From mboxrd@z Thu Jan 1 00:00:00 1970 From: Maxime Coquelin Subject: Re: [PATCH 4/5] net/vhost: remove limit of vhost TX burst size Date: Wed, 1 Mar 2017 10:44:21 +0100 Message-ID: <0ae31488-1ea6-abb8-7bd0-b5526dc9fc9f@redhat.com> References: <1487926101-4637-1-git-send-email-zhiyong.yang@intel.com> <1487926101-4637-5-git-send-email-zhiyong.yang@intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit Cc: yuanhan.liu@linux.intel.com To: Zhiyong Yang , dev@dpdk.org Return-path: Received: from mx1.redhat.com (mx1.redhat.com [209.132.183.28]) by dpdk.org (Postfix) with ESMTP id A2CDB2B88 for ; Wed, 1 Mar 2017 10:44:25 +0100 (CET) In-Reply-To: <1487926101-4637-5-git-send-email-zhiyong.yang@intel.com> List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org Sender: "dev" On 02/24/2017 09:48 AM, Zhiyong Yang wrote: > vhost removes limit of TX burst size(32 pkts) and supports to make > an best effort to transmit pkts. > > Cc: yuanhan.liu@linux.intel.com > Cc: maxime.coquelin@redhat.com > > Signed-off-by: Zhiyong Yang > --- > drivers/net/vhost/rte_eth_vhost.c | 24 ++++++++++++++++++++++-- > 1 file changed, 22 insertions(+), 2 deletions(-) > > diff --git a/drivers/net/vhost/rte_eth_vhost.c b/drivers/net/vhost/rte_eth_vhost.c > index e98cffd..1e1fa34 100644 > --- a/drivers/net/vhost/rte_eth_vhost.c > +++ b/drivers/net/vhost/rte_eth_vhost.c > @@ -52,6 +52,7 @@ > #define ETH_VHOST_QUEUES_ARG "queues" > #define ETH_VHOST_CLIENT_ARG "client" > #define ETH_VHOST_DEQUEUE_ZERO_COPY "dequeue-zero-copy" > +#define VHOST_MAX_PKT_BURST 32 > > static const char *valid_arguments[] = { > ETH_VHOST_IFACE_ARG, > @@ -434,8 +435,27 @@ eth_vhost_tx(void *q, struct rte_mbuf **bufs, uint16_t nb_bufs) > goto out; > > /* Enqueue packets to guest RX queue */ > - nb_tx = rte_vhost_enqueue_burst(r->vid, > - r->virtqueue_id, bufs, nb_bufs); > + if (likely(nb_bufs <= VHOST_MAX_PKT_BURST)) > + nb_tx = rte_vhost_enqueue_burst(r->vid, r->virtqueue_id, > + bufs, nb_bufs); > + else { > + uint16_t nb_send = nb_bufs; > + > + while (nb_send) { > + uint16_t nb_pkts; > + uint16_t num = (uint16_t)RTE_MIN(nb_send, > + VHOST_MAX_PKT_BURST); > + > + nb_pkts = rte_vhost_enqueue_burst(r->vid, > + r->virtqueue_id, > + &bufs[nb_tx], num); > + > + nb_tx += nb_pkts; > + nb_send -= nb_pkts; > + if (nb_pkts < num) > + break; > + } > + } It looks like the if/else could be avoided, but maybe you did so for performance reason? If this is the case, maybe you could add a comment or at least state this in the commit message. Thanks, Maxime > > r->stats.pkts += nb_tx; > r->stats.missed_pkts += nb_bufs - nb_tx; >