QEMU源码全解析 —— virtio（8）

接前一篇文章：

上一回讲解了 virtio balloon相关类所涉及的realize函数，如下表所示：

本回继续接着讲。

再来回顾一下，virtio balloon PCI 代理设备类型的初始化函数virtio_balloon_pci_class_init()，在hw/virtio/virtio-balloon-pci.c中，代码如下：

static void virtio_balloon_pci_class_init(ObjectClass *klass, void *data)
{
    DeviceClass *dc = DEVICE_CLASS(klass);
    VirtioPCIClass *k = VIRTIO_PCI_CLASS(klass);
    PCIDeviceClass *pcidev_k = PCI_DEVICE_CLASS(klass);
    k->realize = virtio_balloon_pci_realize;
    set_bit(DEVICE_CATEGORY_MISC, dc->categories);
    pcidev_k->vendor_id = PCI_VENDOR_ID_REDHAT_QUMRANET;
    pcidev_k->device_id = PCI_DEVICE_ID_VIRTIO_BALLOON;
    pcidev_k->revision = VIRTIO_PCI_ABI_VERSION;
    pcidev_k->class_id = PCI_CLASS_OTHERS;
}

device_set_realized函数在hw/core/qdev.c中，代码如下：

static void device_set_realized(Object *obj, bool value, Error **errp)
{
    DeviceState *dev = DEVICE(obj);
    DeviceClass *dc = DEVICE_GET_CLASS(dev);
    HotplugHandler *hotplug_ctrl;
    BusState *bus;
    NamedClockList *ncl;
    Error *local_err = NULL;
    bool unattached_parent = false;
    static int unattached_count;
 
    if (dev->hotplugged && !dc->hotpluggable) {
        error_setg(errp, QERR_DEVICE_NO_HOTPLUG, object_get_typename(obj));
        return;
    }
 
    if (value && !dev->realized) {
        if (!check_only_migratable(obj, errp)) {
            goto fail;
        }
 
        if (!obj->parent) {
            gchar *name = g_strdup_printf("device[%d]", unattached_count++);
 
            object_property_add_child(container_get(qdev_get_machine(),
                                                    "/unattached"),
                                      name, obj);
            unattached_parent = true;
            g_free(name);
        }
 
        hotplug_ctrl = qdev_get_hotplug_handler(dev);
        if (hotplug_ctrl) {
            hotplug_handler_pre_plug(hotplug_ctrl, dev, &local_err);
            if (local_err != NULL) {
                goto fail;
            }
        }
 
        if (dc->realize) {
            dc->realize(dev, &local_err);
            if (local_err != NULL) {
                goto fail;
            }
        }
 
        DEVICE_LISTENER_CALL(realize, Forward, dev);
 
        /*
         * always free/re-initialize here since the value cannot be cleaned up
         * in device_unrealize due to its usage later on in the unplug path
         */
        g_free(dev->canonical_path);
        dev->canonical_path = object_get_canonical_path(OBJECT(dev));
        QLIST_FOREACH(ncl, &dev->clocks, node) {
            if (ncl->alias) {
                continue;
            } else {
                clock_setup_canonical_path(ncl->clock);
            }
        }
 
        if (qdev_get_vmsd(dev)) {
            if (vmstate_register_with_alias_id(VMSTATE_IF(dev),
                                               VMSTATE_INSTANCE_ID_ANY,
                                               qdev_get_vmsd(dev), dev,
                                               dev->instance_id_alias,
                                               dev->alias_required_for_version,
                                               &local_err) < 0) {
                goto post_realize_fail;
            }
        }
 
        /*
         * Clear the reset state, in case the object was previously unrealized
         * with a dirty state.
         */
        resettable_state_clear(&dev->reset);
 
        QLIST_FOREACH(bus, &dev->child_bus, sibling) {
            if (!qbus_realize(bus, errp)) {
                goto child_realize_fail;
            }
        }
        if (dev->hotplugged) {
            /*
             * Reset the device, as well as its subtree which, at this point,
             * should be realized too.
             */
            resettable_assert_reset(OBJECT(dev), RESET_TYPE_COLD);
            resettable_change_parent(OBJECT(dev), OBJECT(dev->parent_bus),
                                     NULL);
            resettable_release_reset(OBJECT(dev), RESET_TYPE_COLD);
        }
        dev->pending_deleted_event = false;
 
        if (hotplug_ctrl) {
            hotplug_handler_plug(hotplug_ctrl, dev, &local_err);
            if (local_err != NULL) {
                goto child_realize_fail;
            }
       }
 
       qatomic_store_release(&dev->realized, value);
 
    } else if (!value && dev->realized) {
 
        /*
         * Change the value so that any concurrent users are aware
         * that the device is going to be unrealized
         *
         * TODO: change .realized property to enum that states
         * each phase of the device realization/unrealization
         */
 
        qatomic_set(&dev->realized, value);
        /*
         * Ensure that concurrent users see this update prior to
         * any other changes done by unrealize.
         */
        smp_wmb();
 
        QLIST_FOREACH(bus, &dev->child_bus, sibling) {
            qbus_unrealize(bus);
        }
        if (qdev_get_vmsd(dev)) {
            vmstate_unregister(VMSTATE_IF(dev), qdev_get_vmsd(dev), dev);
        }
        if (dc->unrealize) {
            dc->unrealize(dev);
        }
        dev->pending_deleted_event = true;
        DEVICE_LISTENER_CALL(unrealize, Reverse, dev);
    }
 
    assert(local_err == NULL);
    return;
 
child_realize_fail:
    QLIST_FOREACH(bus, &dev->child_bus, sibling) {
        qbus_unrealize(bus);
    }
 
    if (qdev_get_vmsd(dev)) {
        vmstate_unregister(VMSTATE_IF(dev), qdev_get_vmsd(dev), dev);
    }
 
post_realize_fail:
    g_free(dev->canonical_path);
    dev->canonical_path = NULL;
    if (dc->unrealize) {
        dc->unrealize(dev);
    }
 
fail:
    error_propagate(errp, local_err);
    if (unattached_parent) {
        /*
         * Beware, this doesn't just revert
         * object_property_add_child(), it also runs bus_remove()!
         */
        object_unparent(OBJECT(dev));
        unattached_count--;
    }
}

设置virtio PCI代理设备的realize属性时，device_set_realized函数中会首先调用DeviceClass->realize函数指针所指向的函数，也就是virtio_pci_dc_realize函数。代码片段如下：

        if (dc->realize) {
            dc->realize(dev, &local_err);
            if (local_err != NULL) {
                goto fail;
            }
        }

virtio_pci_dc_realize函数在hw/virtio/virtio-pci.c中，代码如下：

static void virtio_pci_dc_realize(DeviceState *qdev, Error **errp)
{
    VirtioPCIClass *vpciklass = VIRTIO_PCI_GET_CLASS(qdev);
    VirtIOPCIProxy *proxy = VIRTIO_PCI(qdev);
    PCIDevice *pci_dev = &proxy->pci_dev;
 
    if (!(proxy->flags & VIRTIO_PCI_FLAG_DISABLE_PCIE) &&
        virtio_pci_modern(proxy)) {
        pci_dev->cap_present |= QEMU_PCI_CAP_EXPRESS;
    }
 
    vpciklass->parent_dc_realize(qdev, errp);
}

virtio_pci_dc_realize函数中会调用VirtioPCIClass->parent_dc_realize，也就是pci_qdev_realize函数。

这里再复习强调一下，原本是：

经过VirtioPCIClass类的 初始化 函数virtio_pci_class_init后，是：

pci_qdev_realize函数在hw/pci/pci.c中，代码如下：

static void pci_qdev_realize(DeviceState *qdev, Error **errp)
{
    PCIDevice *pci_dev = (PCIDevice *)qdev;
    PCIDeviceClass *pc = PCI_DEVICE_GET_CLASS(pci_dev);
    ObjectClass *klass = OBJECT_CLASS(pc);
    Error *local_err = NULL;
    bool is_default_rom;
    uint16_t class_id;
 
    /*
     * capped by systemd (see: udev-builtin-net_id.c)
     * as it's the only known user honor it to avoid users
     * misconfigure QEMU and then wonder why acpi-index doesn't work
     */
    if (pci_dev->acpi_index > ONBOARD_INDEX_MAX) {
        error_setg(errp, "acpi-index should be less or equal to %u",
                   ONBOARD_INDEX_MAX);
        return;
    }
 
    /*
     * make sure that acpi-index is unique across all present PCI devices
     */
    if (pci_dev->acpi_index) {
        GSequence *used_indexes = pci_acpi_index_list();
 
        if (g_sequence_lookup(used_indexes,
                              GINT_TO_POINTER(pci_dev->acpi_index),
                              g_cmp_uint32, NULL)) {
            error_setg(errp, "a PCI device with acpi-index = %" PRIu32
                       " already exist", pci_dev->acpi_index);
            return;
        }
        g_sequence_insert_sorted(used_indexes,
                                 GINT_TO_POINTER(pci_dev->acpi_index),
                                 g_cmp_uint32, NULL);
    }
 
    if (pci_dev->romsize != -1 && !is_power_of_2(pci_dev->romsize)) {
        error_setg(errp, "ROM size %u is not a power of two", pci_dev->romsize);
        return;
    }
 
    /* initialize cap_present for pci_is_express() and pci_config_size(),
     * Note that hybrid PCIs are not set automatically and need to manage
     * QEMU_PCI_CAP_EXPRESS manually */
    if (object_class_dynamic_cast(klass, INTERFACE_PCIE_DEVICE) &&
       !object_class_dynamic_cast(klass, INTERFACE_CONVENTIONAL_PCI_DEVICE)) {
        pci_dev->cap_present |= QEMU_PCI_CAP_EXPRESS;
    }
 
    if (object_class_dynamic_cast(klass, INTERFACE_CXL_DEVICE)) {
        pci_dev->cap_present |= QEMU_PCIE_CAP_CXL;
    }
 
    pci_dev = do_pci_register_device(pci_dev,
                                     object_get_typename(OBJECT(qdev)),
                                     pci_dev->devfn, errp);
    if (pci_dev == NULL)
        return;
 
    if (pc->realize) {
        pc->realize(pci_dev, &local_err);
        if (local_err) {
            error_propagate(errp, local_err);
            do_pci_unregister_device(pci_dev);
            return;
        }
    }
 
    /*
     * A PCIe Downstream Port that do not have ARI Forwarding enabled must
     * associate only Device 0 with the device attached to the bus
     * representing the Link from the Port (PCIe base spec rev 4.0 ver 0.3,
     * sec 7.3.1).
     * With ARI, PCI_SLOT() can return non-zero value as the traditional
     * 5-bit Device Number and 3-bit Function Number fields in its associated
     * Routing IDs, Requester IDs and Completer IDs are interpreted as a
     * single 8-bit Function Number. Hence, ignore ARI capable devices.
     */
    if (pci_is_express(pci_dev) &&
        !pcie_find_capability(pci_dev, PCI_EXT_CAP_ID_ARI) &&
        pcie_has_upstream_port(pci_dev) &&
        PCI_SLOT(pci_dev->devfn)) {
        warn_report("PCI: slot %d is not valid for %s,"
                    " parent device only allows plugging into slot 0.",
                    PCI_SLOT(pci_dev->devfn), pci_dev->name);
    }
 
    if (pci_dev->failover_pair_id) {
        if (!pci_bus_is_express(pci_get_bus(pci_dev))) {
            error_setg(errp, "failover primary device must be on "
                             "PCIExpress bus");
            pci_qdev_unrealize(DEVICE(pci_dev));
            return;
        }
        class_id = pci_get_word(pci_dev->config + PCI_CLASS_DEVICE);
        if (class_id != PCI_CLASS_NETWORK_ETHERNET) {
            error_setg(errp, "failover primary device is not an "
                             "Ethernet device");
            pci_qdev_unrealize(DEVICE(pci_dev));
            return;
        }
        if ((pci_dev->cap_present & QEMU_PCI_CAP_MULTIFUNCTION)
            || (PCI_FUNC(pci_dev->devfn) != 0)) {
            error_setg(errp, "failover: primary device must be in its own "
                              "PCI slot");
            pci_qdev_unrealize(DEVICE(pci_dev));
            return;
        }
        qdev->allow_unplug_during_migration = true;
    }
 
    /* rom loading */
    is_default_rom = false;
    if (pci_dev->romfile == NULL && pc->romfile != NULL) {
        pci_dev->romfile = g_strdup(pc->romfile);
        is_default_rom = true;
    }
 
    pci_add_option_rom(pci_dev, is_default_rom, &local_err);
    if (local_err) {
        error_propagate(errp, local_err);
        pci_qdev_unrealize(DEVICE(pci_dev));
        return;
    }
 
    pci_set_power(pci_dev, true);
 
    pci_dev->msi_trigger = pci_msi_trigger;
}

在pci_qdev_realize函数中会调用PCIDeviceClass的realize函数指针所指向的函数，也就是virtio_pci_realize函数。代码片段如下：

    PCIDeviceClass *pc = PCI_DEVICE_GET_CLASS(pci_dev);
    ……
    if (pc->realize) {
        pc->realize(pci_dev, &local_err);
        if (local_err) {
            error_propagate(errp, local_err);
            do_pci_unregister_device(pci_dev);
            return;
        }
    }

virtio_pci_realize函数在hw/virtio/virtio-pci.c中，代码如下：

static void virtio_pci_realize(PCIDevice *pci_dev, Error **errp)
{
    VirtIOPCIProxy *proxy = VIRTIO_PCI(pci_dev);
    VirtioPCIClass *k = VIRTIO_PCI_GET_CLASS(pci_dev);
    bool pcie_port = pci_bus_is_express(pci_get_bus(pci_dev)) &&
                     !pci_bus_is_root(pci_get_bus(pci_dev));
 
    if (kvm_enabled() && !kvm_has_many_ioeventfds()) {
        proxy->flags &= ~VIRTIO_PCI_FLAG_USE_IOEVENTFD;
    }
 
    /* fd-based ioevents can't be synchronized in record/replay */
    if (replay_mode != REPLAY_MODE_NONE) {
        proxy->flags &= ~VIRTIO_PCI_FLAG_USE_IOEVENTFD;
    }
 
    /*
     * virtio pci bar layout used by default.
     * subclasses can re-arrange things if needed.
     *
     *   region 0   --  virtio legacy io bar
     *   region 1   --  msi-x bar
     *   region 2   --  virtio modern io bar (off by default)
     *   region 4+5 --  virtio modern memory (64bit) bar
     *
     */
    proxy->legacy_io_bar_idx  = 0;
    proxy->msix_bar_idx       = 1;
    proxy->modern_io_bar_idx  = 2;
    proxy->modern_mem_bar_idx = 4;
 
    proxy->common.offset = 0x0;
    proxy->common.size = 0x1000;
    proxy->common.type = VIRTIO_PCI_CAP_COMMON_CFG;
 
    proxy->isr.offset = 0x1000;
    proxy->isr.size = 0x1000;
    proxy->isr.type = VIRTIO_PCI_CAP_ISR_CFG;
 
    proxy->device.offset = 0x2000;
    proxy->device.size = 0x1000;
    proxy->device.type = VIRTIO_PCI_CAP_DEVICE_CFG;
 
    proxy->notify.offset = 0x3000;
    proxy->notify.size = virtio_pci_queue_mem_mult(proxy) * VIRTIO_QUEUE_MAX;
    proxy->notify.type = VIRTIO_PCI_CAP_NOTIFY_CFG;
 
    proxy->notify_pio.offset = 0x0;
    proxy->notify_pio.size = 0x4;
    proxy->notify_pio.type = VIRTIO_PCI_CAP_NOTIFY_CFG;
 
    /* subclasses can enforce modern, so do this unconditionally */
    memory_region_init(&proxy->modern_bar, OBJECT(proxy), "virtio-pci",
                       /* PCI BAR regions must be powers of 2 */
                       pow2ceil(proxy->notify.offset + proxy->notify.size));
 
    if (proxy->disable_legacy == ON_OFF_AUTO_AUTO) {
        proxy->disable_legacy = pcie_port ? ON_OFF_AUTO_ON : ON_OFF_AUTO_OFF;
    }
 
    if (!virtio_pci_modern(proxy) && !virtio_pci_legacy(proxy)) {
        error_setg(errp, "device cannot work as neither modern nor legacy mode"
                   " is enabled");
        error_append_hint(errp, "Set either disable-modern or disable-legacy"
                          " to off\n");
        return;
    }
 
    if (pcie_port && pci_is_express(pci_dev)) {
        int pos;
        uint16_t last_pcie_cap_offset = PCI_CONFIG_SPACE_SIZE;
 
        pos = pcie_endpoint_cap_init(pci_dev, 0);
        assert(pos > 0);
 
        pos = pci_add_capability(pci_dev, PCI_CAP_ID_PM, 0,
                                 PCI_PM_SIZEOF, errp);
        if (pos < 0) {
            return;
        }
 
        pci_dev->exp.pm_cap = pos;
 
        /*
         * Indicates that this function complies with revision 1.2 of the
         * PCI Power Management Interface Specification.
         */
        pci_set_word(pci_dev->config + pos + PCI_PM_PMC, 0x3);
 
        if (proxy->flags & VIRTIO_PCI_FLAG_AER) {
            pcie_aer_init(pci_dev, PCI_ERR_VER, last_pcie_cap_offset,
                          PCI_ERR_SIZEOF, NULL);
            last_pcie_cap_offset += PCI_ERR_SIZEOF;
        }
 
        if (proxy->flags & VIRTIO_PCI_FLAG_INIT_DEVERR) {
            /* Init error enabling flags */
            pcie_cap_deverr_init(pci_dev);
        }
 
        if (proxy->flags & VIRTIO_PCI_FLAG_INIT_LNKCTL) {
            /* Init Link Control Register */
            pcie_cap_lnkctl_init(pci_dev);
        }
 
        if (proxy->flags & VIRTIO_PCI_FLAG_INIT_PM) {
            /* Init Power Management Control Register */
            pci_set_word(pci_dev->wmask + pos + PCI_PM_CTRL,
                         PCI_PM_CTRL_STATE_MASK);
        }
 
        if (proxy->flags & VIRTIO_PCI_FLAG_ATS) {
            pcie_ats_init(pci_dev, last_pcie_cap_offset,
                          proxy->flags & VIRTIO_PCI_FLAG_ATS_PAGE_ALIGNED);
            last_pcie_cap_offset += PCI_EXT_CAP_ATS_SIZEOF;
        }
 
        if (proxy->flags & VIRTIO_PCI_FLAG_INIT_FLR) {
            /* Set Function Level Reset capability bit */
            pcie_cap_flr_init(pci_dev);
        }
    } else {
        /*
         * make future invocations of pci_is_express() return false
         * and pci_config_size() return PCI_CONFIG_SPACE_SIZE.
         */
        pci_dev->cap_present &= ~QEMU_PCI_CAP_EXPRESS;
    }
 
    virtio_pci_bus_new(&proxy->bus, sizeof(proxy->bus), proxy);
    if (k->realize) {
        k->realize(proxy, errp);
    }
}

在这个函数的最后会调用VirtioPCIClass的realize函数指针所指向的函数，也就是virtio_balloon_pci_realize函数。代码片段如下：

static void virtio_pci_realize(PCIDevice *pci_dev, Error **errp)
{
    VirtIOPCIProxy *proxy = VIRTIO_PCI(pci_dev);
    VirtioPCIClass *k = VIRTIO_PCI_GET_CLASS(pci_dev);
    ……
 
    if (k->realize) {
        k->realize(proxy, errp);
    }
}

更多内容请看下回。