From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mx1.redhat.com (mx1.redhat.com [209.132.183.28]) by dpdk.org (Postfix) with ESMTP id 5C7AE1B192 for ; Sun, 24 Sep 2017 18:21:01 +0200 (CEST) Received: from smtp.corp.redhat.com (int-mx01.intmail.prod.int.phx2.redhat.com [10.5.11.11]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id A7B14356DA; Sun, 24 Sep 2017 16:21:00 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mx1.redhat.com A7B14356DA Authentication-Results: ext-mx06.extmail.prod.ext.phx2.redhat.com; dmarc=none (p=none dis=none) header.from=redhat.com Authentication-Results: ext-mx06.extmail.prod.ext.phx2.redhat.com; spf=fail smtp.mailfrom=maxime.coquelin@redhat.com Received: from localhost.localdomain (ovpn-112-17.ams2.redhat.com [10.36.112.17]) by smtp.corp.redhat.com (Postfix) with ESMTP id C8B1069FB8; Sun, 24 Sep 2017 16:20:56 +0000 (UTC) From: Maxime Coquelin To: dev@dpdk.org, remy.horton@intel.com, tiwei.bie@intel.com, yliu@fridaylinux.org Cc: mst@redhat.com, jfreiman@redhat.com, vkaplans@redhat.com, jasowang@redhat.com, Maxime Coquelin Date: Sun, 24 Sep 2017 18:19:20 +0200 Message-Id: <20170924161921.30010-19-maxime.coquelin@redhat.com> In-Reply-To: <20170924161921.30010-1-maxime.coquelin@redhat.com> References: <20170924161921.30010-1-maxime.coquelin@redhat.com> X-Scanned-By: MIMEDefang 2.79 on 10.5.11.11 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.30]); Sun, 24 Sep 2017 16:21:00 +0000 (UTC) Subject: [dpdk-dev] [PATCH v2 18/19] vhost: iommu: Invalidate vring in case of matching IOTLB invalidate 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: , X-List-Received-Date: Sun, 24 Sep 2017 16:21:02 -0000 As soon as a page used by a ring is invalidated, the access_ok flag is cleared, so that processing threads try to map them again. Signed-off-by: Maxime Coquelin --- lib/librte_vhost/vhost.c | 14 ++++++++++++++ lib/librte_vhost/vhost.h | 1 + lib/librte_vhost/vhost_user.c | 38 +++++++++++++++++++++++++++++++++----- 3 files changed, 48 insertions(+), 5 deletions(-) diff --git a/lib/librte_vhost/vhost.c b/lib/librte_vhost/vhost.c index ef54835a6..061f08a69 100644 --- a/lib/librte_vhost/vhost.c +++ b/lib/librte_vhost/vhost.c @@ -172,6 +172,20 @@ vring_translate(struct virtio_net *dev, struct vhost_virtqueue *vq) return 0; } +void vring_invalidate(struct virtio_net *dev, struct vhost_virtqueue *vq) +{ + if (dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM)) + vhost_user_iotlb_wr_lock(vq); + + vq->access_ok = 0; + vq->desc = NULL; + vq->avail = NULL; + vq->used = NULL; + + if (dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM)) + vhost_user_iotlb_wr_unlock(vq); +} + static void init_vring_queue(struct virtio_net *dev, uint32_t vring_idx) { diff --git a/lib/librte_vhost/vhost.h b/lib/librte_vhost/vhost.h index b3fe6bb8e..fb48f3012 100644 --- a/lib/librte_vhost/vhost.h +++ b/lib/librte_vhost/vhost.h @@ -380,6 +380,7 @@ void vhost_backend_cleanup(struct virtio_net *dev); uint64_t __vhost_iova_to_vva(struct virtio_net *dev, struct vhost_virtqueue *vq, uint64_t iova, uint64_t size, uint8_t perm); int vring_translate(struct virtio_net *dev, struct vhost_virtqueue *vq); +void vring_invalidate(struct virtio_net *dev, struct vhost_virtqueue *vq); static __rte_always_inline uint64_t vhost_iova_to_vva(struct virtio_net *dev, struct vhost_virtqueue *vq, diff --git a/lib/librte_vhost/vhost_user.c b/lib/librte_vhost/vhost_user.c index 6d2c4d191..cef044307 100644 --- a/lib/librte_vhost/vhost_user.c +++ b/lib/librte_vhost/vhost_user.c @@ -391,11 +391,7 @@ vhost_user_set_vring_addr(struct virtio_net *dev, VhostUserMsg *msg) */ memcpy(&vq->ring_addrs, addr, sizeof(*addr)); - vq->desc = NULL; - vq->avail = NULL; - vq->used = NULL; - - vq->access_ok = 0; + vring_invalidate(dev, vq); return 0; } @@ -1012,6 +1008,35 @@ is_vring_iotlb_update(struct vhost_virtqueue *vq, struct vhost_iotlb_msg *imsg) } static int +is_vring_iotlb_invalidate(struct vhost_virtqueue *vq, + struct vhost_iotlb_msg *imsg) +{ + uint64_t istart, iend, vstart, vend; + + istart = imsg->iova; + iend = istart + imsg->size - 1; + + vstart = (uint64_t)vq->desc; + vend = vstart + sizeof(struct vring_desc) * vq->size - 1; + if (vstart <= iend && istart <= vend) + return 1; + + vstart = (uint64_t)vq->avail; + vend = vstart + sizeof(struct vring_avail); + vend += sizeof(uint16_t) * vq->size - 1; + if (vstart <= iend && istart <= vend) + return 1; + + vstart = (uint64_t)vq->used; + vend = vstart + sizeof(struct vring_used); + vend += sizeof(struct vring_used_elem) * vq->size - 1; + if (vstart <= iend && istart <= vend) + return 1; + + return 0; +} + +static int vhost_user_iotlb_msg(struct virtio_net **pdev, struct VhostUserMsg *msg) { struct virtio_net *dev = *pdev; @@ -1041,6 +1066,9 @@ vhost_user_iotlb_msg(struct virtio_net **pdev, struct VhostUserMsg *msg) vhost_user_iotlb_cache_remove(vq, imsg->iova, imsg->size); + + if (is_vring_iotlb_invalidate(vq, imsg)) + vring_invalidate(dev, vq); } break; default: -- 2.13.5