From: "Loftus, Ciara" <ciara.loftus@intel.com>
To: "Wang, Zhihong" <zhihong.wang@intel.com>, "dev@dpdk.org" <dev@dpdk.org>
Cc: "yuanhan.liu@linux.intel.com" <yuanhan.liu@linux.intel.com>,
"Wang, Zhihong" <zhihong.wang@intel.com>
Subject: Re: [dpdk-dev] [PATCH] vhost: support rx_queue_count
Date: Tue, 23 May 2017 13:10:07 +0000 [thread overview]
Message-ID: <74F120C019F4A64C9B78E802F6AD4CC24FA4D84F@IRSMSX106.ger.corp.intel.com> (raw)
In-Reply-To: <1495483268-60184-1-git-send-email-zhihong.wang@intel.com>
>
> This patch implements the ops rx_queue_count for vhost PMD by adding
> a helper function rte_vhost_rx_queue_count in vhost lib.
>
> The ops ops rx_queue_count gets vhost RX queue avail count and helps
> to understand the queue fill level.
>
> Signed-off-by: Zhihong Wang <zhihong.wang@intel.com>
> ---
> drivers/net/vhost/rte_eth_vhost.c | 13 +++++++++++++
> lib/librte_vhost/rte_vhost.h | 12 ++++++++++++
> lib/librte_vhost/rte_vhost_version.map | 7 +++++++
> lib/librte_vhost/vhost.c | 23 +++++++++++++++++++++++
> 4 files changed, 55 insertions(+)
>
> diff --git a/drivers/net/vhost/rte_eth_vhost.c
> b/drivers/net/vhost/rte_eth_vhost.c
> index 257bf6d..e3a3fe0 100644
> --- a/drivers/net/vhost/rte_eth_vhost.c
> +++ b/drivers/net/vhost/rte_eth_vhost.c
> @@ -973,6 +973,18 @@ eth_link_update(struct rte_eth_dev *dev
> __rte_unused,
> return 0;
> }
>
> +static uint32_t
> +eth_rx_queue_count(struct rte_eth_dev *dev, uint16_t rx_queue_id)
> +{
> + struct vhost_queue *vq;
> +
> + vq = dev->data->rx_queues[rx_queue_id];
> + if (!vq)
> + return 0;
> +
> + return rte_vhost_rx_queue_count(vq->vid, vq->virtqueue_id);
> +}
> +
> static const struct eth_dev_ops ops = {
> .dev_start = eth_dev_start,
> .dev_stop = eth_dev_stop,
> @@ -984,6 +996,7 @@ static const struct eth_dev_ops ops = {
> .rx_queue_release = eth_queue_release,
> .tx_queue_release = eth_queue_release,
> .tx_done_cleanup = eth_tx_done_cleanup,
> + .rx_queue_count = eth_rx_queue_count,
> .link_update = eth_link_update,
> .stats_get = eth_stats_get,
> .stats_reset = eth_stats_reset,
> diff --git a/lib/librte_vhost/rte_vhost.h b/lib/librte_vhost/rte_vhost.h
> index 605e47c..f64ed20 100644
> --- a/lib/librte_vhost/rte_vhost.h
> +++ b/lib/librte_vhost/rte_vhost.h
> @@ -432,6 +432,18 @@ int rte_vhost_get_mem_table(int vid, struct
> rte_vhost_memory **mem);
> int rte_vhost_get_vhost_vring(int vid, uint16_t vring_idx,
> struct rte_vhost_vring *vring);
>
> +/**
> + * Get vhost RX queue avail count.
> + *
> + * @param vid
> + * vhost device ID
> + * @param qid
> + * virtio queue index in mq case
> + * @return
> + * num of desc available
> + */
> +uint32_t rte_vhost_rx_queue_count(int vid, uint16_t qid);
> +
> #ifdef __cplusplus
> }
> #endif
> diff --git a/lib/librte_vhost/rte_vhost_version.map
> b/lib/librte_vhost/rte_vhost_version.map
> index 0785873..1e70495 100644
> --- a/lib/librte_vhost/rte_vhost_version.map
> +++ b/lib/librte_vhost/rte_vhost_version.map
> @@ -45,3 +45,10 @@ DPDK_17.05 {
> rte_vhost_log_write;
>
> } DPDK_16.07;
> +
> +DPDK_17.08 {
> + global:
> +
> + rte_vhost_rx_queue_count;
> +
> +} DPDK_17.05;
> diff --git a/lib/librte_vhost/vhost.c b/lib/librte_vhost/vhost.c
> index 0b19d2e..140d2ae 100644
> --- a/lib/librte_vhost/vhost.c
> +++ b/lib/librte_vhost/vhost.c
> @@ -475,3 +475,26 @@ rte_vhost_log_used_vring(int vid, uint16_t
> vring_idx,
>
> vhost_log_used_vring(dev, vq, offset, len);
> }
> +
> +uint32_t
> +rte_vhost_rx_queue_count(int vid, uint16_t qid)
> +{
> + struct virtio_net *dev;
> + struct vhost_virtqueue *vq;
> +
> + dev = get_device(vid);
> + if (!dev)
> + return 0;
> +
> + if (unlikely(qid >= dev->nr_vring || (qid & 1) == 0)) {
I assume the '& 1' is to ensure it's a virtio txq? It might be clearer to use the VIRTIO_TXQ var or similar to make this clearer.
> + RTE_LOG(ERR, VHOST_DATA, "(%d) %s: invalid virtqueue idx
> %d.\n",
> + dev->vid, __func__, qid);
> + return 0;
> + }
> +
> + vq = dev->virtqueue[qid];
> + if (unlikely(vq->enabled == 0))
> + return 0;
> +
> + return *((volatile uint16_t *)&vq->avail->idx) - vq->last_avail_idx;
> +}
> --
> 2.7.4
Acked-by: Ciara Loftus <ciara.loftus@intel.com>
next prev parent reply other threads:[~2017-05-23 13:10 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-05-22 20:01 Zhihong Wang
2017-05-23 11:53 ` Jens Freimann
2017-05-23 12:51 ` Wang, Zhihong
2017-05-24 8:14 ` Yuanhan Liu
2017-05-24 8:36 ` Jens Freimann
2017-05-24 8:42 ` Yuanhan Liu
2017-05-26 4:35 ` Wang, Zhihong
2017-05-23 13:10 ` Loftus, Ciara [this message]
2017-05-24 0:54 ` [dpdk-dev] [PATCH v2] " Zhihong Wang
2017-05-26 17:18 ` [dpdk-dev] [PATCH v3] " Zhihong Wang
2017-05-26 5:54 ` Yuanhan Liu
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=74F120C019F4A64C9B78E802F6AD4CC24FA4D84F@IRSMSX106.ger.corp.intel.com \
--to=ciara.loftus@intel.com \
--cc=dev@dpdk.org \
--cc=yuanhan.liu@linux.intel.com \
--cc=zhihong.wang@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).