From patchwork Tue Oct 23 10:07:10 2018 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Maxime Coquelin X-Patchwork-Id: 47232 X-Patchwork-Delegate: maxime.coquelin@redhat.com Return-Path: X-Original-To: patchwork@dpdk.org Delivered-To: patchwork@dpdk.org Received: from [92.243.14.124] (localhost [127.0.0.1]) by dpdk.org (Postfix) with ESMTP id 981771B4A1; Tue, 23 Oct 2018 12:07:25 +0200 (CEST) Received: from mx1.redhat.com (mx1.redhat.com [209.132.183.28]) by dpdk.org (Postfix) with ESMTP id 469D71B49E; Tue, 23 Oct 2018 12:07:20 +0200 (CEST) Received: from smtp.corp.redhat.com (int-mx05.intmail.prod.int.phx2.redhat.com [10.5.11.15]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id 732B5C0587EE; Tue, 23 Oct 2018 10:07:19 +0000 (UTC) Received: from max-t460s.localdomain (ovpn-112-43.ams2.redhat.com [10.36.112.43]) by smtp.corp.redhat.com (Postfix) with ESMTP id D543E5D6AA; Tue, 23 Oct 2018 10:07:12 +0000 (UTC) From: Maxime Coquelin To: dev@dpdk.org, tiwei.bie@intel.com, jfreimann@redhat.com, zhihong.wang@intel.com Cc: Maxime Coquelin , stable@dpdk.org Date: Tue, 23 Oct 2018 12:07:10 +0200 Message-Id: <20181023100710.14739-1-maxime.coquelin@redhat.com> X-Scanned-By: MIMEDefang 2.79 on 10.5.11.15 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.32]); Tue, 23 Oct 2018 10:07:19 +0000 (UTC) Subject: [dpdk-dev] [PATCH v2] vhost: avoid memory barriers when no descriptors dequeued X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org Sender: "dev" In both split and packed dequeue paths, flush_shadow_used_ring and vhost_ring_call variants gets called even if not packets have been dequeued, and so no descriptors updates happened. It has an impact on CPU pipeline, as memory barriers are used in these functions. This patch don't call these functions if no descriptors have been dequeued. The performance gain with split ring when dequeue zero-copy is disabled should be null, but should be noticeable with packed ring or dequeue zero-copy enabled. Fixes: ae999ce49dcb ("vhost: add Tx support for packed ring") Fixes: 915cf9404225 ("vhost: use shadow used ring in dequeue path") Cc: stable@dpdk.org Signed-off-by: Maxime Coquelin Reviewed-by: Jens Freimann Tested-by: Jens Freimann Reviewed-by: Tiwei Bie --- Changes in v2: - Fix shadow_used_idx reset in error path (Tiwei) lib/librte_vhost/virtio_net.c | 24 ++++++++++++++++-------- 1 file changed, 16 insertions(+), 8 deletions(-) diff --git a/lib/librte_vhost/virtio_net.c b/lib/librte_vhost/virtio_net.c index 1c31c0562..8ad30c94a 100644 --- a/lib/librte_vhost/virtio_net.c +++ b/lib/librte_vhost/virtio_net.c @@ -1360,8 +1360,10 @@ virtio_dev_tx_split(struct virtio_net *dev, struct vhost_virtqueue *vq, } } - flush_shadow_used_ring_split(dev, vq); - vhost_vring_call_split(dev, vq); + if (likely(vq->shadow_used_idx)) { + flush_shadow_used_ring_split(dev, vq); + vhost_vring_call_split(dev, vq); + } } rte_prefetch0(&vq->avail->ring[vq->last_avail_idx & (vq->size - 1)]); @@ -1440,8 +1442,10 @@ virtio_dev_tx_split(struct virtio_net *dev, struct vhost_virtqueue *vq, do_data_copy_dequeue(vq); if (unlikely(i < count)) vq->shadow_used_idx = i; - flush_shadow_used_ring_split(dev, vq); - vhost_vring_call_split(dev, vq); + if (likely(vq->shadow_used_idx)) { + flush_shadow_used_ring_split(dev, vq); + vhost_vring_call_split(dev, vq); + } } return i; @@ -1476,8 +1480,10 @@ virtio_dev_tx_packed(struct virtio_net *dev, struct vhost_virtqueue *vq, } } - flush_shadow_used_ring_packed(dev, vq); - vhost_vring_call_packed(dev, vq); + if (likely(vq->shadow_used_idx)) { + flush_shadow_used_ring_packed(dev, vq); + vhost_vring_call_packed(dev, vq); + } } VHOST_LOG_DEBUG(VHOST_DATA, "(%d) %s\n", dev->vid, __func__); @@ -1555,8 +1561,10 @@ virtio_dev_tx_packed(struct virtio_net *dev, struct vhost_virtqueue *vq, do_data_copy_dequeue(vq); if (unlikely(i < count)) vq->shadow_used_idx = i; - flush_shadow_used_ring_packed(dev, vq); - vhost_vring_call_packed(dev, vq); + if (likely(vq->shadow_used_idx)) { + flush_shadow_used_ring_packed(dev, vq); + vhost_vring_call_packed(dev, vq); + } } return i;