From mboxrd@z Thu Jan  1 00:00:00 1970
From: "Michael S. Tsirkin" <mst@redhat.com>
Subject: Re: [PATCH v3] vhost: batch used descs chains write-back
 with packed ring
Date: Thu, 20 Dec 2018 09:30:30 -0500
Message-ID: <20181220092201-mutt-send-email-mst@kernel.org>
References: <20181220100022.3531-1-maxime.coquelin@redhat.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: dev@dpdk.org, i.maximets@samsung.com, tiwei.bie@intel.com,
 zhihong.wang@intel.com, jfreiman@redhat.com
To: Maxime Coquelin <maxime.coquelin@redhat.com>
Return-path: <dev-bounces@dpdk.org>
Received: from mx1.redhat.com (mx1.redhat.com [209.132.183.28])
 by dpdk.org (Postfix) with ESMTP id 5778A1B9A5
 for <dev@dpdk.org>; Thu, 20 Dec 2018 15:30:38 +0100 (CET)
Content-Disposition: inline
In-Reply-To: <20181220100022.3531-1-maxime.coquelin@redhat.com>
List-Id: DPDK patches and discussions <dev.dpdk.org>
List-Unsubscribe: <https://mails.dpdk.org/options/dev>,
 <mailto:dev-request@dpdk.org?subject=unsubscribe>
List-Archive: <http://mails.dpdk.org/archives/dev/>
List-Post: <mailto:dev@dpdk.org>
List-Help: <mailto:dev-request@dpdk.org?subject=help>
List-Subscribe: <https://mails.dpdk.org/listinfo/dev>,
 <mailto:dev-request@dpdk.org?subject=subscribe>
Errors-To: dev-bounces@dpdk.org
Sender: "dev" <dev-bounces@dpdk.org>

On Thu, Dec 20, 2018 at 11:00:22AM +0100, Maxime Coquelin wrote:
> Instead of writing back descriptors chains in order, let's
> write the first chain flags last in order to improve batching.
> 
> With Kernel's pktgen benchmark, ~3% performance gain is measured.
> 
> Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
> ---
>  lib/librte_vhost/virtio_net.c | 19 +++++++++++++++++--
>  1 file changed, 17 insertions(+), 2 deletions(-)
> 
> diff --git a/lib/librte_vhost/virtio_net.c b/lib/librte_vhost/virtio_net.c
> index 8c657a101..66ccd3c35 100644
> --- a/lib/librte_vhost/virtio_net.c
> +++ b/lib/librte_vhost/virtio_net.c
> @@ -97,6 +97,8 @@ flush_shadow_used_ring_packed(struct virtio_net *dev,
>  {
>  	int i;
>  	uint16_t used_idx = vq->last_used_idx;
> +	uint16_t head_idx = vq->last_used_idx;
> +	uint16_t head_flags = 0;
>  
>  	/* Split loop in two to save memory barriers */
>  	for (i = 0; i < vq->shadow_used_idx; i++) {
> @@ -126,12 +128,17 @@ flush_shadow_used_ring_packed(struct virtio_net *dev,
>  			flags &= ~VRING_DESC_F_AVAIL;
>  		}
>  
> -		vq->desc_packed[vq->last_used_idx].flags = flags;
> +		if (i > 0) {
> +			vq->desc_packed[vq->last_used_idx].flags = flags;
>  
> -		vhost_log_cache_used_vring(dev, vq,
> +			vhost_log_cache_used_vring(dev, vq,
>  					vq->last_used_idx *
>  					sizeof(struct vring_packed_desc),
>  					sizeof(struct vring_packed_desc));
> +		} else {
> +			head_idx = vq->last_used_idx;
> +			head_flags = flags;
> +		}
>  
>  		vq->last_used_idx += vq->shadow_used_packed[i].count;
>  		if (vq->last_used_idx >= vq->size) {
> @@ -140,7 +147,15 @@ flush_shadow_used_ring_packed(struct virtio_net *dev,
>  		}
>  	}
>  
> +	vq->desc_packed[head_idx].flags = head_flags;
> +
>  	rte_smp_wmb();
> +
> +	vhost_log_cache_used_vring(dev, vq,
> +				head_idx *
> +				sizeof(struct vring_packed_desc),
> +				sizeof(struct vring_packed_desc));
> +
>  	vq->shadow_used_idx = 0;
>  	vhost_log_cache_sync(dev, vq);

How about moving rte_smp_wmb into logging functions?
This way it's free with log disabled even on arm...

>  }
> -- 
> 2.17.2