Re: [PATCH v11 30/40] virtio_pci: support VIRTIO_F_RING_RESET

[Jason Wang <jasowang@redhat.com>]

在 2022/6/29 14:56, Xuan Zhuo 写道:
> This patch implements virtio pci support for QUEUE RESET.
>
> Performing reset on a queue is divided into these steps:
>
>   1. notify the device to reset the queue
>   2. recycle the buffer submitted
>   3. reset the vring (may re-alloc)
>   4. mmap vring to device, and enable the queue
>
> This patch implements virtio_reset_vq(), virtio_enable_resetq() in the
> pci scenario.
>
> Signed-off-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
> ---
>   drivers/virtio/virtio_pci_common.c | 12 +++-
>   drivers/virtio/virtio_pci_modern.c | 96 ++++++++++++++++++++++++++++++
>   drivers/virtio/virtio_ring.c       |  2 +
>   include/linux/virtio.h             |  1 +
>   4 files changed, 108 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/virtio/virtio_pci_common.c b/drivers/virtio/virtio_pci_common.c
> index ca51fcc9daab..ad258a9d3b9f 100644
> --- a/drivers/virtio/virtio_pci_common.c
> +++ b/drivers/virtio/virtio_pci_common.c
> @@ -214,9 +214,15 @@ 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 it fails during re-enable reset vq. This way we won't rejoin
> +	 * info->node to the queue. Prevent unexpected irqs.
> +	 */
> +	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);
> diff --git a/drivers/virtio/virtio_pci_modern.c b/drivers/virtio/virtio_pci_modern.c
> index 9041d9a41b7d..754e5e10386b 100644
> --- a/drivers/virtio/virtio_pci_modern.c
> +++ b/drivers/virtio/virtio_pci_modern.c
> @@ -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 */
> @@ -199,6 +202,95 @@ static int vp_active_vq(struct virtqueue *vq, u16 msix_vec)
>   	return 0;
>   }
>   
> +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);
> +
> +	/* For the case where vq has an exclusive irq, to prevent the irq from
> +	 * being received again and the pending irq, call synchronize_irq(), and
> +	 * break it.
> +	 *
> +	 * We can't use disable_irq() since it conflicts with the affinity
> +	 * managed IRQ that is used by some drivers. So this is done on top of
> +	 * IRQ hardening.
> +	 *
> +	 * In the scenario based on shared interrupts, vq will be searched from
> +	 * the queue virtqueues. Since the previous list_del() has been deleted
> +	 * from the queue, it is impossible for vq to be called in this case.
> +	 * There is no need to close the corresponding interrupt.
> +	 */
> +	if (vp_dev->per_vq_vectors && info->msix_vector != VIRTIO_MSI_NO_VECTOR) {
> +#ifdef CONFIG_VIRTIO_HARDEN_NOTIFICATION
> +		__virtqueue_break(vq);
> +#endif
> +		synchronize_irq(pci_irq_vector(vp_dev->pci_dev, info->msix_vector));
> +	}
> +
> +	vq->reset = true;
> +
> +	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;
> +	unsigned long flags, index;
> +	int err;
> +
> +	if (!vq->reset)
> +		return -EBUSY;
> +
> +	index = vq->index;
> +	info = vp_dev->vqs[index];
> +
> +	if (vp_modern_get_queue_reset(mdev, index))
> +		return -EBUSY;
> +
> +	if (vp_modern_get_queue_enable(mdev, index))
> +		return -EBUSY;
> +
> +	err = vp_active_vq(vq, info->msix_vector);
> +	if (err)
> +		return err;
> +
> +	if (vq->callback) {
> +		spin_lock_irqsave(&vp_dev->lock, flags);
> +		list_add(&info->node, &vp_dev->virtqueues);
> +		spin_unlock_irqrestore(&vp_dev->lock, flags);
> +	} else {
> +		INIT_LIST_HEAD(&info->node);
> +	}
> +
> +#ifdef CONFIG_VIRTIO_HARDEN_NOTIFICATION
> +	if (vp_dev->per_vq_vectors && info->msix_vector != VIRTIO_MSI_NO_VECTOR)
> +		__virtqueue_unbreak(vq);
> +#endif
> +
> +	vp_modern_set_queue_enable(&vp_dev->mdev, index, true);
> +	vq->reset = false;
> +
> +	return 0;
> +}
> +
>   static u16 vp_config_vector(struct virtio_pci_device *vp_dev, u16 vector)
>   {
>   	return vp_modern_config_vector(&vp_dev->mdev, vector);
> @@ -413,6 +505,8 @@ 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 = {
> @@ -431,6 +525,8 @@ 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 */
> diff --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.c
> index 7b02be7fce67..82b058e8ce34 100644
> --- a/drivers/virtio/virtio_ring.c
> +++ b/drivers/virtio/virtio_ring.c
> @@ -2008,6 +2008,7 @@ static struct virtqueue *vring_create_virtqueue_packed(
>   	vq->vq.vdev = vdev;
>   	vq->vq.name = name;
>   	vq->vq.index = index;
> +	vq->vq.reset = false;
>   	vq->we_own_ring = true;
>   	vq->notify = notify;
>   	vq->weak_barriers = weak_barriers;
> @@ -2487,6 +2488,7 @@ struct virtqueue *__vring_new_virtqueue(unsigned int index,
>   	vq->vq.vdev = vdev;
>   	vq->vq.name = name;
>   	vq->vq.index = index;
> +	vq->vq.reset = false;
>   	vq->we_own_ring = false;
>   	vq->notify = notify;
>   	vq->weak_barriers = weak_barriers;
> diff --git a/include/linux/virtio.h b/include/linux/virtio.h
> index dc474a0d48d1..88f21796b1c3 100644
> --- a/include/linux/virtio.h
> +++ b/include/linux/virtio.h
> @@ -33,6 +33,7 @@ struct virtqueue {
>   	unsigned int num_free;
>   	unsigned int num_max;
>   	void *priv;
> +	bool reset;
>   };


I wonder if it's better to move virtio_ring part into a separate patch.

Other looks good.

Thanks


>   
>   int virtqueue_add_outbuf(struct virtqueue *vq,