From patchwork Mon Aug 19 11:34:56 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Tiwei Bie X-Patchwork-Id: 57739 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 157931BEA4; Mon, 19 Aug 2019 13:37:32 +0200 (CEST) Received: from mga02.intel.com (mga02.intel.com [134.134.136.20]) by dpdk.org (Postfix) with ESMTP id B256A1BE92; Mon, 19 Aug 2019 13:37:27 +0200 (CEST) X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from fmsmga008.fm.intel.com ([10.253.24.58]) by orsmga101.jf.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 19 Aug 2019 04:37:27 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.64,403,1559545200"; d="scan'208";a="177856871" Received: from dpdk-virtio-tbie-2.sh.intel.com ([10.67.104.71]) by fmsmga008.fm.intel.com with ESMTP; 19 Aug 2019 04:37:25 -0700 From: Tiwei Bie To: dev@dpdk.org, maxime.coquelin@redhat.com, zhihong.wang@intel.com Cc: lvyilong.lyl@alibaba-inc.com, yinan.wang@intel.com, xnhp0320@icloud.com, stable@dpdk.org Date: Mon, 19 Aug 2019 19:34:56 +0800 Message-Id: <20190819113457.15569-3-tiwei.bie@intel.com> X-Mailer: git-send-email 2.17.1 In-Reply-To: <20190819113457.15569-1-tiwei.bie@intel.com> References: <20190819113457.15569-1-tiwei.bie@intel.com> Subject: [dpdk-dev] [PATCH 2/3] vhost: fix vring addr handling during live migration 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" When live migration starts, QEMU will set ring addrs again for each virtqueue. In this case, we should try to translate ring addrs after we invalidating the ring, otherwise virtqueues can be enabled with the addrs untranslated. Besides, also leverage the access_ok flag in non-IOMMU case to prevent the data path accessing invalidated virtqueues. Fixes: 5a4933e56be4 ("vhost: postpone ring address translations at kick time only") Cc: stable@dpdk.org Reported-by: Yilong Lv Signed-off-by: Tiwei Bie Reviewed-by: Maxime Coquelin --- lib/librte_vhost/vhost.c | 3 +-- lib/librte_vhost/vhost_user.c | 12 ++++++++++-- 2 files changed, 11 insertions(+), 4 deletions(-) diff --git a/lib/librte_vhost/vhost.c b/lib/librte_vhost/vhost.c index 981837b5d..77be16069 100644 --- a/lib/librte_vhost/vhost.c +++ b/lib/librte_vhost/vhost.c @@ -358,7 +358,7 @@ vring_translate(struct virtio_net *dev, struct vhost_virtqueue *vq) { if (!(dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM))) - goto out; + return -1; if (vq_is_packed(dev)) { if (vring_translate_packed(dev, vq) < 0) @@ -367,7 +367,6 @@ vring_translate(struct virtio_net *dev, struct vhost_virtqueue *vq) if (vring_translate_split(dev, vq) < 0) return -1; } -out: vq->access_ok = 1; return 0; diff --git a/lib/librte_vhost/vhost_user.c b/lib/librte_vhost/vhost_user.c index ea43ab139..381250cbc 100644 --- a/lib/librte_vhost/vhost_user.c +++ b/lib/librte_vhost/vhost_user.c @@ -622,6 +622,7 @@ translate_ring_addresses(struct virtio_net *dev, int vq_index) return dev; } + vq->access_ok = 1; return dev; } @@ -680,6 +681,7 @@ translate_ring_addresses(struct virtio_net *dev, int vq_index) } vq->log_guest_addr = addr->log_guest_addr; + vq->access_ok = 1; VHOST_LOG_DEBUG(VHOST_CONFIG, "(%d) mapped address desc: %p\n", dev->vid, vq->desc); @@ -704,6 +706,7 @@ vhost_user_set_vring_addr(struct virtio_net **pdev, struct VhostUserMsg *msg, struct virtio_net *dev = *pdev; struct vhost_virtqueue *vq; struct vhost_vring_addr *addr = &msg->payload.addr; + bool access_ok; if (dev->mem == NULL) return RTE_VHOST_MSG_RESULT_ERR; @@ -711,6 +714,8 @@ vhost_user_set_vring_addr(struct virtio_net **pdev, struct VhostUserMsg *msg, /* addr->index refers to the queue index. The txq 1, rxq is 0. */ vq = dev->virtqueue[msg->payload.addr.index]; + access_ok = vq->access_ok; + /* * Rings addresses should not be interpreted as long as the ring is not * started and enabled @@ -719,8 +724,9 @@ vhost_user_set_vring_addr(struct virtio_net **pdev, struct VhostUserMsg *msg, vring_invalidate(dev, vq); - if (vq->enabled && (dev->features & - (1ULL << VHOST_USER_F_PROTOCOL_FEATURES))) { + if ((vq->enabled && (dev->features & + (1ULL << VHOST_USER_F_PROTOCOL_FEATURES))) || + access_ok) { dev = translate_ring_addresses(dev, msg->payload.addr.index); if (!dev) return RTE_VHOST_MSG_RESULT_ERR; @@ -1325,6 +1331,8 @@ vhost_user_get_vring_base(struct virtio_net **pdev, msg->size = sizeof(msg->payload.state); msg->fd_num = 0; + vring_invalidate(dev, vq); + return RTE_VHOST_MSG_RESULT_REPLY; }