From patchwork Tue Jun 6 08:18:32 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Maxime Coquelin X-Patchwork-Id: 128171 X-Patchwork-Delegate: maxime.coquelin@redhat.com Return-Path: X-Original-To: patchwork@inbox.dpdk.org Delivered-To: patchwork@inbox.dpdk.org Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id D225742C3E; Tue, 6 Jun 2023 10:19:45 +0200 (CEST) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id B5BA442D35; Tue, 6 Jun 2023 10:19:20 +0200 (CEST) Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) by mails.dpdk.org (Postfix) with ESMTP id 2909D40697 for ; Tue, 6 Jun 2023 10:19:19 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1686039558; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version:content-type:content-type: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=pRtT2U6/p62Goq2GE+Yawodl2EdFsJLJutV0Bqwp8lg=; b=UI4rV3z0/OD8jIpCi5LbCDirCMqKU13rfNqry6TD7LqvQY4HLozAwnfOKn5tHHjPbb1zTO 3+ig8lafFVxxCrFPVRSIkXH5ODrS+pi4guqAVEuorb1cQ/ccDxqGfo18aTKqftq9KAhXbb +jNr6sOyo15bhYxPD0I4JDoE+k/4IyY= Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-70-A5w1xqoqMqGJ0jr36D-spQ-1; Tue, 06 Jun 2023 04:19:15 -0400 X-MC-Unique: A5w1xqoqMqGJ0jr36D-spQ-1 Received: from smtp.corp.redhat.com (int-mx01.intmail.prod.int.rdu2.redhat.com [10.11.54.1]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 1364485A5BE; Tue, 6 Jun 2023 08:19:15 +0000 (UTC) Received: from max-t490s.redhat.com (unknown [10.39.208.25]) by smtp.corp.redhat.com (Postfix) with ESMTP id 5AAAC40CF8F6; Tue, 6 Jun 2023 08:19:12 +0000 (UTC) From: Maxime Coquelin To: dev@dpdk.org, chenbo.xia@intel.com, david.marchand@redhat.com, mkp@redhat.com, fbl@redhat.com, jasowang@redhat.com, cunming.liang@intel.com, xieyongji@bytedance.com, echaudro@redhat.com, eperezma@redhat.com, amorenoz@redhat.com, lulu@redhat.com Cc: Maxime Coquelin Subject: [PATCH v5 06/26] vhost: add offset field to IOTLB entries Date: Tue, 6 Jun 2023 10:18:32 +0200 Message-Id: <20230606081852.71003-7-maxime.coquelin@redhat.com> In-Reply-To: <20230606081852.71003-1-maxime.coquelin@redhat.com> References: <20230606081852.71003-1-maxime.coquelin@redhat.com> MIME-Version: 1.0 X-Scanned-By: MIMEDefang 3.1 on 10.11.54.1 X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: redhat.com 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 This patch is a preliminary work to prepare for VDUSE support, for which we need to keep track of the mmaped base address and offset in order to be able to unmap it later when IOTLB entry is invalidated. Signed-off-by: Maxime Coquelin Reviewed-by: Chenbo Xia --- lib/vhost/iotlb.c | 30 ++++++++++++++++++------------ lib/vhost/iotlb.h | 2 +- lib/vhost/vhost_user.c | 2 +- 3 files changed, 20 insertions(+), 14 deletions(-) diff --git a/lib/vhost/iotlb.c b/lib/vhost/iotlb.c index 6d49bf6b30..aa5100e6e7 100644 --- a/lib/vhost/iotlb.c +++ b/lib/vhost/iotlb.c @@ -17,6 +17,7 @@ struct vhost_iotlb_entry { uint64_t iova; uint64_t uaddr; + uint64_t uoffset; uint64_t size; uint8_t perm; }; @@ -27,15 +28,18 @@ static bool vhost_user_iotlb_share_page(struct vhost_iotlb_entry *a, struct vhost_iotlb_entry *b, uint64_t align) { - uint64_t a_end, b_start; + uint64_t a_start, a_end, b_start; if (a == NULL || b == NULL) return false; + a_start = a->uaddr + a->uoffset; + b_start = b->uaddr + b->uoffset; + /* Assumes entry a lower than entry b */ - RTE_ASSERT(a->uaddr < b->uaddr); - a_end = RTE_ALIGN_CEIL(a->uaddr + a->size, align); - b_start = RTE_ALIGN_FLOOR(b->uaddr, align); + RTE_ASSERT(a_start < b_start); + a_end = RTE_ALIGN_CEIL(a_start + a->size, align); + b_start = RTE_ALIGN_FLOOR(b_start, align); return a_end > b_start; } @@ -43,11 +47,12 @@ vhost_user_iotlb_share_page(struct vhost_iotlb_entry *a, struct vhost_iotlb_entr static void vhost_user_iotlb_set_dump(struct virtio_net *dev, struct vhost_iotlb_entry *node) { - uint64_t align; + uint64_t align, start; - align = hua_to_alignment(dev->mem, (void *)(uintptr_t)node->uaddr); + start = node->uaddr + node->uoffset; + align = hua_to_alignment(dev->mem, (void *)(uintptr_t)start); - mem_set_dump((void *)(uintptr_t)node->uaddr, node->size, true, align); + mem_set_dump((void *)(uintptr_t)start, node->size, true, align); } static void @@ -56,10 +61,10 @@ vhost_user_iotlb_clear_dump(struct virtio_net *dev, struct vhost_iotlb_entry *no { uint64_t align, start, end; - start = node->uaddr; - end = node->uaddr + node->size; + start = node->uaddr + node->uoffset; + end = start + node->size; - align = hua_to_alignment(dev->mem, (void *)(uintptr_t)node->uaddr); + align = hua_to_alignment(dev->mem, (void *)(uintptr_t)start); /* Skip first page if shared with previous entry. */ if (vhost_user_iotlb_share_page(prev, node, align)) @@ -234,7 +239,7 @@ vhost_user_iotlb_cache_random_evict(struct virtio_net *dev) void vhost_user_iotlb_cache_insert(struct virtio_net *dev, uint64_t iova, uint64_t uaddr, - uint64_t size, uint8_t perm) + uint64_t uoffset, uint64_t size, uint8_t perm) { struct vhost_iotlb_entry *node, *new_node; @@ -256,6 +261,7 @@ vhost_user_iotlb_cache_insert(struct virtio_net *dev, uint64_t iova, uint64_t ua new_node->iova = iova; new_node->uaddr = uaddr; + new_node->uoffset = uoffset; new_node->size = size; new_node->perm = perm; @@ -344,7 +350,7 @@ vhost_user_iotlb_cache_find(struct virtio_net *dev, uint64_t iova, uint64_t *siz offset = iova - node->iova; if (!vva) - vva = node->uaddr + offset; + vva = node->uaddr + node->uoffset + offset; mapped += node->size - offset; iova = node->iova + node->size; diff --git a/lib/vhost/iotlb.h b/lib/vhost/iotlb.h index 3490b9e6be..bee36c5903 100644 --- a/lib/vhost/iotlb.h +++ b/lib/vhost/iotlb.h @@ -58,7 +58,7 @@ vhost_user_iotlb_wr_unlock_all(struct virtio_net *dev) } void vhost_user_iotlb_cache_insert(struct virtio_net *dev, uint64_t iova, uint64_t uaddr, - uint64_t size, uint8_t perm); + uint64_t uoffset, uint64_t size, uint8_t perm); void vhost_user_iotlb_cache_remove(struct virtio_net *dev, uint64_t iova, uint64_t size); uint64_t vhost_user_iotlb_cache_find(struct virtio_net *dev, uint64_t iova, uint64_t *size, uint8_t perm); diff --git a/lib/vhost/vhost_user.c b/lib/vhost/vhost_user.c index f2fe7ebc93..7f88a8754f 100644 --- a/lib/vhost/vhost_user.c +++ b/lib/vhost/vhost_user.c @@ -2641,7 +2641,7 @@ vhost_user_iotlb_msg(struct virtio_net **pdev, if (!vva) return RTE_VHOST_MSG_RESULT_ERR; - vhost_user_iotlb_cache_insert(dev, imsg->iova, vva, len, imsg->perm); + vhost_user_iotlb_cache_insert(dev, imsg->iova, vva, 0, len, imsg->perm); for (i = 0; i < dev->nr_vring; i++) { struct vhost_virtqueue *vq = dev->virtqueue[i];