From nobody Wed Apr 16 03:26:11 2025 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=fail; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=fail(p=none dis=none) header.from=redhat.com Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1646880437434264.59295304537056; Wed, 9 Mar 2022 18:47:17 -0800 (PST) Received: from localhost ([::1]:50624 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1nS8pg-0001Cj-Dy for importer@patchew.org; Wed, 09 Mar 2022 21:47:16 -0500 Received: from eggs.gnu.org ([209.51.188.92]:38896) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1nS8jZ-0005Wd-T6 for qemu-devel@nongnu.org; Wed, 09 Mar 2022 21:40:57 -0500 Received: from us-smtp-delivery-124.mimecast.com ([170.10.133.124]:60365) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1nS8jG-0004NE-60 for qemu-devel@nongnu.org; Wed, 09 Mar 2022 21:40:40 -0500 Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-650-TAluQo9ZPqKQDhX2tbzr6g-1; Wed, 09 Mar 2022 21:40:36 -0500 Received: from smtp.corp.redhat.com (int-mx07.intmail.prod.int.phx2.redhat.com [10.5.11.22]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id 8E9D2801AFE; Thu, 10 Mar 2022 02:40:35 +0000 (UTC) Received: from localhost.localdomain (ovpn-13-194.pek2.redhat.com [10.72.13.194]) by smtp.corp.redhat.com (Postfix) with ESMTP id 97B2A100685F; Thu, 10 Mar 2022 02:40:33 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1646880037; 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=Pf+Oh+kaqPK5hQeYp7bMskoan+bjH8zR9jz15a9c2Ws=; b=KlDk0Dv5spcPuNVWcXP6ks8iD4hHhJDvva36C3eECnZWFTHzhi4aQoH/64PYBp7ezlCd7Y ZAZIP8ihTEPUKxxjxaQpTCMraG8eZt+e1VAzNnZRsEnnz3t50sczYA6MkysGrMiQ0ZoQO5 tCLulmz/tQ2wRBbY983VM8sV4I3exMU= X-MC-Unique: TAluQo9ZPqKQDhX2tbzr6g-1 From: Jason Wang To: qemu-devel@nongnu.org, peter.maydell@linaro.org Subject: [PULL V2 12/15] vdpa: Add custom IOTLB translations to SVQ Date: Thu, 10 Mar 2022 10:39:04 +0800 Message-Id: <20220310023907.2443-13-jasowang@redhat.com> In-Reply-To: <20220310023907.2443-1-jasowang@redhat.com> References: <20220310023907.2443-1-jasowang@redhat.com> MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.84 on 10.5.11.22 Authentication-Results: relay.mimecast.com; auth=pass smtp.auth=CUSA124A263 smtp.mailfrom=jasowang@redhat.com X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: redhat.com Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=170.10.133.124; envelope-from=jasowang@redhat.com; helo=us-smtp-delivery-124.mimecast.com X-Spam_score_int: -21 X-Spam_score: -2.2 X-Spam_bar: -- X-Spam_report: (-2.2 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.082, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_NONE=-0.0001, RCVD_IN_MSPIKE_H5=0.001, RCVD_IN_MSPIKE_WL=0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, T_SCC_BODY_TEXT_LINE=-0.01 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: =?UTF-8?q?Eugenio=20P=C3=A9rez?= , Jason Wang , "Michael S . Tsirkin" Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: "Qemu-devel" X-ZohoMail-DKIM: fail (Header signature does not verify) X-ZM-MESSAGEID: 1646880439026100001 From: Eugenio P=C3=A9rez Use translations added in VhostIOVATree in SVQ. Only introduce usage here, not allocation and deallocation. As with previous patches, we use the dead code paths of shadow_vqs_enabled to avoid commiting too many changes at once. These are impossible to take at the moment. Acked-by: Michael S. Tsirkin Signed-off-by: Eugenio P=C3=A9rez Signed-off-by: Jason Wang --- hw/virtio/vhost-shadow-virtqueue.c | 75 +++++++++++++++++++++-- hw/virtio/vhost-shadow-virtqueue.h | 6 +- hw/virtio/vhost-vdpa.c | 122 +++++++++++++++++++++++++++++++--= ---- include/hw/virtio/vhost-vdpa.h | 3 + 4 files changed, 181 insertions(+), 25 deletions(-) diff --git a/hw/virtio/vhost-shadow-virtqueue.c b/hw/virtio/vhost-shadow-vi= rtqueue.c index b044e30..fd00506 100644 --- a/hw/virtio/vhost-shadow-virtqueue.c +++ b/hw/virtio/vhost-shadow-virtqueue.c @@ -70,7 +70,58 @@ static uint16_t vhost_svq_available_slots(const VhostSha= dowVirtqueue *svq) return svq->vring.num - (svq->shadow_avail_idx - svq->shadow_used_idx); } =20 +/** + * Translate addresses between the qemu's virtual address and the SVQ IOVA + * + * @svq: Shadow VirtQueue + * @vaddr: Translated IOVA addresses + * @iovec: Source qemu's VA addresses + * @num: Length of iovec and minimum length of vaddr + */ +static bool vhost_svq_translate_addr(const VhostShadowVirtqueue *svq, + void **addrs, const struct iovec *iov= ec, + size_t num) +{ + if (num =3D=3D 0) { + return true; + } + + for (size_t i =3D 0; i < num; ++i) { + DMAMap needle =3D { + .translated_addr =3D (hwaddr)iovec[i].iov_base, + .size =3D iovec[i].iov_len, + }; + size_t off; + + const DMAMap *map =3D vhost_iova_tree_find_iova(svq->iova_tree, &n= eedle); + /* + * Map cannot be NULL since iova map contains all guest space and + * qemu already has a physical address mapped + */ + if (unlikely(!map)) { + qemu_log_mask(LOG_GUEST_ERROR, + "Invalid address 0x%"HWADDR_PRIx" given by guest= ", + needle.translated_addr); + return false; + } + + off =3D needle.translated_addr - map->translated_addr; + addrs[i] =3D (void *)(map->iova + off); + + if (unlikely(int128_gt(int128_add(needle.translated_addr, + iovec[i].iov_len), + map->translated_addr + map->size))) { + qemu_log_mask(LOG_GUEST_ERROR, + "Guest buffer expands over iova range"); + return false; + } + } + + return true; +} + static void vhost_vring_write_descs(VhostShadowVirtqueue *svq, + void * const *sg, const struct iovec *iovec, size_t num, bool more_descs, bool writ= e) { @@ -89,7 +140,7 @@ static void vhost_vring_write_descs(VhostShadowVirtqueue= *svq, } else { descs[i].flags =3D flags; } - descs[i].addr =3D cpu_to_le64((hwaddr)iovec[n].iov_base); + descs[i].addr =3D cpu_to_le64((hwaddr)sg[n]); descs[i].len =3D cpu_to_le32(iovec[n].iov_len); =20 last =3D i; @@ -105,6 +156,8 @@ static bool vhost_svq_add_split(VhostShadowVirtqueue *s= vq, { unsigned avail_idx; vring_avail_t *avail =3D svq->vring.avail; + bool ok; + g_autofree void **sgs =3D g_new(void *, MAX(elem->out_num, elem->in_nu= m)); =20 *head =3D svq->free_head; =20 @@ -115,9 +168,20 @@ static bool vhost_svq_add_split(VhostShadowVirtqueue *= svq, return false; } =20 - vhost_vring_write_descs(svq, elem->out_sg, elem->out_num, + ok =3D vhost_svq_translate_addr(svq, sgs, elem->out_sg, elem->out_num); + if (unlikely(!ok)) { + return false; + } + vhost_vring_write_descs(svq, sgs, elem->out_sg, elem->out_num, elem->in_num > 0, false); - vhost_vring_write_descs(svq, elem->in_sg, elem->in_num, false, true); + + + ok =3D vhost_svq_translate_addr(svq, sgs, elem->in_sg, elem->in_num); + if (unlikely(!ok)) { + return false; + } + + vhost_vring_write_descs(svq, sgs, elem->in_sg, elem->in_num, false, tr= ue); =20 /* * Put the entry in the available array (but don't update avail->idx u= ntil @@ -521,11 +585,13 @@ void vhost_svq_stop(VhostShadowVirtqueue *svq) * Creates vhost shadow virtqueue, and instructs the vhost device to use t= he * shadow methods and file descriptors. * + * @iova_tree: Tree to perform descriptors translations + * * Returns the new virtqueue or NULL. * * In case of error, reason is reported through error_report. */ -VhostShadowVirtqueue *vhost_svq_new(void) +VhostShadowVirtqueue *vhost_svq_new(VhostIOVATree *iova_tree) { g_autofree VhostShadowVirtqueue *svq =3D g_new0(VhostShadowVirtqueue, = 1); int r; @@ -546,6 +612,7 @@ VhostShadowVirtqueue *vhost_svq_new(void) =20 event_notifier_init_fd(&svq->svq_kick, VHOST_FILE_UNBIND); event_notifier_set_handler(&svq->hdev_call, vhost_svq_handle_call); + svq->iova_tree =3D iova_tree; return g_steal_pointer(&svq); =20 err_init_hdev_call: diff --git a/hw/virtio/vhost-shadow-virtqueue.h b/hw/virtio/vhost-shadow-vi= rtqueue.h index 38b3b91..e5e24c5 100644 --- a/hw/virtio/vhost-shadow-virtqueue.h +++ b/hw/virtio/vhost-shadow-virtqueue.h @@ -13,6 +13,7 @@ #include "qemu/event_notifier.h" #include "hw/virtio/virtio.h" #include "standard-headers/linux/vhost_types.h" +#include "hw/virtio/vhost-iova-tree.h" =20 /* Shadow virtqueue to relay notifications */ typedef struct VhostShadowVirtqueue { @@ -43,6 +44,9 @@ typedef struct VhostShadowVirtqueue { /* Virtio device */ VirtIODevice *vdev; =20 + /* IOVA mapping */ + VhostIOVATree *iova_tree; + /* Map for use the guest's descriptors */ VirtQueueElement **ring_id_maps; =20 @@ -75,7 +79,7 @@ void vhost_svq_start(VhostShadowVirtqueue *svq, VirtIODev= ice *vdev, VirtQueue *vq); void vhost_svq_stop(VhostShadowVirtqueue *svq); =20 -VhostShadowVirtqueue *vhost_svq_new(void); +VhostShadowVirtqueue *vhost_svq_new(VhostIOVATree *iova_tree); =20 void vhost_svq_free(gpointer vq); G_DEFINE_AUTOPTR_CLEANUP_FUNC(VhostShadowVirtqueue, vhost_svq_free); diff --git a/hw/virtio/vhost-vdpa.c b/hw/virtio/vhost-vdpa.c index f07e53a..a7db129 100644 --- a/hw/virtio/vhost-vdpa.c +++ b/hw/virtio/vhost-vdpa.c @@ -209,6 +209,21 @@ static void vhost_vdpa_listener_region_add(MemoryListe= ner *listener, vaddr, section->readonly); =20 llsize =3D int128_sub(llend, int128_make64(iova)); + if (v->shadow_vqs_enabled) { + DMAMap mem_region =3D { + .translated_addr =3D (hwaddr)vaddr, + .size =3D int128_get64(llsize) - 1, + .perm =3D IOMMU_ACCESS_FLAG(true, section->readonly), + }; + + int r =3D vhost_iova_tree_map_alloc(v->iova_tree, &mem_region); + if (unlikely(r !=3D IOVA_OK)) { + error_report("Can't allocate a mapping (%d)", r); + goto fail; + } + + iova =3D mem_region.iova; + } =20 vhost_vdpa_iotlb_batch_begin_once(v); ret =3D vhost_vdpa_dma_map(v, iova, int128_get64(llsize), @@ -261,6 +276,20 @@ static void vhost_vdpa_listener_region_del(MemoryListe= ner *listener, =20 llsize =3D int128_sub(llend, int128_make64(iova)); =20 + if (v->shadow_vqs_enabled) { + const DMAMap *result; + const void *vaddr =3D memory_region_get_ram_ptr(section->mr) + + section->offset_within_region + + (iova - section->offset_within_address_space); + DMAMap mem_region =3D { + .translated_addr =3D (hwaddr)vaddr, + .size =3D int128_get64(llsize) - 1, + }; + + result =3D vhost_iova_tree_find_iova(v->iova_tree, &mem_region); + iova =3D result->iova; + vhost_iova_tree_remove(v->iova_tree, &mem_region); + } vhost_vdpa_iotlb_batch_begin_once(v); ret =3D vhost_vdpa_dma_unmap(v, iova, int128_get64(llsize)); if (ret) { @@ -370,7 +399,7 @@ static int vhost_vdpa_init_svq(struct vhost_dev *hdev, = struct vhost_vdpa *v, =20 shadow_vqs =3D g_ptr_array_new_full(hdev->nvqs, vhost_svq_free); for (unsigned n =3D 0; n < hdev->nvqs; ++n) { - g_autoptr(VhostShadowVirtqueue) svq =3D vhost_svq_new(); + g_autoptr(VhostShadowVirtqueue) svq =3D vhost_svq_new(v->iova_tree= ); =20 if (unlikely(!svq)) { error_setg(errp, "Cannot create svq %u", n); @@ -808,33 +837,70 @@ static int vhost_vdpa_svq_set_fds(struct vhost_dev *d= ev, /** * Unmap a SVQ area in the device */ -static bool vhost_vdpa_svq_unmap_ring(struct vhost_vdpa *v, hwaddr iova, - hwaddr size) +static bool vhost_vdpa_svq_unmap_ring(struct vhost_vdpa *v, + const DMAMap *needle) { + const DMAMap *result =3D vhost_iova_tree_find_iova(v->iova_tree, needl= e); + hwaddr size; int r; =20 - size =3D ROUND_UP(size, qemu_real_host_page_size); - r =3D vhost_vdpa_dma_unmap(v, iova, size); + if (unlikely(!result)) { + error_report("Unable to find SVQ address to unmap"); + return false; + } + + size =3D ROUND_UP(result->size, qemu_real_host_page_size); + r =3D vhost_vdpa_dma_unmap(v, result->iova, size); return r =3D=3D 0; } =20 static bool vhost_vdpa_svq_unmap_rings(struct vhost_dev *dev, const VhostShadowVirtqueue *svq) { + DMAMap needle =3D {}; struct vhost_vdpa *v =3D dev->opaque; struct vhost_vring_addr svq_addr; - size_t device_size =3D vhost_svq_device_area_size(svq); - size_t driver_size =3D vhost_svq_driver_area_size(svq); bool ok; =20 vhost_svq_get_vring_addr(svq, &svq_addr); =20 - ok =3D vhost_vdpa_svq_unmap_ring(v, svq_addr.desc_user_addr, driver_si= ze); + needle.translated_addr =3D svq_addr.desc_user_addr; + ok =3D vhost_vdpa_svq_unmap_ring(v, &needle); if (unlikely(!ok)) { return false; } =20 - return vhost_vdpa_svq_unmap_ring(v, svq_addr.used_user_addr, device_si= ze); + needle.translated_addr =3D svq_addr.used_user_addr; + return vhost_vdpa_svq_unmap_ring(v, &needle); +} + +/** + * Map the SVQ area in the device + * + * @v: Vhost-vdpa device + * @needle: The area to search iova + * @errorp: Error pointer + */ +static bool vhost_vdpa_svq_map_ring(struct vhost_vdpa *v, DMAMap *needle, + Error **errp) +{ + int r; + + r =3D vhost_iova_tree_map_alloc(v->iova_tree, needle); + if (unlikely(r !=3D IOVA_OK)) { + error_setg(errp, "Cannot allocate iova (%d)", r); + return false; + } + + r =3D vhost_vdpa_dma_map(v, needle->iova, needle->size + 1, + (void *)needle->translated_addr, + needle->perm =3D=3D IOMMU_RO); + if (unlikely(r !=3D 0)) { + error_setg_errno(errp, -r, "Cannot map region to device"); + vhost_iova_tree_remove(v->iova_tree, needle); + } + + return r =3D=3D 0; } =20 /** @@ -850,28 +916,44 @@ static bool vhost_vdpa_svq_map_rings(struct vhost_dev= *dev, struct vhost_vring_addr *addr, Error **errp) { + DMAMap device_region, driver_region; + struct vhost_vring_addr svq_addr; struct vhost_vdpa *v =3D dev->opaque; size_t device_size =3D vhost_svq_device_area_size(svq); size_t driver_size =3D vhost_svq_driver_area_size(svq); - int r; + size_t avail_offset; + bool ok; =20 ERRP_GUARD(); - vhost_svq_get_vring_addr(svq, addr); + vhost_svq_get_vring_addr(svq, &svq_addr); =20 - r =3D vhost_vdpa_dma_map(v, addr->desc_user_addr, driver_size, - (void *)addr->desc_user_addr, true); - if (unlikely(r !=3D 0)) { - error_setg_errno(errp, -r, "Cannot create vq driver region: "); + driver_region =3D (DMAMap) { + .translated_addr =3D svq_addr.desc_user_addr, + .size =3D driver_size - 1, + .perm =3D IOMMU_RO, + }; + ok =3D vhost_vdpa_svq_map_ring(v, &driver_region, errp); + if (unlikely(!ok)) { + error_prepend(errp, "Cannot create vq driver region: "); return false; } + addr->desc_user_addr =3D driver_region.iova; + avail_offset =3D svq_addr.avail_user_addr - svq_addr.desc_user_addr; + addr->avail_user_addr =3D driver_region.iova + avail_offset; =20 - r =3D vhost_vdpa_dma_map(v, addr->used_user_addr, device_size, - (void *)addr->used_user_addr, false); - if (unlikely(r !=3D 0)) { - error_setg_errno(errp, -r, "Cannot create vq device region: "); + device_region =3D (DMAMap) { + .translated_addr =3D svq_addr.used_user_addr, + .size =3D device_size - 1, + .perm =3D IOMMU_RW, + }; + ok =3D vhost_vdpa_svq_map_ring(v, &device_region, errp); + if (unlikely(!ok)) { + error_prepend(errp, "Cannot create vq device region: "); + vhost_vdpa_svq_unmap_ring(v, &driver_region); } + addr->used_user_addr =3D device_region.iova; =20 - return r =3D=3D 0; + return ok; } =20 static bool vhost_vdpa_svq_setup(struct vhost_dev *dev, diff --git a/include/hw/virtio/vhost-vdpa.h b/include/hw/virtio/vhost-vdpa.h index 009a9f3..ee8e939 100644 --- a/include/hw/virtio/vhost-vdpa.h +++ b/include/hw/virtio/vhost-vdpa.h @@ -14,6 +14,7 @@ =20 #include =20 +#include "hw/virtio/vhost-iova-tree.h" #include "hw/virtio/virtio.h" #include "standard-headers/linux/vhost_types.h" =20 @@ -30,6 +31,8 @@ typedef struct vhost_vdpa { MemoryListener listener; struct vhost_vdpa_iova_range iova_range; bool shadow_vqs_enabled; + /* IOVA mapping used by the Shadow Virtqueue */ + VhostIOVATree *iova_tree; GPtrArray *shadow_vqs; struct vhost_dev *dev; VhostVDPAHostNotifier notifier[VIRTIO_QUEUE_MAX]; --=20 2.7.4