summary refs log tree commit diff stats
path: root/hw/vfio/container.c
diff options
context:
space:
mode:
Diffstat (limited to 'hw/vfio/container.c')
-rw-r--r--hw/vfio/container.c1325
1 files changed, 200 insertions, 1125 deletions
diff --git a/hw/vfio/container.c b/hw/vfio/container.c
index 030c6d3f89..250b20f424 100644
--- a/hw/vfio/container.c
+++ b/hw/vfio/container.c
@@ -1,1275 +1,350 @@
 /*
- * generic functions used by VFIO devices
+ * VFIO BASE CONTAINER
  *
- * Copyright Red Hat, Inc. 2012
+ * Copyright (C) 2023 Intel Corporation.
+ * Copyright Red Hat, Inc. 2023
  *
- * Authors:
- *  Alex Williamson <alex.williamson@redhat.com>
+ * Authors: Yi Liu <yi.l.liu@intel.com>
+ *          Eric Auger <eric.auger@redhat.com>
  *
- * This work is licensed under the terms of the GNU GPL, version 2.  See
- * the COPYING file in the top-level directory.
- *
- * Based on qemu-kvm device-assignment:
- *  Adapted for KVM by Qumranet.
- *  Copyright (c) 2007, Neocleus, Alex Novik (alex@neocleus.com)
- *  Copyright (c) 2007, Neocleus, Guy Zana (guy@neocleus.com)
- *  Copyright (C) 2008, Qumranet, Amit Shah (amit.shah@qumranet.com)
- *  Copyright (C) 2008, Red Hat, Amit Shah (amit.shah@redhat.com)
- *  Copyright (C) 2008, IBM, Muli Ben-Yehuda (muli@il.ibm.com)
+ * SPDX-License-Identifier: GPL-2.0-or-later
  */
 
-#include "qemu/osdep.h"
 #include <sys/ioctl.h>
 #include <linux/vfio.h>
 
-#include "hw/vfio/vfio-device.h"
-#include "system/address-spaces.h"
-#include "system/memory.h"
+#include "qemu/osdep.h"
+#include "system/tcg.h"
 #include "system/ram_addr.h"
-#include "qemu/error-report.h"
-#include "qemu/range.h"
-#include "system/reset.h"
-#include "trace.h"
 #include "qapi/error.h"
-#include "migration/cpr.h"
-#include "migration/blocker.h"
-#include "pci.h"
+#include "qemu/error-report.h"
 #include "hw/vfio/vfio-container.h"
+#include "hw/vfio/vfio-device.h" /* vfio_device_reset_handler */
+#include "system/reset.h"
 #include "vfio-helpers.h"
-#include "vfio-listener.h"
-
-#define TYPE_HOST_IOMMU_DEVICE_LEGACY_VFIO TYPE_HOST_IOMMU_DEVICE "-legacy-vfio"
-
-typedef QLIST_HEAD(VFIOGroupList, VFIOGroup) VFIOGroupList;
-static VFIOGroupList vfio_group_list =
-    QLIST_HEAD_INITIALIZER(vfio_group_list);
-
-static int vfio_ram_block_discard_disable(VFIOContainer *container, bool state)
-{
-    switch (container->iommu_type) {
-    case VFIO_TYPE1v2_IOMMU:
-    case VFIO_TYPE1_IOMMU:
-        /*
-         * We support coordinated discarding of RAM via the RamDiscardManager.
-         */
-        return ram_block_uncoordinated_discard_disable(state);
-    default:
-        /*
-         * VFIO_SPAPR_TCE_IOMMU most probably works just fine with
-         * RamDiscardManager, however, it is completely untested.
-         *
-         * VFIO_SPAPR_TCE_v2_IOMMU with "DMA memory preregistering" does
-         * completely the opposite of managing mapping/pinning dynamically as
-         * required by RamDiscardManager. We would have to special-case sections
-         * with a RamDiscardManager.
-         */
-        return ram_block_discard_disable(state);
-    }
-}
-
-static int vfio_dma_unmap_bitmap(const VFIOContainer *container,
-                                 hwaddr iova, ram_addr_t size,
-                                 IOMMUTLBEntry *iotlb)
-{
-    const VFIOContainerBase *bcontainer = VFIO_IOMMU(container);
-    struct vfio_iommu_type1_dma_unmap *unmap;
-    struct vfio_bitmap *bitmap;
-    VFIOBitmap vbmap;
-    int ret;
-
-    ret = vfio_bitmap_alloc(&vbmap, size);
-    if (ret) {
-        return ret;
-    }
-
-    unmap = g_malloc0(sizeof(*unmap) + sizeof(*bitmap));
-
-    unmap->argsz = sizeof(*unmap) + sizeof(*bitmap);
-    unmap->iova = iova;
-    unmap->size = size;
-    unmap->flags |= VFIO_DMA_UNMAP_FLAG_GET_DIRTY_BITMAP;
-    bitmap = (struct vfio_bitmap *)&unmap->data;
-
-    /*
-     * cpu_physical_memory_set_dirty_lebitmap() supports pages in bitmap of
-     * qemu_real_host_page_size to mark those dirty. Hence set bitmap_pgsize
-     * to qemu_real_host_page_size.
-     */
-    bitmap->pgsize = qemu_real_host_page_size();
-    bitmap->size = vbmap.size;
-    bitmap->data = (__u64 *)vbmap.bitmap;
-
-    if (vbmap.size > bcontainer->max_dirty_bitmap_size) {
-        error_report("UNMAP: Size of bitmap too big 0x%"PRIx64, vbmap.size);
-        ret = -E2BIG;
-        goto unmap_exit;
-    }
-
-    ret = ioctl(container->fd, VFIO_IOMMU_UNMAP_DMA, unmap);
-    if (!ret) {
-        cpu_physical_memory_set_dirty_lebitmap(vbmap.bitmap,
-                iotlb->translated_addr, vbmap.pages);
-    } else {
-        error_report("VFIO_UNMAP_DMA with DIRTY_BITMAP : %m");
-    }
-
-unmap_exit:
-    g_free(unmap);
-    g_free(vbmap.bitmap);
-
-    return ret;
-}
-
-static int vfio_legacy_dma_unmap_one(const VFIOContainerBase *bcontainer,
-                                     hwaddr iova, ram_addr_t size,
-                                     IOMMUTLBEntry *iotlb)
-{
-    const VFIOContainer *container = VFIO_IOMMU_LEGACY(bcontainer);
-    struct vfio_iommu_type1_dma_unmap unmap = {
-        .argsz = sizeof(unmap),
-        .flags = 0,
-        .iova = iova,
-        .size = size,
-    };
-    bool need_dirty_sync = false;
-    int ret;
-    Error *local_err = NULL;
-
-    g_assert(!cpr_is_incoming());
-
-    if (iotlb && vfio_container_dirty_tracking_is_started(bcontainer)) {
-        if (!vfio_container_devices_dirty_tracking_is_supported(bcontainer) &&
-            bcontainer->dirty_pages_supported) {
-            return vfio_dma_unmap_bitmap(container, iova, size, iotlb);
-        }
 
-        need_dirty_sync = true;
-    }
-
-    while (ioctl(container->fd, VFIO_IOMMU_UNMAP_DMA, &unmap)) {
-        /*
-         * The type1 backend has an off-by-one bug in the kernel (71a7d3d78e3c
-         * v4.15) where an overflow in its wrap-around check prevents us from
-         * unmapping the last page of the address space.  Test for the error
-         * condition and re-try the unmap excluding the last page.  The
-         * expectation is that we've never mapped the last page anyway and this
-         * unmap request comes via vIOMMU support which also makes it unlikely
-         * that this page is used.  This bug was introduced well after type1 v2
-         * support was introduced, so we shouldn't need to test for v1.  A fix
-         * is queued for kernel v5.0 so this workaround can be removed once
-         * affected kernels are sufficiently deprecated.
-         */
-        if (errno == EINVAL && unmap.size && !(unmap.iova + unmap.size) &&
-            container->iommu_type == VFIO_TYPE1v2_IOMMU) {
-            trace_vfio_legacy_dma_unmap_overflow_workaround();
-            unmap.size -= 1ULL << ctz64(bcontainer->pgsizes);
-            continue;
-        }
-        return -errno;
-    }
-
-    if (need_dirty_sync) {
-        ret = vfio_container_query_dirty_bitmap(bcontainer, iova, size,
-                                    iotlb->translated_addr, &local_err);
-        if (ret) {
-            error_report_err(local_err);
-            return ret;
-        }
-    }
+#include "trace.h"
 
-    return 0;
-}
+static QLIST_HEAD(, VFIOAddressSpace) vfio_address_spaces =
+    QLIST_HEAD_INITIALIZER(vfio_address_spaces);
 
-/*
- * DMA - Mapping and unmapping for the "type1" IOMMU interface used on x86
- */
-static int vfio_legacy_dma_unmap(const VFIOContainerBase *bcontainer,
-                                 hwaddr iova, ram_addr_t size,
-                                 IOMMUTLBEntry *iotlb, bool unmap_all)
+VFIOAddressSpace *vfio_address_space_get(AddressSpace *as)
 {
-    int ret;
-
-    if (unmap_all) {
-        /* The unmap ioctl doesn't accept a full 64-bit span. */
-        Int128 llsize = int128_rshift(int128_2_64(), 1);
-
-        ret = vfio_legacy_dma_unmap_one(bcontainer, 0, int128_get64(llsize),
-                                        iotlb);
+    VFIOAddressSpace *space;
 
-        if (ret == 0) {
-            ret = vfio_legacy_dma_unmap_one(bcontainer, int128_get64(llsize),
-                                            int128_get64(llsize), iotlb);
+    QLIST_FOREACH(space, &vfio_address_spaces, list) {
+        if (space->as == as) {
+            return space;
         }
-
-    } else {
-        ret = vfio_legacy_dma_unmap_one(bcontainer, iova, size, iotlb);
-    }
-
-    return ret;
-}
-
-static int vfio_legacy_dma_map(const VFIOContainerBase *bcontainer, hwaddr iova,
-                               ram_addr_t size, void *vaddr, bool readonly,
-                               MemoryRegion *mr)
-{
-    const VFIOContainer *container = VFIO_IOMMU_LEGACY(bcontainer);
-    struct vfio_iommu_type1_dma_map map = {
-        .argsz = sizeof(map),
-        .flags = VFIO_DMA_MAP_FLAG_READ,
-        .vaddr = (__u64)(uintptr_t)vaddr,
-        .iova = iova,
-        .size = size,
-    };
-
-    if (!readonly) {
-        map.flags |= VFIO_DMA_MAP_FLAG_WRITE;
-    }
-
-    /*
-     * Try the mapping, if it fails with EBUSY, unmap the region and try
-     * again.  This shouldn't be necessary, but we sometimes see it in
-     * the VGA ROM space.
-     */
-    if (ioctl(container->fd, VFIO_IOMMU_MAP_DMA, &map) == 0 ||
-        (errno == EBUSY &&
-         vfio_legacy_dma_unmap(bcontainer, iova, size, NULL, false) == 0 &&
-         ioctl(container->fd, VFIO_IOMMU_MAP_DMA, &map) == 0)) {
-        return 0;
-    }
-
-    return -errno;
-}
-
-static int
-vfio_legacy_set_dirty_page_tracking(const VFIOContainerBase *bcontainer,
-                                    bool start, Error **errp)
-{
-    const VFIOContainer *container = VFIO_IOMMU_LEGACY(bcontainer);
-    int ret;
-    struct vfio_iommu_type1_dirty_bitmap dirty = {
-        .argsz = sizeof(dirty),
-    };
-
-    if (start) {
-        dirty.flags = VFIO_IOMMU_DIRTY_PAGES_FLAG_START;
-    } else {
-        dirty.flags = VFIO_IOMMU_DIRTY_PAGES_FLAG_STOP;
-    }
-
-    ret = ioctl(container->fd, VFIO_IOMMU_DIRTY_PAGES, &dirty);
-    if (ret) {
-        ret = -errno;
-        error_setg_errno(errp, errno, "Failed to set dirty tracking flag 0x%x",
-                         dirty.flags);
-    }
-
-    return ret;
-}
-
-static int vfio_legacy_query_dirty_bitmap(const VFIOContainerBase *bcontainer,
-                      VFIOBitmap *vbmap, hwaddr iova, hwaddr size, Error **errp)
-{
-    const VFIOContainer *container = VFIO_IOMMU_LEGACY(bcontainer);
-    struct vfio_iommu_type1_dirty_bitmap *dbitmap;
-    struct vfio_iommu_type1_dirty_bitmap_get *range;
-    int ret;
-
-    dbitmap = g_malloc0(sizeof(*dbitmap) + sizeof(*range));
-
-    dbitmap->argsz = sizeof(*dbitmap) + sizeof(*range);
-    dbitmap->flags = VFIO_IOMMU_DIRTY_PAGES_FLAG_GET_BITMAP;
-    range = (struct vfio_iommu_type1_dirty_bitmap_get *)&dbitmap->data;
-    range->iova = iova;
-    range->size = size;
-
-    /*
-     * cpu_physical_memory_set_dirty_lebitmap() supports pages in bitmap of
-     * qemu_real_host_page_size to mark those dirty. Hence set bitmap's pgsize
-     * to qemu_real_host_page_size.
-     */
-    range->bitmap.pgsize = qemu_real_host_page_size();
-    range->bitmap.size = vbmap->size;
-    range->bitmap.data = (__u64 *)vbmap->bitmap;
-
-    ret = ioctl(container->fd, VFIO_IOMMU_DIRTY_PAGES, dbitmap);
-    if (ret) {
-        ret = -errno;
-        error_setg_errno(errp, errno,
-                         "Failed to get dirty bitmap for iova: 0x%"PRIx64
-                         " size: 0x%"PRIx64, (uint64_t)range->iova,
-                         (uint64_t)range->size);
-    }
-
-    g_free(dbitmap);
-
-    return ret;
-}
-
-static bool vfio_get_info_iova_range(struct vfio_iommu_type1_info *info,
-                                     VFIOContainerBase *bcontainer)
-{
-    struct vfio_info_cap_header *hdr;
-    struct vfio_iommu_type1_info_cap_iova_range *cap;
-
-    hdr = vfio_get_iommu_type1_info_cap(info,
-                                        VFIO_IOMMU_TYPE1_INFO_CAP_IOVA_RANGE);
-    if (!hdr) {
-        return false;
     }
 
-    cap = (void *)hdr;
-
-    for (int i = 0; i < cap->nr_iovas; i++) {
-        Range *range = g_new(Range, 1);
+    /* No suitable VFIOAddressSpace, create a new one */
+    space = g_malloc0(sizeof(*space));
+    space->as = as;
+    QLIST_INIT(&space->containers);
 
-        range_set_bounds(range, cap->iova_ranges[i].start,
-                         cap->iova_ranges[i].end);
-        bcontainer->iova_ranges =
-            range_list_insert(bcontainer->iova_ranges, range);
+    if (QLIST_EMPTY(&vfio_address_spaces)) {
+        qemu_register_reset(vfio_device_reset_handler, NULL);
     }
 
-    return true;
-}
-
-static void vfio_group_add_kvm_device(VFIOGroup *group)
-{
-    Error *err = NULL;
+    QLIST_INSERT_HEAD(&vfio_address_spaces, space, list);
 
-    if (vfio_kvm_device_add_fd(group->fd, &err)) {
-        error_reportf_err(err, "group ID %d: ", group->groupid);
-    }
+    return space;
 }
 
-static void vfio_group_del_kvm_device(VFIOGroup *group)
+void vfio_address_space_put(VFIOAddressSpace *space)
 {
-    Error *err = NULL;
-
-    if (vfio_kvm_device_del_fd(group->fd, &err)) {
-        error_reportf_err(err, "group ID %d: ", group->groupid);
+    if (!QLIST_EMPTY(&space->containers)) {
+        return;
     }
-}
 
-/*
- * vfio_get_iommu_type - selects the richest iommu_type (v2 first)
- */
-static int vfio_get_iommu_type(int container_fd,
-                               Error **errp)
-{
-    int iommu_types[] = { VFIO_TYPE1v2_IOMMU, VFIO_TYPE1_IOMMU,
-                          VFIO_SPAPR_TCE_v2_IOMMU, VFIO_SPAPR_TCE_IOMMU };
-    int i;
+    QLIST_REMOVE(space, list);
+    g_free(space);
 
-    for (i = 0; i < ARRAY_SIZE(iommu_types); i++) {
-        if (ioctl(container_fd, VFIO_CHECK_EXTENSION, iommu_types[i])) {
-            return iommu_types[i];
-        }
+    if (QLIST_EMPTY(&vfio_address_spaces)) {
+        qemu_unregister_reset(vfio_device_reset_handler, NULL);
     }
-    error_setg(errp, "No available IOMMU models");
-    return -EINVAL;
 }
 
-/*
- * vfio_get_iommu_ops - get a VFIOIOMMUClass associated with a type
- */
-static const char *vfio_get_iommu_class_name(int iommu_type)
+void vfio_address_space_insert(VFIOAddressSpace *space,
+                               VFIOContainer *bcontainer)
 {
-    switch (iommu_type) {
-    case VFIO_TYPE1v2_IOMMU:
-    case VFIO_TYPE1_IOMMU:
-        return TYPE_VFIO_IOMMU_LEGACY;
-        break;
-    case VFIO_SPAPR_TCE_v2_IOMMU:
-    case VFIO_SPAPR_TCE_IOMMU:
-        return TYPE_VFIO_IOMMU_SPAPR;
-        break;
-    default:
-        g_assert_not_reached();
-    };
+    QLIST_INSERT_HEAD(&space->containers, bcontainer, next);
+    bcontainer->space = space;
 }
 
-static bool vfio_set_iommu(int container_fd, int group_fd,
-                           int *iommu_type, Error **errp)
+int vfio_container_dma_map(VFIOContainer *bcontainer,
+                           hwaddr iova, ram_addr_t size,
+                           void *vaddr, bool readonly, MemoryRegion *mr)
 {
-    if (ioctl(group_fd, VFIO_GROUP_SET_CONTAINER, &container_fd)) {
-        error_setg_errno(errp, errno, "Failed to set group container");
-        return false;
-    }
-
-    while (ioctl(container_fd, VFIO_SET_IOMMU, *iommu_type)) {
-        if (*iommu_type == VFIO_SPAPR_TCE_v2_IOMMU) {
-            /*
-             * On sPAPR, despite the IOMMU subdriver always advertises v1 and
-             * v2, the running platform may not support v2 and there is no
-             * way to guess it until an IOMMU group gets added to the container.
-             * So in case it fails with v2, try v1 as a fallback.
-             */
-            *iommu_type = VFIO_SPAPR_TCE_IOMMU;
-            continue;
-        }
-        error_setg_errno(errp, errno, "Failed to set iommu for container");
-        return false;
-    }
-
-    return true;
-}
-
-static VFIOContainer *vfio_create_container(int fd, VFIOGroup *group,
-                                            Error **errp)
-{
-    int iommu_type;
-    const char *vioc_name;
-    VFIOContainer *container;
+    VFIOIOMMUClass *vioc = VFIO_IOMMU_GET_CLASS(bcontainer);
+    RAMBlock *rb = mr->ram_block;
+    int mfd = rb ? qemu_ram_get_fd(rb) : -1;
 
-    iommu_type = vfio_get_iommu_type(fd, errp);
-    if (iommu_type < 0) {
-        return NULL;
-    }
+    if (mfd >= 0 && vioc->dma_map_file) {
+        unsigned long start = vaddr - qemu_ram_get_host_addr(rb);
+        unsigned long offset = qemu_ram_get_fd_offset(rb);
 
-    /*
-     * During CPR, just set the container type and skip the ioctls, as the
-     * container and group are already configured in the kernel.
-     */
-    if (!cpr_is_incoming() &&
-        !vfio_set_iommu(fd, group->fd, &iommu_type, errp)) {
-        return NULL;
+        return vioc->dma_map_file(bcontainer, iova, size, mfd, start + offset,
+                                  readonly);
     }
-
-    vioc_name = vfio_get_iommu_class_name(iommu_type);
-
-    container = VFIO_IOMMU_LEGACY(object_new(vioc_name));
-    container->fd = fd;
-    container->iommu_type = iommu_type;
-    return container;
+    g_assert(vioc->dma_map);
+    return vioc->dma_map(bcontainer, iova, size, vaddr, readonly, mr);
 }
 
-static int vfio_get_iommu_info(VFIOContainer *container,
-                               struct vfio_iommu_type1_info **info)
+int vfio_container_dma_unmap(VFIOContainer *bcontainer,
+                             hwaddr iova, ram_addr_t size,
+                             IOMMUTLBEntry *iotlb, bool unmap_all)
 {
+    VFIOIOMMUClass *vioc = VFIO_IOMMU_GET_CLASS(bcontainer);
 
-    size_t argsz = sizeof(struct vfio_iommu_type1_info);
-
-    *info = g_new0(struct vfio_iommu_type1_info, 1);
-again:
-    (*info)->argsz = argsz;
-
-    if (ioctl(container->fd, VFIO_IOMMU_GET_INFO, *info)) {
-        g_free(*info);
-        *info = NULL;
-        return -errno;
-    }
-
-    if (((*info)->argsz > argsz)) {
-        argsz = (*info)->argsz;
-        *info = g_realloc(*info, argsz);
-        goto again;
-    }
-
-    return 0;
+    g_assert(vioc->dma_unmap);
+    return vioc->dma_unmap(bcontainer, iova, size, iotlb, unmap_all);
 }
 
-static struct vfio_info_cap_header *
-vfio_get_iommu_info_cap(struct vfio_iommu_type1_info *info, uint16_t id)
+bool vfio_container_add_section_window(VFIOContainer *bcontainer,
+                                       MemoryRegionSection *section,
+                                       Error **errp)
 {
-    struct vfio_info_cap_header *hdr;
-    void *ptr = info;
+    VFIOIOMMUClass *vioc = VFIO_IOMMU_GET_CLASS(bcontainer);
 
-    if (!(info->flags & VFIO_IOMMU_INFO_CAPS)) {
-        return NULL;
+    if (!vioc->add_window) {
+        return true;
     }
 
-    for (hdr = ptr + info->cap_offset; hdr != ptr; hdr = ptr + hdr->next) {
-        if (hdr->id == id) {
-            return hdr;
-        }
-    }
-
-    return NULL;
+    return vioc->add_window(bcontainer, section, errp);
 }
 
-static void vfio_get_iommu_info_migration(VFIOContainer *container,
-                                          struct vfio_iommu_type1_info *info)
+void vfio_container_del_section_window(VFIOContainer *bcontainer,
+                                       MemoryRegionSection *section)
 {
-    struct vfio_info_cap_header *hdr;
-    struct vfio_iommu_type1_info_cap_migration *cap_mig;
-    VFIOContainerBase *bcontainer = VFIO_IOMMU(container);
+    VFIOIOMMUClass *vioc = VFIO_IOMMU_GET_CLASS(bcontainer);
 
-    hdr = vfio_get_iommu_info_cap(info, VFIO_IOMMU_TYPE1_INFO_CAP_MIGRATION);
-    if (!hdr) {
+    if (!vioc->del_window) {
         return;
     }
 
-    cap_mig = container_of(hdr, struct vfio_iommu_type1_info_cap_migration,
-                            header);
-
-    /*
-     * cpu_physical_memory_set_dirty_lebitmap() supports pages in bitmap of
-     * qemu_real_host_page_size to mark those dirty.
-     */
-    if (cap_mig->pgsize_bitmap & qemu_real_host_page_size()) {
-        bcontainer->dirty_pages_supported = true;
-        bcontainer->max_dirty_bitmap_size = cap_mig->max_dirty_bitmap_size;
-        bcontainer->dirty_pgsizes = cap_mig->pgsize_bitmap;
-    }
+    return vioc->del_window(bcontainer, section);
 }
 
-static bool vfio_legacy_setup(VFIOContainerBase *bcontainer, Error **errp)
+int vfio_container_set_dirty_page_tracking(VFIOContainer *bcontainer,
+                                           bool start, Error **errp)
 {
-    VFIOContainer *container = VFIO_IOMMU_LEGACY(bcontainer);
-    g_autofree struct vfio_iommu_type1_info *info = NULL;
+    VFIOIOMMUClass *vioc = VFIO_IOMMU_GET_CLASS(bcontainer);
     int ret;
 
-    ret = vfio_get_iommu_info(container, &info);
-    if (ret) {
-        error_setg_errno(errp, -ret, "Failed to get VFIO IOMMU info");
-        return false;
+    if (!bcontainer->dirty_pages_supported) {
+        return 0;
     }
 
-    if (info->flags & VFIO_IOMMU_INFO_PGSIZES) {
-        bcontainer->pgsizes = info->iova_pgsizes;
-    } else {
-        bcontainer->pgsizes = qemu_real_host_page_size();
+    g_assert(vioc->set_dirty_page_tracking);
+    if (bcontainer->dirty_pages_started == start) {
+        return 0;
     }
 
-    if (!vfio_get_info_dma_avail(info, &bcontainer->dma_max_mappings)) {
-        bcontainer->dma_max_mappings = 65535;
+    ret = vioc->set_dirty_page_tracking(bcontainer, start, errp);
+    if (!ret) {
+        bcontainer->dirty_pages_started = start;
     }
 
-    vfio_get_info_iova_range(info, bcontainer);
-
-    vfio_get_iommu_info_migration(container, info);
-    return true;
+    return ret;
 }
 
-static bool vfio_container_attach_discard_disable(VFIOContainer *container,
-                                            VFIOGroup *group, Error **errp)
+static bool vfio_container_devices_dirty_tracking_is_started(
+    const VFIOContainer *bcontainer)
 {
-    int ret;
+    VFIODevice *vbasedev;
 
-    /*
-     * VFIO is currently incompatible with discarding of RAM insofar as the
-     * madvise to purge (zap) the page from QEMU's address space does not
-     * interact with the memory API and therefore leaves stale virtual to
-     * physical mappings in the IOMMU if the page was previously pinned.  We
-     * therefore set discarding broken for each group added to a container,
-     * whether the container is used individually or shared.  This provides
-     * us with options to allow devices within a group to opt-in and allow
-     * discarding, so long as it is done consistently for a group (for instance
-     * if the device is an mdev device where it is known that the host vendor
-     * driver will never pin pages outside of the working set of the guest
-     * driver, which would thus not be discarding candidates).
-     *
-     * The first opportunity to induce pinning occurs here where we attempt to
-     * attach the group to existing containers within the AddressSpace.  If any
-     * pages are already zapped from the virtual address space, such as from
-     * previous discards, new pinning will cause valid mappings to be
-     * re-established.  Likewise, when the overall MemoryListener for a new
-     * container is registered, a replay of mappings within the AddressSpace
-     * will occur, re-establishing any previously zapped pages as well.
-     *
-     * Especially virtio-balloon is currently only prevented from discarding
-     * new memory, it will not yet set ram_block_discard_set_required() and
-     * therefore, neither stops us here or deals with the sudden memory
-     * consumption of inflated memory.
-     *
-     * We do support discarding of memory coordinated via the RamDiscardManager
-     * with some IOMMU types. vfio_ram_block_discard_disable() handles the
-     * details once we know which type of IOMMU we are using.
-     */
-
-    ret = vfio_ram_block_discard_disable(container, true);
-    if (ret) {
-        error_setg_errno(errp, -ret, "Cannot set discarding of RAM broken");
-        if (ioctl(group->fd, VFIO_GROUP_UNSET_CONTAINER, &container->fd)) {
-            error_report("vfio: error disconnecting group %d from"
-                         " container", group->groupid);
+    QLIST_FOREACH(vbasedev, &bcontainer->device_list, container_next) {
+        if (!vbasedev->dirty_tracking) {
+            return false;
         }
     }
-    return !ret;
-}
 
-static bool vfio_container_group_add(VFIOContainer *container, VFIOGroup *group,
-                                     Error **errp)
-{
-    if (!vfio_container_attach_discard_disable(container, group, errp)) {
-        return false;
-    }
-    group->container = container;
-    QLIST_INSERT_HEAD(&container->group_list, group, container_next);
-    vfio_group_add_kvm_device(group);
-    /*
-     * Remember the container fd for each group, so we can attach to the same
-     * container after CPR.
-     */
-    cpr_resave_fd("vfio_container_for_group", group->groupid, container->fd);
     return true;
 }
 
-static void vfio_container_group_del(VFIOContainer *container, VFIOGroup *group)
+bool vfio_container_dirty_tracking_is_started(
+    const VFIOContainer *bcontainer)
 {
-    QLIST_REMOVE(group, container_next);
-    group->container = NULL;
-    vfio_group_del_kvm_device(group);
-    vfio_ram_block_discard_disable(container, false);
-    cpr_delete_fd("vfio_container_for_group", group->groupid);
+    return vfio_container_devices_dirty_tracking_is_started(bcontainer) ||
+           bcontainer->dirty_pages_started;
 }
 
-static bool vfio_container_connect(VFIOGroup *group, AddressSpace *as,
-                                   Error **errp)
+bool vfio_container_devices_dirty_tracking_is_supported(
+    const VFIOContainer *bcontainer)
 {
-    VFIOContainer *container;
-    VFIOContainerBase *bcontainer;
-    int ret, fd = -1;
-    VFIOAddressSpace *space;
-    VFIOIOMMUClass *vioc = NULL;
-    bool new_container = false;
-    bool group_was_added = false;
-
-    space = vfio_address_space_get(as);
-    fd = cpr_find_fd("vfio_container_for_group", group->groupid);
-
-    if (!cpr_is_incoming()) {
-        QLIST_FOREACH(bcontainer, &space->containers, next) {
-            container = VFIO_IOMMU_LEGACY(bcontainer);
-            if (!ioctl(group->fd, VFIO_GROUP_SET_CONTAINER, &container->fd)) {
-                return vfio_container_group_add(container, group, errp);
-            }
-        }
+    VFIODevice *vbasedev;
 
-        fd = qemu_open("/dev/vfio/vfio", O_RDWR, errp);
-        if (fd < 0) {
-            goto fail;
+    QLIST_FOREACH(vbasedev, &bcontainer->device_list, container_next) {
+        if (vbasedev->device_dirty_page_tracking == ON_OFF_AUTO_OFF) {
+            return false;
         }
-    } else {
-        /*
-         * For incoming CPR, the group is already attached in the kernel.
-         * If a container with matching fd is found, then update the
-         * userland group list and return.  If not, then after the loop,
-         * create the container struct and group list.
-         */
-        QLIST_FOREACH(bcontainer, &space->containers, next) {
-            container = VFIO_IOMMU_LEGACY(bcontainer);
-
-            if (vfio_cpr_container_match(container, group, fd)) {
-                return vfio_container_group_add(container, group, errp);
-            }
-        }
-    }
-
-    ret = ioctl(fd, VFIO_GET_API_VERSION);
-    if (ret != VFIO_API_VERSION) {
-        error_setg(errp, "supported vfio version: %d, "
-                   "reported version: %d", VFIO_API_VERSION, ret);
-        goto fail;
-    }
-
-    container = vfio_create_container(fd, group, errp);
-    if (!container) {
-        goto fail;
-    }
-    new_container = true;
-    bcontainer = VFIO_IOMMU(container);
-
-    if (!vfio_legacy_cpr_register_container(container, errp)) {
-        goto fail;
-    }
-
-    vioc = VFIO_IOMMU_GET_CLASS(bcontainer);
-    assert(vioc->setup);
-
-    if (!vioc->setup(bcontainer, errp)) {
-        goto fail;
-    }
-
-    vfio_address_space_insert(space, bcontainer);
-
-    if (!vfio_container_group_add(container, group, errp)) {
-        goto fail;
-    }
-    group_was_added = true;
-
-    /*
-     * If CPR, register the listener later, after all state that may
-     * affect regions and mapping boundaries has been cpr load'ed.  Later,
-     * the listener will invoke its callback on each flat section and call
-     * dma_map to supply the new vaddr, and the calls will match the mappings
-     * remembered by the kernel.
-     */
-    if (!cpr_is_incoming()) {
-        if (!vfio_listener_register(bcontainer, errp)) {
-            goto fail;
+        if (!vbasedev->dirty_pages_supported) {
+            return false;
         }
     }
 
-    bcontainer->initialized = true;
-
     return true;
-
-fail:
-    if (new_container) {
-        vfio_listener_unregister(bcontainer);
-    }
-
-    if (group_was_added) {
-        vfio_container_group_del(container, group);
-    }
-    if (vioc && vioc->release) {
-        vioc->release(bcontainer);
-    }
-    if (new_container) {
-        vfio_legacy_cpr_unregister_container(container);
-        object_unref(container);
-    }
-    if (fd >= 0) {
-        close(fd);
-    }
-    vfio_address_space_put(space);
-
-    return false;
-}
-
-static void vfio_container_disconnect(VFIOGroup *group)
-{
-    VFIOContainer *container = group->container;
-    VFIOContainerBase *bcontainer = VFIO_IOMMU(container);
-    VFIOIOMMUClass *vioc = VFIO_IOMMU_GET_CLASS(bcontainer);
-
-    QLIST_REMOVE(group, container_next);
-    group->container = NULL;
-    cpr_delete_fd("vfio_container_for_group", group->groupid);
-
-    /*
-     * Explicitly release the listener first before unset container,
-     * since unset may destroy the backend container if it's the last
-     * group.
-     */
-    if (QLIST_EMPTY(&container->group_list)) {
-        vfio_listener_unregister(bcontainer);
-        if (vioc->release) {
-            vioc->release(bcontainer);
-        }
-    }
-
-    if (ioctl(group->fd, VFIO_GROUP_UNSET_CONTAINER, &container->fd)) {
-        error_report("vfio: error disconnecting group %d from container",
-                     group->groupid);
-    }
-
-    if (QLIST_EMPTY(&container->group_list)) {
-        VFIOAddressSpace *space = bcontainer->space;
-
-        trace_vfio_container_disconnect(container->fd);
-        vfio_legacy_cpr_unregister_container(container);
-        close(container->fd);
-        object_unref(container);
-
-        vfio_address_space_put(space);
-    }
 }
 
-static VFIOGroup *vfio_group_get(int groupid, AddressSpace *as, Error **errp)
+static int vfio_device_dma_logging_report(VFIODevice *vbasedev, hwaddr iova,
+                                          hwaddr size, void *bitmap)
 {
-    ERRP_GUARD();
-    VFIOGroup *group;
-    char path[32];
-    struct vfio_group_status status = { .argsz = sizeof(status) };
-
-    QLIST_FOREACH(group, &vfio_group_list, next) {
-        if (group->groupid == groupid) {
-            /* Found it.  Now is it already in the right context? */
-            if (VFIO_IOMMU(group->container)->space->as == as) {
-                return group;
-            } else {
-                error_setg(errp, "group %d used in multiple address spaces",
-                           group->groupid);
-                return NULL;
-            }
-        }
-    }
+    uint64_t buf[DIV_ROUND_UP(sizeof(struct vfio_device_feature) +
+                        sizeof(struct vfio_device_feature_dma_logging_report),
+                        sizeof(uint64_t))] = {};
+    struct vfio_device_feature *feature = (struct vfio_device_feature *)buf;
+    struct vfio_device_feature_dma_logging_report *report =
+        (struct vfio_device_feature_dma_logging_report *)feature->data;
 
-    group = g_malloc0(sizeof(*group));
+    report->iova = iova;
+    report->length = size;
+    report->page_size = qemu_real_host_page_size();
+    report->bitmap = (uintptr_t)bitmap;
 
-    snprintf(path, sizeof(path), "/dev/vfio/%d", groupid);
-    group->fd = cpr_open_fd(path, O_RDWR, "vfio_group", groupid, errp);
-    if (group->fd < 0) {
-        goto free_group_exit;
-    }
+    feature->argsz = sizeof(buf);
+    feature->flags = VFIO_DEVICE_FEATURE_GET |
+                     VFIO_DEVICE_FEATURE_DMA_LOGGING_REPORT;
 
-    if (ioctl(group->fd, VFIO_GROUP_GET_STATUS, &status)) {
-        error_setg_errno(errp, errno, "failed to get group %d status", groupid);
-        goto close_fd_exit;
-    }
-
-    if (!(status.flags & VFIO_GROUP_FLAGS_VIABLE)) {
-        error_setg(errp, "group %d is not viable", groupid);
-        error_append_hint(errp,
-                          "Please ensure all devices within the iommu_group "
-                          "are bound to their vfio bus driver.\n");
-        goto close_fd_exit;
-    }
-
-    group->groupid = groupid;
-    QLIST_INIT(&group->device_list);
-
-    if (!vfio_container_connect(group, as, errp)) {
-        error_prepend(errp, "failed to setup container for group %d: ",
-                      groupid);
-        goto close_fd_exit;
-    }
-
-    QLIST_INSERT_HEAD(&vfio_group_list, group, next);
-
-    return group;
-
-close_fd_exit:
-    cpr_delete_fd("vfio_group", groupid);
-    close(group->fd);
-
-free_group_exit:
-    g_free(group);
-
-    return NULL;
-}
-
-static void vfio_group_put(VFIOGroup *group)
-{
-    if (!group || !QLIST_EMPTY(&group->device_list)) {
-        return;
-    }
-
-    if (!group->ram_block_discard_allowed) {
-        vfio_ram_block_discard_disable(group->container, false);
-    }
-    vfio_group_del_kvm_device(group);
-    vfio_container_disconnect(group);
-    QLIST_REMOVE(group, next);
-    trace_vfio_group_put(group->fd);
-    cpr_delete_fd("vfio_group", group->groupid);
-    close(group->fd);
-    g_free(group);
-}
-
-static bool vfio_device_get(VFIOGroup *group, const char *name,
-                            VFIODevice *vbasedev, Error **errp)
-{
-    g_autofree struct vfio_device_info *info = NULL;
-    int fd;
-
-    fd = vfio_cpr_group_get_device_fd(group->fd, name);
-    if (fd < 0) {
-        error_setg_errno(errp, errno, "error getting device from group %d",
-                         group->groupid);
-        error_append_hint(errp,
-                      "Verify all devices in group %d are bound to vfio-<bus> "
-                      "or pci-stub and not already in use\n", group->groupid);
-        return false;
-    }
-
-    info = vfio_get_device_info(fd);
-    if (!info) {
-        error_setg_errno(errp, errno, "error getting device info");
-        goto fail;
-    }
-
-    /*
-     * Set discarding of RAM as not broken for this group if the driver knows
-     * the device operates compatibly with discarding.  Setting must be
-     * consistent per group, but since compatibility is really only possible
-     * with mdev currently, we expect singleton groups.
-     */
-    if (vbasedev->ram_block_discard_allowed !=
-        group->ram_block_discard_allowed) {
-        if (!QLIST_EMPTY(&group->device_list)) {
-            error_setg(errp, "Inconsistent setting of support for discarding "
-                       "RAM (e.g., balloon) within group");
-            goto fail;
-        }
-
-        if (!group->ram_block_discard_allowed) {
-            group->ram_block_discard_allowed = true;
-            vfio_ram_block_discard_disable(group->container, false);
-        }
-    }
-
-    vfio_device_prepare(vbasedev, VFIO_IOMMU(group->container), info);
-
-    vbasedev->fd = fd;
-    vbasedev->group = group;
-    QLIST_INSERT_HEAD(&group->device_list, vbasedev, next);
-
-    trace_vfio_device_get(name, info->flags, info->num_regions, info->num_irqs);
-
-    return true;
-
-fail:
-    close(fd);
-    cpr_delete_fd(name, 0);
-    return false;
-}
-
-static void vfio_device_put(VFIODevice *vbasedev)
-{
-    if (!vbasedev->group) {
-        return;
-    }
-    QLIST_REMOVE(vbasedev, next);
-    vbasedev->group = NULL;
-    trace_vfio_device_put(vbasedev->fd);
-    cpr_delete_fd(vbasedev->name, 0);
-    close(vbasedev->fd);
+    return vbasedev->io_ops->device_feature(vbasedev, feature);
 }
 
-static int vfio_device_get_groupid(VFIODevice *vbasedev, Error **errp)
+static int vfio_container_iommu_query_dirty_bitmap(
+    const VFIOContainer *bcontainer, VFIOBitmap *vbmap, hwaddr iova,
+    hwaddr size, Error **errp)
 {
-    char *tmp, group_path[PATH_MAX];
-    g_autofree char *group_name = NULL;
-    int ret, groupid;
-    ssize_t len;
-
-    tmp = g_strdup_printf("%s/iommu_group", vbasedev->sysfsdev);
-    len = readlink(tmp, group_path, sizeof(group_path));
-    g_free(tmp);
-
-    if (len <= 0 || len >= sizeof(group_path)) {
-        ret = len < 0 ? -errno : -ENAMETOOLONG;
-        error_setg_errno(errp, -ret, "no iommu_group found");
-        return ret;
-    }
-
-    group_path[len] = 0;
+    VFIOIOMMUClass *vioc = VFIO_IOMMU_GET_CLASS(bcontainer);
 
-    group_name = g_path_get_basename(group_path);
-    if (sscanf(group_name, "%d", &groupid) != 1) {
-        error_setg_errno(errp, errno, "failed to read %s", group_path);
-        return -errno;
-    }
-    return groupid;
+    g_assert(vioc->query_dirty_bitmap);
+    return vioc->query_dirty_bitmap(bcontainer, vbmap, iova, size,
+                                               errp);
 }
 
-/*
- * vfio_device_attach: attach a device to a security context
- * @name and @vbasedev->name are likely to be different depending
- * on the type of the device, hence the need for passing @name
- */
-static bool vfio_legacy_attach_device(const char *name, VFIODevice *vbasedev,
-                                      AddressSpace *as, Error **errp)
+static int vfio_container_devices_query_dirty_bitmap(
+    const VFIOContainer *bcontainer, VFIOBitmap *vbmap, hwaddr iova,
+    hwaddr size, Error **errp)
 {
-    int groupid = vfio_device_get_groupid(vbasedev, errp);
-    VFIODevice *vbasedev_iter;
-    VFIOGroup *group;
-
-    if (groupid < 0) {
-        return false;
-    }
-
-    trace_vfio_device_attach(vbasedev->name, groupid);
-
-    group = vfio_group_get(groupid, as, errp);
-    if (!group) {
-        return false;
-    }
-
-    QLIST_FOREACH(vbasedev_iter, &group->device_list, next) {
-        if (strcmp(vbasedev_iter->name, vbasedev->name) == 0) {
-            error_setg(errp, "device is already attached");
-            goto group_put_exit;
-        }
-    }
-    if (!vfio_device_get(group, name, vbasedev, errp)) {
-        goto group_put_exit;
-    }
+    VFIODevice *vbasedev;
+    int ret;
 
-    if (!vfio_device_hiod_create_and_realize(vbasedev,
-                                             TYPE_HOST_IOMMU_DEVICE_LEGACY_VFIO,
-                                             errp)) {
-        goto device_put_exit;
-    }
+    QLIST_FOREACH(vbasedev, &bcontainer->device_list, container_next) {
+        ret = vfio_device_dma_logging_report(vbasedev, iova, size,
+                                             vbmap->bitmap);
+        if (ret) {
+            error_setg_errno(errp, -ret,
+                             "%s: Failed to get DMA logging report, iova: "
+                             "0x%" HWADDR_PRIx ", size: 0x%" HWADDR_PRIx,
+                             vbasedev->name, iova, size);
 
-    if (vbasedev->mdev) {
-        error_setg(&vbasedev->cpr.mdev_blocker,
-                   "CPR does not support vfio mdev %s", vbasedev->name);
-        if (migrate_add_blocker_modes(&vbasedev->cpr.mdev_blocker, errp,
-                                      MIG_MODE_CPR_TRANSFER, -1) < 0) {
-            goto hiod_unref_exit;
+            return ret;
         }
     }
 
-    return true;
-
-hiod_unref_exit:
-    object_unref(vbasedev->hiod);
-device_put_exit:
-    vfio_device_put(vbasedev);
-group_put_exit:
-    vfio_group_put(group);
-    return false;
+    return 0;
 }
 
-static void vfio_legacy_detach_device(VFIODevice *vbasedev)
+int vfio_container_query_dirty_bitmap(const VFIOContainer *bcontainer,
+                                      uint64_t iova, uint64_t size,
+                                      ram_addr_t ram_addr, Error **errp)
 {
-    VFIOGroup *group = vbasedev->group;
-
-    trace_vfio_device_detach(vbasedev->name, group->groupid);
-
-    vfio_device_unprepare(vbasedev);
-
-    migrate_del_blocker(&vbasedev->cpr.mdev_blocker);
-    object_unref(vbasedev->hiod);
-    vfio_device_put(vbasedev);
-    vfio_group_put(group);
-}
+    bool all_device_dirty_tracking =
+        vfio_container_devices_dirty_tracking_is_supported(bcontainer);
+    uint64_t dirty_pages;
+    VFIOBitmap vbmap;
+    int ret;
 
-static int vfio_legacy_pci_hot_reset(VFIODevice *vbasedev, bool single)
-{
-    VFIOPCIDevice *vdev = container_of(vbasedev, VFIOPCIDevice, vbasedev);
-    VFIOGroup *group;
-    struct vfio_pci_hot_reset_info *info = NULL;
-    struct vfio_pci_dependent_device *devices;
-    struct vfio_pci_hot_reset *reset;
-    int32_t *fds;
-    int ret, i, count;
-    bool multi = false;
-
-    trace_vfio_pci_hot_reset(vdev->vbasedev.name, single ? "one" : "multi");
-
-    if (!single) {
-        vfio_pci_pre_reset(vdev);
+    if (!bcontainer->dirty_pages_supported && !all_device_dirty_tracking) {
+        cpu_physical_memory_set_dirty_range(ram_addr, size,
+                                            tcg_enabled() ? DIRTY_CLIENTS_ALL :
+                                            DIRTY_CLIENTS_NOCODE);
+        return 0;
     }
-    vdev->vbasedev.needs_reset = false;
-
-    ret = vfio_pci_get_pci_hot_reset_info(vdev, &info);
 
+    ret = vfio_bitmap_alloc(&vbmap, size);
     if (ret) {
-        goto out_single;
-    }
-    devices = &info->devices[0];
-
-    trace_vfio_pci_hot_reset_has_dep_devices(vdev->vbasedev.name);
-
-    /* Verify that we have all the groups required */
-    for (i = 0; i < info->count; i++) {
-        PCIHostDeviceAddress host;
-        VFIOPCIDevice *tmp;
-        VFIODevice *vbasedev_iter;
-
-        host.domain = devices[i].segment;
-        host.bus = devices[i].bus;
-        host.slot = PCI_SLOT(devices[i].devfn);
-        host.function = PCI_FUNC(devices[i].devfn);
-
-        trace_vfio_pci_hot_reset_dep_devices(host.domain,
-                host.bus, host.slot, host.function, devices[i].group_id);
-
-        if (vfio_pci_host_match(&host, vdev->vbasedev.name)) {
-            continue;
-        }
-
-        QLIST_FOREACH(group, &vfio_group_list, next) {
-            if (group->groupid == devices[i].group_id) {
-                break;
-            }
-        }
-
-        if (!group) {
-            if (!vdev->has_pm_reset) {
-                error_report("vfio: Cannot reset device %s, "
-                             "depends on group %d which is not owned.",
-                             vdev->vbasedev.name, devices[i].group_id);
-            }
-            ret = -EPERM;
-            goto out;
-        }
-
-        /* Prep dependent devices for reset and clear our marker. */
-        QLIST_FOREACH(vbasedev_iter, &group->device_list, next) {
-            if (!vbasedev_iter->dev->realized ||
-                !vfio_pci_from_vfio_device(vbasedev_iter)) {
-                continue;
-            }
-            tmp = container_of(vbasedev_iter, VFIOPCIDevice, vbasedev);
-            if (vfio_pci_host_match(&host, tmp->vbasedev.name)) {
-                if (single) {
-                    ret = -EINVAL;
-                    goto out_single;
-                }
-                vfio_pci_pre_reset(tmp);
-                tmp->vbasedev.needs_reset = false;
-                multi = true;
-                break;
-            }
-        }
-    }
-
-    if (!single && !multi) {
-        ret = -EINVAL;
-        goto out_single;
-    }
-
-    /* Determine how many group fds need to be passed */
-    count = 0;
-    QLIST_FOREACH(group, &vfio_group_list, next) {
-        for (i = 0; i < info->count; i++) {
-            if (group->groupid == devices[i].group_id) {
-                count++;
-                break;
-            }
-        }
+        error_setg_errno(errp, -ret,
+                         "Failed to allocate dirty tracking bitmap");
+        return ret;
     }
 
-    reset = g_malloc0(sizeof(*reset) + (count * sizeof(*fds)));
-    reset->argsz = sizeof(*reset) + (count * sizeof(*fds));
-    fds = &reset->group_fds[0];
-
-    /* Fill in group fds */
-    QLIST_FOREACH(group, &vfio_group_list, next) {
-        for (i = 0; i < info->count; i++) {
-            if (group->groupid == devices[i].group_id) {
-                fds[reset->count++] = group->fd;
-                break;
-            }
-        }
+    if (all_device_dirty_tracking) {
+        ret = vfio_container_devices_query_dirty_bitmap(bcontainer, &vbmap, iova, size,
+                                                        errp);
+    } else {
+        ret = vfio_container_iommu_query_dirty_bitmap(bcontainer, &vbmap, iova, size,
+                                                     errp);
     }
 
-    /* Bus reset! */
-    ret = ioctl(vdev->vbasedev.fd, VFIO_DEVICE_PCI_HOT_RESET, reset);
-    g_free(reset);
     if (ret) {
-        ret = -errno;
+        goto out;
     }
 
-    trace_vfio_pci_hot_reset_result(vdev->vbasedev.name,
-                                    ret ? strerror(errno) : "Success");
+    dirty_pages = cpu_physical_memory_set_dirty_lebitmap(vbmap.bitmap, ram_addr,
+                                                         vbmap.pages);
 
+    trace_vfio_container_query_dirty_bitmap(iova, size, vbmap.size, ram_addr,
+                                            dirty_pages);
 out:
-    /* Re-enable INTx on affected devices */
-    for (i = 0; i < info->count; i++) {
-        PCIHostDeviceAddress host;
-        VFIOPCIDevice *tmp;
-        VFIODevice *vbasedev_iter;
-
-        host.domain = devices[i].segment;
-        host.bus = devices[i].bus;
-        host.slot = PCI_SLOT(devices[i].devfn);
-        host.function = PCI_FUNC(devices[i].devfn);
-
-        if (vfio_pci_host_match(&host, vdev->vbasedev.name)) {
-            continue;
-        }
-
-        QLIST_FOREACH(group, &vfio_group_list, next) {
-            if (group->groupid == devices[i].group_id) {
-                break;
-            }
-        }
-
-        if (!group) {
-            break;
-        }
-
-        QLIST_FOREACH(vbasedev_iter, &group->device_list, next) {
-            if (!vbasedev_iter->dev->realized ||
-                !vfio_pci_from_vfio_device(vbasedev_iter)) {
-                continue;
-            }
-            tmp = container_of(vbasedev_iter, VFIOPCIDevice, vbasedev);
-            if (vfio_pci_host_match(&host, tmp->vbasedev.name)) {
-                vfio_pci_post_reset(tmp);
-                break;
-            }
-        }
-    }
-out_single:
-    if (!single) {
-        vfio_pci_post_reset(vdev);
-    }
-    g_free(info);
+    g_free(vbmap.bitmap);
 
     return ret;
 }
 
-static void vfio_iommu_legacy_class_init(ObjectClass *klass, const void *data)
+static gpointer copy_iova_range(gconstpointer src, gpointer data)
 {
-    VFIOIOMMUClass *vioc = VFIO_IOMMU_CLASS(klass);
-
-    vioc->setup = vfio_legacy_setup;
-    vioc->dma_map = vfio_legacy_dma_map;
-    vioc->dma_unmap = vfio_legacy_dma_unmap;
-    vioc->attach_device = vfio_legacy_attach_device;
-    vioc->detach_device = vfio_legacy_detach_device;
-    vioc->set_dirty_page_tracking = vfio_legacy_set_dirty_page_tracking;
-    vioc->query_dirty_bitmap = vfio_legacy_query_dirty_bitmap;
-    vioc->pci_hot_reset = vfio_legacy_pci_hot_reset;
-};
-
-static bool hiod_legacy_vfio_realize(HostIOMMUDevice *hiod, void *opaque,
-                                     Error **errp)
-{
-    VFIODevice *vdev = opaque;
+     Range *source = (Range *)src;
+     Range *dest = g_new(Range, 1);
 
-    hiod->name = g_strdup(vdev->name);
-    hiod->agent = opaque;
-
-    return true;
+     range_set_bounds(dest, range_lob(source), range_upb(source));
+     return dest;
 }
 
-static int hiod_legacy_vfio_get_cap(HostIOMMUDevice *hiod, int cap,
-                                    Error **errp)
+GList *vfio_container_get_iova_ranges(const VFIOContainer *bcontainer)
 {
-    switch (cap) {
-    case HOST_IOMMU_DEVICE_CAP_AW_BITS:
-        return vfio_device_get_aw_bits(hiod->agent);
-    default:
-        error_setg(errp, "%s: unsupported capability %x", hiod->name, cap);
-        return -EINVAL;
-    }
+    assert(bcontainer);
+    return g_list_copy_deep(bcontainer->iova_ranges, copy_iova_range, NULL);
 }
 
-static GList *
-hiod_legacy_vfio_get_iova_ranges(HostIOMMUDevice *hiod)
+static void vfio_container_instance_finalize(Object *obj)
 {
-    VFIODevice *vdev = hiod->agent;
+    VFIOContainer *bcontainer = VFIO_IOMMU(obj);
+    VFIOGuestIOMMU *giommu, *tmp;
 
-    g_assert(vdev);
-    return vfio_container_get_iova_ranges(vdev->bcontainer);
-}
+    QLIST_SAFE_REMOVE(bcontainer, next);
 
-static uint64_t
-hiod_legacy_vfio_get_page_size_mask(HostIOMMUDevice *hiod)
-{
-    VFIODevice *vdev = hiod->agent;
+    QLIST_FOREACH_SAFE(giommu, &bcontainer->giommu_list, giommu_next, tmp) {
+        memory_region_unregister_iommu_notifier(
+                MEMORY_REGION(giommu->iommu_mr), &giommu->n);
+        QLIST_REMOVE(giommu, giommu_next);
+        g_free(giommu);
+    }
 
-    g_assert(vdev);
-    return vfio_container_get_page_size_mask(vdev->bcontainer);
+    g_list_free_full(bcontainer->iova_ranges, g_free);
 }
 
-static void vfio_iommu_legacy_instance_init(Object *obj)
+static void vfio_container_instance_init(Object *obj)
 {
-    VFIOContainer *container = VFIO_IOMMU_LEGACY(obj);
+    VFIOContainer *bcontainer = VFIO_IOMMU(obj);
 
-    QLIST_INIT(&container->group_list);
+    bcontainer->error = NULL;
+    bcontainer->dirty_pages_supported = false;
+    bcontainer->dma_max_mappings = 0;
+    bcontainer->iova_ranges = NULL;
+    QLIST_INIT(&bcontainer->giommu_list);
+    QLIST_INIT(&bcontainer->vrdl_list);
 }
 
-static void hiod_legacy_vfio_class_init(ObjectClass *oc, const void *data)
-{
-    HostIOMMUDeviceClass *hioc = HOST_IOMMU_DEVICE_CLASS(oc);
-
-    hioc->realize = hiod_legacy_vfio_realize;
-    hioc->get_cap = hiod_legacy_vfio_get_cap;
-    hioc->get_iova_ranges = hiod_legacy_vfio_get_iova_ranges;
-    hioc->get_page_size_mask = hiod_legacy_vfio_get_page_size_mask;
-};
-
 static const TypeInfo types[] = {
     {
-        .name = TYPE_VFIO_IOMMU_LEGACY,
-        .parent = TYPE_VFIO_IOMMU,
-        .instance_init = vfio_iommu_legacy_instance_init,
+        .name = TYPE_VFIO_IOMMU,
+        .parent = TYPE_OBJECT,
+        .instance_init = vfio_container_instance_init,
+        .instance_finalize = vfio_container_instance_finalize,
         .instance_size = sizeof(VFIOContainer),
-        .class_init = vfio_iommu_legacy_class_init,
-    }, {
-        .name = TYPE_HOST_IOMMU_DEVICE_LEGACY_VFIO,
-        .parent = TYPE_HOST_IOMMU_DEVICE,
-        .class_init = hiod_legacy_vfio_class_init,
-    }
+        .class_size = sizeof(VFIOIOMMUClass),
+        .abstract = true,
+    },
 };
 
 DEFINE_TYPES(types)