULL);
 	if (ret)