From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 3B586A0C50; Fri, 16 Jul 2021 09:41:39 +0200 (CEST) Received: from [217.70.189.124] (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id BACDF4133A; Fri, 16 Jul 2021 09:41:33 +0200 (CEST) Received: from mga17.intel.com (mga17.intel.com [192.55.52.151]) by mails.dpdk.org (Postfix) with ESMTP id B1349412FB for ; Fri, 16 Jul 2021 09:41:31 +0200 (CEST) X-IronPort-AV: E=McAfee;i="6200,9189,10046"; a="191062523" X-IronPort-AV: E=Sophos;i="5.84,244,1620716400"; d="scan'208";a="191062523" Received: from fmsmga002.fm.intel.com ([10.253.24.26]) by fmsmga107.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 16 Jul 2021 00:41:30 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.84,244,1620716400"; d="scan'208";a="506384315" Received: from dpdk_jiangcheng.sh.intel.com ([10.67.119.149]) by fmsmga002.fm.intel.com with ESMTP; 16 Jul 2021 00:41:27 -0700 From: Cheng Jiang To: maxime.coquelin@redhat.com, Chenbo.Xia@intel.com Cc: dev@dpdk.org, jiayu.hu@intel.com, yvonnex.yang@intel.com, Cheng Jiang Date: Fri, 16 Jul 2021 07:24:35 +0000 Message-Id: <20210716072436.19017-5-cheng1.jiang@intel.com> X-Mailer: git-send-email 2.29.2 In-Reply-To: <20210716072436.19017-1-cheng1.jiang@intel.com> References: <20210602042802.31943-1-cheng1.jiang@intel.com> <20210716072436.19017-1-cheng1.jiang@intel.com> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Subject: [dpdk-dev] [PATCH v5 4/5] examples/vhost: handle memory hotplug for async vhost X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 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" When the guest memory is hotplugged, the vhost application which enables DMA acceleration must stop DMA transfers before the vhost re-maps the guest memory. To accomplish that, we need to do these changes in the vhost sample: 1. add inflight pkt count. 2. add vring_state_changed() callback. 3. add inflight pkt drain process in destroy_device() and vring_state_changed(). Signed-off-by: Cheng Jiang --- examples/vhost/main.c | 48 +++++++++++++++++++++++++++++++++++++++++-- examples/vhost/main.h | 1 + 2 files changed, 47 insertions(+), 2 deletions(-) diff --git a/examples/vhost/main.c b/examples/vhost/main.c index d2179eadb9..9014c999be 100644 --- a/examples/vhost/main.c +++ b/examples/vhost/main.c @@ -851,8 +851,11 @@ complete_async_pkts(struct vhost_dev *vdev) complete_count = rte_vhost_poll_enqueue_completed(vdev->vid, VIRTIO_RXQ, p_cpl, MAX_PKT_BURST); - if (complete_count) + if (complete_count) { free_pkts(p_cpl, complete_count); + __atomic_sub_fetch(&vdev->pkts_inflight, complete_count, __ATOMIC_SEQ_CST); + } + } static __rte_always_inline void @@ -895,6 +898,7 @@ drain_vhost(struct vhost_dev *vdev) complete_async_pkts(vdev); ret = rte_vhost_submit_enqueue_burst(vdev->vid, VIRTIO_RXQ, m, nr_xmit, m_cpu_cpl, &cpu_cpl_nr); + __atomic_add_fetch(&vdev->pkts_inflight, ret - cpu_cpl_nr, __ATOMIC_SEQ_CST); if (cpu_cpl_nr) free_pkts(m_cpu_cpl, cpu_cpl_nr); @@ -1226,6 +1230,9 @@ drain_eth_rx(struct vhost_dev *vdev) enqueue_count = rte_vhost_submit_enqueue_burst(vdev->vid, VIRTIO_RXQ, pkts, rx_count, m_cpu_cpl, &cpu_cpl_nr); + __atomic_add_fetch(&vdev->pkts_inflight, enqueue_count - cpu_cpl_nr, + __ATOMIC_SEQ_CST); + if (cpu_cpl_nr) free_pkts(m_cpu_cpl, cpu_cpl_nr); @@ -1397,8 +1404,15 @@ destroy_device(int vid) "(%d) device has been removed from data core\n", vdev->vid); - if (async_vhost_driver) + if (async_vhost_driver) { + uint16_t n_pkt = 0; + struct rte_mbuf *m_cpl[vdev->pkts_inflight]; + n_pkt = rte_vhost_try_drain_queue_thread_unsafe(vid, VIRTIO_RXQ, m_cpl, + vdev->pkts_inflight, 2); + + free_pkts(m_cpl, n_pkt); rte_vhost_async_channel_unregister(vid, VIRTIO_RXQ); + } rte_free(vdev); } @@ -1487,6 +1501,35 @@ new_device(int vid) return 0; } +static int +vring_state_changed(int vid, uint16_t queue_id, int enable) +{ + struct vhost_dev *vdev = NULL; + + TAILQ_FOREACH(vdev, &vhost_dev_list, global_vdev_entry) { + if (vdev->vid == vid) + break; + } + if (!vdev) + return -1; + + if (queue_id != VIRTIO_RXQ) + return 0; + + if (async_vhost_driver) { + if (!enable) { + uint16_t n_pkt; + struct rte_mbuf *m_cpl[vdev->pkts_inflight]; + + n_pkt = rte_vhost_try_drain_queue_thread_unsafe(vid, queue_id, + m_cpl, vdev->pkts_inflight, 2); + free_pkts(m_cpl, n_pkt); + } + } + + return 0; +} + /* * These callback allow devices to be added to the data core when configuration * has been fully complete. @@ -1495,6 +1538,7 @@ static const struct vhost_device_ops virtio_net_device_ops = { .new_device = new_device, .destroy_device = destroy_device, + .vring_state_changed = vring_state_changed, }; /* diff --git a/examples/vhost/main.h b/examples/vhost/main.h index 0ccdce4b4a..e7b1ac60a6 100644 --- a/examples/vhost/main.h +++ b/examples/vhost/main.h @@ -51,6 +51,7 @@ struct vhost_dev { uint64_t features; size_t hdr_len; uint16_t nr_vrings; + uint16_t pkts_inflight; struct rte_vhost_memory *mem; struct device_statistics stats; TAILQ_ENTRY(vhost_dev) global_vdev_entry; -- 2.29.2