Re: [dpdk-dev] [PATCH 2/2] vhost: support thread-safe API for clearing in-flight packets in async vhost

["Xia, Chenbo" <chenbo.xia@intel.com>]

Hi Yuan,

> -----Original Message-----
> From: Wang, YuanX <yuanx.wang@intel.com>
> Sent: Thursday, September 9, 2021 2:58 PM
> To: dev@dpdk.org
> Cc: maxime.coquelin@redhat.com; Xia, Chenbo <chenbo.xia@intel.com>; Pai G,
> Sunil <sunil.pai.g@intel.com>; Hu, Jiayu <jiayu.hu@intel.com>; Ding, Xuan
> <xuan.ding@intel.com>; Jiang, Cheng1 <cheng1.jiang@intel.com>; Ma, WenwuX
> <wenwux.ma@intel.com>; Yang, YvonneX <yvonnex.yang@intel.com>; Wang, YuanX
> <yuanx.wang@intel.com>
> Subject: [PATCH 2/2] vhost: support thread-safe API for clearing in-flight
> packets in async vhost

support -> add

> 
> This patch adds thread-safe version for
> clearing in-flight packets function.
> 
> Signed-off-by: Yuan Wang <yuanx.wang@intel.com>
> ---
>  lib/vhost/rte_vhost_async.h | 21 +++++++++++++++++++++
>  lib/vhost/version.map       |  1 +
>  lib/vhost/virtio_net.c      | 36 ++++++++++++++++++++++++++++++++++++
>  3 files changed, 58 insertions(+)

Miss update of release note.

> 
> diff --git a/lib/vhost/rte_vhost_async.h b/lib/vhost/rte_vhost_async.h
> index 5e2429ab70..a418e0a03d 100644
> --- a/lib/vhost/rte_vhost_async.h
> +++ b/lib/vhost/rte_vhost_async.h
> @@ -261,6 +261,27 @@ int rte_vhost_async_get_inflight(int vid, uint16_t
> queue_id);
>  __rte_experimental
>  uint16_t rte_vhost_clear_queue_thread_unsafe(int vid, uint16_t queue_id,
>  		struct rte_mbuf **pkts, uint16_t count);
> +
> +/**
> + * This function checks async completion status and clear packets for
> + * a specific vhost device queue. Packets which are inflight will be
> + * returned in an array.
> + *
> + * @param vid
> + *  ID of vhost device to clear data
> + * @param queue_id
> + *  Queue id to clear data
> + * @param pkts
> + *  Blank array to get return packet pointer
> + * @param count
> + *  Size of the packet array
> + * @return
> + *  Number of packets returned
> + */
> +__rte_experimental
> +uint16_t rte_vhost_clear_queue(int vid, uint16_t queue_id,
> +		struct rte_mbuf **pkts, uint16_t count);
> +
>  /**
>   * This function tries to receive packets from the guest with offloading
>   * copies to the async channel. The packets that are transfer completed
> diff --git a/lib/vhost/version.map b/lib/vhost/version.map
> index 3d566a6d5f..f78cc89b58 100644
> --- a/lib/vhost/version.map
> +++ b/lib/vhost/version.map
> @@ -88,4 +88,5 @@ EXPERIMENTAL {
> 
>  	# added in 21.11
>  	rte_vhost_async_try_dequeue_burst;
> +	rte_vhost_clear_queue;
>  };
> diff --git a/lib/vhost/virtio_net.c b/lib/vhost/virtio_net.c
> index 7f6183a929..51693a7c35 100644
> --- a/lib/vhost/virtio_net.c
> +++ b/lib/vhost/virtio_net.c
> @@ -2142,6 +2142,42 @@ rte_vhost_clear_queue_thread_unsafe(int vid, uint16_t
> queue_id,
>  	return n_pkts_cpl;
>  }
> 
> +uint16_t
> +rte_vhost_clear_queue(int vid, uint16_t queue_id, struct rte_mbuf **pkts,
> uint16_t count)
> +{
> +	struct virtio_net *dev = get_device(vid);
> +	struct vhost_virtqueue *vq;
> +	uint16_t n_pkts_cpl;
> +
> +	if (!dev)
> +		return 0;
> +
> +	VHOST_LOG_DATA(DEBUG, "(%d) %s\n", dev->vid, __func__);
> +

Should check queue id here.

> +	vq = dev->virtqueue[queue_id];
> +
> +	if (unlikely(!vq->async_registered)) {
> +		VHOST_LOG_DATA(ERR, "(%d) %s: async not registered for queue
> id %d.\n",
> +			dev->vid, __func__, queue_id);
> +		return 0;
> +	}
> +
> +	if (!rte_spinlock_trylock(&vq->access_lock)) {
> +		VHOST_LOG_CONFIG(ERR, "Failed to clear async queue, virt queue
> busy.\n");

Should be VHOST_LOG_DATA. And please add vid and qid info in the log.

> +		return 0;
> +	}
> +
> +	if ((queue_id % 2) == 0)

You can remove internal '()'

> +		n_pkts_cpl = vhost_poll_enqueue_completed(dev, queue_id, pkts,
> count);
> +	else
> +		n_pkts_cpl = async_poll_dequeue_completed_split(dev, vq, queue_id,
> pkts, count,

Add check to make sure it's split queue.

Thanks,
Chenbo

> +						dev->flags & VIRTIO_DEV_LEGACY_OL_FLAGS);
> +
> +	rte_spinlock_unlock(&vq->access_lock);
> +
> +	return n_pkts_cpl;
> +}
> +
>  static __rte_always_inline uint32_t
>  virtio_dev_rx_async_submit(struct virtio_net *dev, uint16_t queue_id,
>  	struct rte_mbuf **pkts, uint32_t count)
> --
> 2.25.1