DPDK patches and discussions
 help / color / mirror / Atom feed
From: "Xia, Chenbo" <chenbo.xia@intel.com>
To: "Wang, YuanX" <yuanx.wang@intel.com>,
	"maxime.coquelin@redhat.com" <maxime.coquelin@redhat.com>,
	"dev@dpdk.org" <dev@dpdk.org>
Cc: "Hu, Jiayu" <jiayu.hu@intel.com>,
	"He, Xingguang" <xingguang.he@intel.com>,
	"stable@dpdk.org" <stable@dpdk.org>,
	"Ling, WeiX" <weix.ling@intel.com>
Subject: RE: [PATCH v2] examples/vhost: fix retry logic on eth rx path
Date: Tue, 21 Jun 2022 13:34:51 +0000	[thread overview]
Message-ID: <SN6PR11MB35049FE6DD181B96EA3B0D889CB39@SN6PR11MB3504.namprd11.prod.outlook.com> (raw)
In-Reply-To: <20220617070144.710487-1-yuanx.wang@intel.com>

Hi Yuan,

> -----Original Message-----
> From: Wang, YuanX <yuanx.wang@intel.com>
> Sent: Friday, June 17, 2022 3:02 PM
> To: maxime.coquelin@redhat.com; Xia, Chenbo <chenbo.xia@intel.com>;
> dev@dpdk.org
> Cc: Hu, Jiayu <jiayu.hu@intel.com>; He, Xingguang <xingguang.he@intel.com>;
> Wang, YuanX <yuanx.wang@intel.com>; stable@dpdk.org; Ling, WeiX
> <weix.ling@intel.com>
> Subject: [PATCH v2] examples/vhost: fix retry logic on eth rx path
> 
> drain_eth_rx() uses rte_vhost_avail_entries() to calculate
> the available entries to determine if a retry is required.
> However, this function only works with split rings, and
> calculating packed rings will return the wrong value and cause
> unnecessary retries resulting in a significant performance penalty.
> 
> This patch fix that by using the difference between tx/rx burst
> as the retry condition.
> 
> Fixes: 4ecf22e356de ("vhost: export device id as the interface to
> applications")
> Cc: stable@dpdk.org
> 
> Signed-off-by: Yuan Wang <yuanx.wang@intel.com>
> Tested-by: Wei Ling <weix.ling@intel.com>
> ---
> V2: Rebase to 22.07 rc1.
> ---
>  examples/vhost/main.c | 27 ++++++++++-----------------
>  1 file changed, 10 insertions(+), 17 deletions(-)
> 
> diff --git a/examples/vhost/main.c b/examples/vhost/main.c
> index e7fee5aa1b..e7a84333ed 100644
> --- a/examples/vhost/main.c
> +++ b/examples/vhost/main.c
> @@ -634,7 +634,7 @@ us_vhost_usage(const char *prgname)
>  {
>  	RTE_LOG(INFO, VHOST_CONFIG, "%s [EAL options] -- -p PORTMASK\n"
>  	"		--vm2vm [0|1|2]\n"
> -	"		--rx_retry [0|1] --mergeable [0|1] --stats [0-N]\n"
> +	"		--rx-retry [0|1] --mergeable [0|1] --stats [0-N]\n"
>  	"		--socket-file <path>\n"
>  	"		--nb-devices ND\n"
>  	"		-p PORTMASK: Set mask for ports to be used by
> application\n"
> @@ -1383,27 +1383,20 @@ drain_eth_rx(struct vhost_dev *vdev)
>  	if (!rx_count)
>  		return;
> 
> -	/*
> -	 * When "enable_retry" is set, here we wait and retry when there
> -	 * is no enough free slots in the queue to hold @rx_count packets,
> -	 * to diminish packet loss.
> -	 */
> -	if (enable_retry &&
> -	    unlikely(rx_count > rte_vhost_avail_entries(vdev->vid,
> -			VIRTIO_RXQ))) {
> -		uint32_t retry;
> +	enqueue_count = vdev_queue_ops[vdev->vid].enqueue_pkt_burst(vdev,
> +						VIRTIO_RXQ, pkts, rx_count);
> 
> -		for (retry = 0; retry < burst_rx_retry_num; retry++) {
> +	/* Retry if necessary */
> +	if (enable_retry && unlikely(enqueue_count < rx_count)) {
> +		uint32_t retry = 0;
> +
> +		while (enqueue_count < rx_count && retry++ <
> burst_rx_retry_num) {
>  			rte_delay_us(burst_rx_delay_time);
> -			if (rx_count <= rte_vhost_avail_entries(vdev->vid,
> -					VIRTIO_RXQ))
> -				break;
> +			enqueue_count += vdev_queue_ops[vdev-
> >vid].enqueue_pkt_burst(vdev,
> +								VIRTIO_RXQ, pkts,
> rx_count);

Looking at the code again, it seems current logic will enqueue more than rx_count
packets and same packet may be sent multiple times as you are enqueue multiple times
but using the same mbuf pointer and rx_count.

Thanks,
Chenbo 

>  		}
>  	}
> 
> -	enqueue_count = vdev_queue_ops[vdev->vid].enqueue_pkt_burst(vdev,
> -					VIRTIO_RXQ, pkts, rx_count);
> -
>  	if (enable_stats) {
>  		__atomic_add_fetch(&vdev->stats.rx_total_atomic, rx_count,
>  				__ATOMIC_SEQ_CST);
> --
> 2.25.1


  parent reply	other threads:[~2022-06-21 13:34 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-05-18 16:25 [PATCH] examples/vhost: Fix retry logic on Rx Yuan Wang
2022-05-26  9:30 ` Ling, WeiX
2022-06-17  7:01 ` [PATCH v2] examples/vhost: fix retry logic on eth rx path Yuan Wang
2022-06-20  3:20   ` Xia, Chenbo
2022-06-20  7:36     ` David Marchand
2022-06-20  7:49       ` Xia, Chenbo
2022-06-20  8:59         ` Hu, Jiayu
2022-06-20  9:09           ` Xia, Chenbo
2022-06-20  9:19             ` Wang, YuanX
2022-06-20  9:33               ` Xia, Chenbo
2022-06-20  9:42             ` Hu, Jiayu
2022-06-21 13:34   ` Xia, Chenbo [this message]
2022-06-22  2:26     ` Wang, YuanX
2022-06-22  6:27 ` [PATCH v3] " Yuan Wang
2022-06-22  7:23   ` Maxime Coquelin
2022-06-22  8:50     ` Wang, YuanX
2022-06-22  9:25 ` [PATCH v4] " Yuan Wang
2022-06-23  2:57   ` Xia, Chenbo
2022-06-23  7:20   ` Ling, WeiX
2022-07-01 13:57   ` Maxime Coquelin

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=SN6PR11MB35049FE6DD181B96EA3B0D889CB39@SN6PR11MB3504.namprd11.prod.outlook.com \
    --to=chenbo.xia@intel.com \
    --cc=dev@dpdk.org \
    --cc=jiayu.hu@intel.com \
    --cc=maxime.coquelin@redhat.com \
    --cc=stable@dpdk.org \
    --cc=weix.ling@intel.com \
    --cc=xingguang.he@intel.com \
    --cc=yuanx.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).