Archipelago » qemu

/*

 * QEMU host block devices

 *

 * Copyright (c) 2003-2008 Fabrice Bellard

 *

 * This work is licensed under the terms of the GNU GPL, version 2 or

 * later.  See the COPYING file in the top-level directory.

 *

 * This file incorporates work covered by the following copyright and

 * permission notice:

 *

 * Copyright (c) 2003-2008 Fabrice Bellard

 *

 * Permission is hereby granted, free of charge, to any person obtaining a copy

 * of this software and associated documentation files (the "Software"), to deal

 * in the Software without restriction, including without limitation the rights

 * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell

 * copies of the Software, and to permit persons to whom the Software is

 * furnished to do so, subject to the following conditions:

 *

 * The above copyright notice and this permission notice shall be included in

 * all copies or substantial portions of the Software.

 *

 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR

 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,

 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL

 * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER

 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,

 * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN

 * THE SOFTWARE.

 */

#include "sysemu/blockdev.h"

#include "hw/block/block.h"

#include "block/blockjob.h"

#include "monitor/monitor.h"

#include "qapi/qmp/qerror.h"

#include "qemu/option.h"

#include "qemu/config-file.h"

#include "qapi/qmp/types.h"

#include "sysemu/sysemu.h"

#include "block/block_int.h"

#include "qmp-commands.h"

#include "trace.h"

#include "sysemu/arch_init.h"

static QTAILQ_HEAD(drivelist, DriveInfo) drives = QTAILQ_HEAD_INITIALIZER(drives);

extern QemuOptsList qemu_common_drive_opts;

static const char *const if_name[IF_COUNT] = {

    [IF_NONE] = "none",

    [IF_IDE] = "ide",

    [IF_SCSI] = "scsi",

    [IF_FLOPPY] = "floppy",

    [IF_PFLASH] = "pflash",

    [IF_MTD] = "mtd",

    [IF_SD] = "sd",

    [IF_VIRTIO] = "virtio",

    [IF_XEN] = "xen",

};

static const int if_max_devs[IF_COUNT] = {

    /*

     * Do not change these numbers!  They govern how drive option

     * index maps to unit and bus.  That mapping is ABI.

     *

     * All controllers used to imlement if=T drives need to support

     * if_max_devs[T] units, for any T with if_max_devs[T] != 0.

     * Otherwise, some index values map to "impossible" bus, unit

     * values.

     *

     * For instance, if you change [IF_SCSI] to 255, -drive

     * if=scsi,index=12 no longer means bus=1,unit=5, but

     * bus=0,unit=12.  With an lsi53c895a controller (7 units max),

     * the drive can't be set up.  Regression.

     */

    [IF_IDE] = 2,

    [IF_SCSI] = 7,

};

/*

 * We automatically delete the drive when a device using it gets

 * unplugged.  Questionable feature, but we can't just drop it.

 * Device models call blockdev_mark_auto_del() to schedule the

 * automatic deletion, and generic qdev code calls blockdev_auto_del()

 * when deletion is actually safe.

 */

void blockdev_mark_auto_del(BlockDriverState *bs)

{

    DriveInfo *dinfo = drive_get_by_blockdev(bs);

    if (bs->job) {

        block_job_cancel(bs->job);

    }

    if (dinfo) {

        dinfo->auto_del = 1;

    }

}

void blockdev_auto_del(BlockDriverState *bs)

{

    DriveInfo *dinfo = drive_get_by_blockdev(bs);

    if (dinfo && dinfo->auto_del) {

        drive_put_ref(dinfo);

    }

}

static int drive_index_to_bus_id(BlockInterfaceType type, int index)

{

    int max_devs = if_max_devs[type];

    return max_devs ? index / max_devs : 0;

}

static int drive_index_to_unit_id(BlockInterfaceType type, int index)

{

    int max_devs = if_max_devs[type];

    return max_devs ? index % max_devs : index;

}

QemuOpts *drive_def(const char *optstr)

{

    return qemu_opts_parse(qemu_find_opts("drive"), optstr, 0);

}

QemuOpts *drive_add(BlockInterfaceType type, int index, const char *file,

                    const char *optstr)

{

    QemuOpts *opts;

    char buf[32];

    opts = drive_def(optstr);

    if (!opts) {

        return NULL;

    }

    if (type != IF_DEFAULT) {

        qemu_opt_set(opts, "if", if_name[type]);

    }

    if (index >= 0) {

        snprintf(buf, sizeof(buf), "%d", index);

        qemu_opt_set(opts, "index", buf);

    }

    if (file)

        qemu_opt_set(opts, "file", file);

    return opts;

}

DriveInfo *drive_get(BlockInterfaceType type, int bus, int unit)

{

    DriveInfo *dinfo;

    /* seek interface, bus and unit */

    QTAILQ_FOREACH(dinfo, &drives, next) {

        if (dinfo->type == type &&

            dinfo->bus == bus &&

            dinfo->unit == unit)

            return dinfo;

    }

    return NULL;

}

DriveInfo *drive_get_by_index(BlockInterfaceType type, int index)

{

    return drive_get(type,

                     drive_index_to_bus_id(type, index),

                     drive_index_to_unit_id(type, index));

}

int drive_get_max_bus(BlockInterfaceType type)

{

    int max_bus;

    DriveInfo *dinfo;

    max_bus = -1;

    QTAILQ_FOREACH(dinfo, &drives, next) {

        if(dinfo->type == type &&

           dinfo->bus > max_bus)

            max_bus = dinfo->bus;

    }

    return max_bus;

}

/* Get a block device.  This should only be used for single-drive devices

   (e.g. SD/Floppy/MTD).  Multi-disk devices (scsi/ide) should use the

   appropriate bus.  */

DriveInfo *drive_get_next(BlockInterfaceType type)

{

    static int next_block_unit[IF_COUNT];

    return drive_get(type, 0, next_block_unit[type]++);

}

DriveInfo *drive_get_by_blockdev(BlockDriverState *bs)

{

    DriveInfo *dinfo;

    QTAILQ_FOREACH(dinfo, &drives, next) {

        if (dinfo->bdrv == bs) {

            return dinfo;

        }

    }

    return NULL;

}

static void bdrv_format_print(void *opaque, const char *name)

{

    error_printf(" %s", name);

}

static void drive_uninit(DriveInfo *dinfo)

{

    qemu_opts_del(dinfo->opts);

    bdrv_delete(dinfo->bdrv);

    g_free(dinfo->id);

    QTAILQ_REMOVE(&drives, dinfo, next);

    g_free(dinfo->serial);

    g_free(dinfo);

}

void drive_put_ref(DriveInfo *dinfo)

{

    assert(dinfo->refcount);

    if (--dinfo->refcount == 0) {

        drive_uninit(dinfo);

    }

}

void drive_get_ref(DriveInfo *dinfo)

{

    dinfo->refcount++;

}

typedef struct {

    QEMUBH *bh;

    DriveInfo *dinfo;

} DrivePutRefBH;

static void drive_put_ref_bh(void *opaque)

{

    DrivePutRefBH *s = opaque;

    drive_put_ref(s->dinfo);

    qemu_bh_delete(s->bh);

    g_free(s);

}

/*

 * Release a drive reference in a BH

 *

 * It is not possible to use drive_put_ref() from a callback function when the

 * callers still need the drive.  In such cases we schedule a BH to release the

 * reference.

 */

static void drive_put_ref_bh_schedule(DriveInfo *dinfo)

{

    DrivePutRefBH *s;

    s = g_new(DrivePutRefBH, 1);

    s->bh = qemu_bh_new(drive_put_ref_bh, s);

    s->dinfo = dinfo;

    qemu_bh_schedule(s->bh);

}

static int parse_block_error_action(const char *buf, bool is_read)

{

    if (!strcmp(buf, "ignore")) {

        return BLOCKDEV_ON_ERROR_IGNORE;

    } else if (!is_read && !strcmp(buf, "enospc")) {

        return BLOCKDEV_ON_ERROR_ENOSPC;

    } else if (!strcmp(buf, "stop")) {

        return BLOCKDEV_ON_ERROR_STOP;

    } else if (!strcmp(buf, "report")) {

        return BLOCKDEV_ON_ERROR_REPORT;

    } else {

        error_report("'%s' invalid %s error action",

                     buf, is_read ? "read" : "write");

        return -1;

    }

}

static bool do_check_io_limits(BlockIOLimit *io_limits, Error **errp)

{

    bool bps_flag;

    bool iops_flag;

    assert(io_limits);

    bps_flag  = (io_limits->bps[BLOCK_IO_LIMIT_TOTAL] != 0)

                 && ((io_limits->bps[BLOCK_IO_LIMIT_READ] != 0)

                 || (io_limits->bps[BLOCK_IO_LIMIT_WRITE] != 0));

    iops_flag = (io_limits->iops[BLOCK_IO_LIMIT_TOTAL] != 0)

                 && ((io_limits->iops[BLOCK_IO_LIMIT_READ] != 0)

                 || (io_limits->iops[BLOCK_IO_LIMIT_WRITE] != 0));

    if (bps_flag || iops_flag) {

        error_setg(errp, "bps(iops) and bps_rd/bps_wr(iops_rd/iops_wr) "

                         "cannot be used at the same time");

        return false;

    }

    if (io_limits->bps[BLOCK_IO_LIMIT_TOTAL] < 0 ||

        io_limits->bps[BLOCK_IO_LIMIT_WRITE] < 0 ||

        io_limits->bps[BLOCK_IO_LIMIT_READ] < 0 ||

        io_limits->iops[BLOCK_IO_LIMIT_TOTAL] < 0 ||

        io_limits->iops[BLOCK_IO_LIMIT_WRITE] < 0 ||

        io_limits->iops[BLOCK_IO_LIMIT_READ] < 0) {

        error_setg(errp, "bps and iops values must be 0 or greater");

        return false;

    }

    return true;

}

DriveInfo *drive_init(QemuOpts *all_opts, BlockInterfaceType block_default_type)

{

    const char *buf;

    const char *file = NULL;

    const char *serial;

    const char *mediastr = "";

    BlockInterfaceType type;

    enum { MEDIA_DISK, MEDIA_CDROM } media;

    int bus_id, unit_id;

    int cyls, heads, secs, translation;

    BlockDriver *drv = NULL;

    int max_devs;

    int index;

    int ro = 0;

    int bdrv_flags = 0;

    int on_read_error, on_write_error;

    const char *devaddr;

    DriveInfo *dinfo;

    BlockIOLimit io_limits;

    int snapshot = 0;

    bool copy_on_read;

    int ret;

    Error *error = NULL;

    QemuOpts *opts;

    QDict *bs_opts;

    const char *id;

    translation = BIOS_ATA_TRANSLATION_AUTO;

    media = MEDIA_DISK;

    /* Check common options by copying from all_opts to opts, all other options

     * are stored in bs_opts. */

    id = qemu_opts_id(all_opts);

    opts = qemu_opts_create(&qemu_common_drive_opts, id, 1, &error);

    if (error_is_set(&error)) {

        qerror_report_err(error);

        error_free(error);

        return NULL;

    }

    bs_opts = qdict_new();

    qemu_opts_to_qdict(all_opts, bs_opts);

    qemu_opts_absorb_qdict(opts, bs_opts, &error);

    if (error_is_set(&error)) {

        qerror_report_err(error);

        error_free(error);

        return NULL;

    }

    if (id) {

        qdict_del(bs_opts, "id");

    }

    /* extract parameters */

    bus_id  = qemu_opt_get_number(opts, "bus", 0);

    unit_id = qemu_opt_get_number(opts, "unit", -1);

    index   = qemu_opt_get_number(opts, "index", -1);

    cyls  = qemu_opt_get_number(opts, "cyls", 0);

    heads = qemu_opt_get_number(opts, "heads", 0);

    secs  = qemu_opt_get_number(opts, "secs", 0);

    snapshot = qemu_opt_get_bool(opts, "snapshot", 0);

    ro = qemu_opt_get_bool(opts, "readonly", 0);

    copy_on_read = qemu_opt_get_bool(opts, "copy-on-read", false);

    file = qemu_opt_get(opts, "file");

    serial = qemu_opt_get(opts, "serial");

    if ((buf = qemu_opt_get(opts, "if")) != NULL) {

        for (type = 0; type < IF_COUNT && strcmp(buf, if_name[type]); type++)

            ;

        if (type == IF_COUNT) {

            error_report("unsupported bus type '%s'", buf);

            return NULL;

        }

    } else {

        type = block_default_type;

    }

    max_devs = if_max_devs[type];

    if (cyls || heads || secs) {

        if (cyls < 1) {

            error_report("invalid physical cyls number");

            return NULL;

        }

        if (heads < 1) {

            error_report("invalid physical heads number");

            return NULL;

        }

        if (secs < 1) {

            error_report("invalid physical secs number");

            return NULL;

        }

    }

    if ((buf = qemu_opt_get(opts, "trans")) != NULL) {

        if (!cyls) {

            error_report("'%s' trans must be used with cyls, heads and secs",

                         buf);

            return NULL;

        }

        if (!strcmp(buf, "none"))

            translation = BIOS_ATA_TRANSLATION_NONE;

        else if (!strcmp(buf, "lba"))

            translation = BIOS_ATA_TRANSLATION_LBA;

        else if (!strcmp(buf, "auto"))

            translation = BIOS_ATA_TRANSLATION_AUTO;

        else {

            error_report("'%s' invalid translation type", buf);

            return NULL;

        }

    }

    if ((buf = qemu_opt_get(opts, "media")) != NULL) {

        if (!strcmp(buf, "disk")) {

            media = MEDIA_DISK;

        } else if (!strcmp(buf, "cdrom")) {

            if (cyls || secs || heads) {

                error_report("CHS can't be set with media=%s", buf);

                return NULL;

            }

            media = MEDIA_CDROM;

        } else {

            error_report("'%s' invalid media", buf);

            return NULL;

        }

    }

    if ((buf = qemu_opt_get(opts, "discard")) != NULL) {

        if (bdrv_parse_discard_flags(buf, &bdrv_flags) != 0) {

            error_report("invalid discard option");

            return NULL;

        }

    }

    bdrv_flags |= BDRV_O_CACHE_WB;

    if ((buf = qemu_opt_get(opts, "cache")) != NULL) {

        if (bdrv_parse_cache_flags(buf, &bdrv_flags) != 0) {

            error_report("invalid cache option");

            return NULL;

        }

    }

#ifdef CONFIG_LINUX_AIO

    if ((buf = qemu_opt_get(opts, "aio")) != NULL) {

        if (!strcmp(buf, "native")) {

            bdrv_flags |= BDRV_O_NATIVE_AIO;

        } else if (!strcmp(buf, "threads")) {

            /* this is the default */

        } else {

           error_report("invalid aio option");

           return NULL;

        }

    }

#endif

    if ((buf = qemu_opt_get(opts, "format")) != NULL) {

        if (is_help_option(buf)) {

            error_printf("Supported formats:");

            bdrv_iterate_format(bdrv_format_print, NULL);

            error_printf("\n");

            return NULL;

        }

        drv = bdrv_find_whitelisted_format(buf);

        if (!drv) {

            error_report("'%s' invalid format", buf);

            return NULL;

        }

    }

    /* disk I/O throttling */

    io_limits.bps[BLOCK_IO_LIMIT_TOTAL]  =

                           qemu_opt_get_number(opts, "bps", 0);

    io_limits.bps[BLOCK_IO_LIMIT_READ]   =

                           qemu_opt_get_number(opts, "bps_rd", 0);

    io_limits.bps[BLOCK_IO_LIMIT_WRITE]  =

                           qemu_opt_get_number(opts, "bps_wr", 0);

    io_limits.iops[BLOCK_IO_LIMIT_TOTAL] =

                           qemu_opt_get_number(opts, "iops", 0);

    io_limits.iops[BLOCK_IO_LIMIT_READ]  =

                           qemu_opt_get_number(opts, "iops_rd", 0);

    io_limits.iops[BLOCK_IO_LIMIT_WRITE] =

                           qemu_opt_get_number(opts, "iops_wr", 0);

    if (!do_check_io_limits(&io_limits, &error)) {

        error_report("%s", error_get_pretty(error));

        error_free(error);

        return NULL;

    }

    if (qemu_opt_get(opts, "boot") != NULL) {

        fprintf(stderr, "qemu-kvm: boot=on|off is deprecated and will be "

                "ignored. Future versions will reject this parameter. Please "

                "update your scripts.\n");

    }

    on_write_error = BLOCKDEV_ON_ERROR_ENOSPC;

    if ((buf = qemu_opt_get(opts, "werror")) != NULL) {

        if (type != IF_IDE && type != IF_SCSI && type != IF_VIRTIO && type != IF_NONE) {

            error_report("werror is not supported by this bus type");

            return NULL;

        }

        on_write_error = parse_block_error_action(buf, 0);

        if (on_write_error < 0) {

            return NULL;

        }

    }

    on_read_error = BLOCKDEV_ON_ERROR_REPORT;

    if ((buf = qemu_opt_get(opts, "rerror")) != NULL) {

        if (type != IF_IDE && type != IF_VIRTIO && type != IF_SCSI && type != IF_NONE) {

            error_report("rerror is not supported by this bus type");

            return NULL;

        }

        on_read_error = parse_block_error_action(buf, 1);

        if (on_read_error < 0) {

            return NULL;

        }

    }

    if ((devaddr = qemu_opt_get(opts, "addr")) != NULL) {

        if (type != IF_VIRTIO) {

            error_report("addr is not supported by this bus type");

            return NULL;

        }

    }

    /* compute bus and unit according index */

    if (index != -1) {

        if (bus_id != 0 || unit_id != -1) {

            error_report("index cannot be used with bus and unit");

            return NULL;

        }

        bus_id = drive_index_to_bus_id(type, index);

        unit_id = drive_index_to_unit_id(type, index);

    }

    /* if user doesn't specify a unit_id,

     * try to find the first free

     */

    if (unit_id == -1) {

       unit_id = 0;

       while (drive_get(type, bus_id, unit_id) != NULL) {

           unit_id++;

           if (max_devs && unit_id >= max_devs) {

               unit_id -= max_devs;

               bus_id++;

           }

       }

    }

    /* check unit id */

    if (max_devs && unit_id >= max_devs) {

        error_report("unit %d too big (max is %d)",

                     unit_id, max_devs - 1);

        return NULL;

    }

    /*

     * catch multiple definitions

     */

    if (drive_get(type, bus_id, unit_id) != NULL) {

        error_report("drive with bus=%d, unit=%d (index=%d) exists",

                     bus_id, unit_id, index);

        return NULL;

    }

    /* init */

    dinfo = g_malloc0(sizeof(*dinfo));

    if ((buf = qemu_opts_id(opts)) != NULL) {

        dinfo->id = g_strdup(buf);

    } else {

        /* no id supplied -> create one */

        dinfo->id = g_malloc0(32);

        if (type == IF_IDE || type == IF_SCSI)

            mediastr = (media == MEDIA_CDROM) ? "-cd" : "-hd";

        if (max_devs)

            snprintf(dinfo->id, 32, "%s%i%s%i",

                     if_name[type], bus_id, mediastr, unit_id);

        else

            snprintf(dinfo->id, 32, "%s%s%i",

                     if_name[type], mediastr, unit_id);

    }

    dinfo->bdrv = bdrv_new(dinfo->id);

    dinfo->bdrv->open_flags = snapshot ? BDRV_O_SNAPSHOT : 0;

    dinfo->bdrv->read_only = ro;

    dinfo->devaddr = devaddr;

    dinfo->type = type;

    dinfo->bus = bus_id;

    dinfo->unit = unit_id;

    dinfo->cyls = cyls;

    dinfo->heads = heads;

    dinfo->secs = secs;

    dinfo->trans = translation;

    dinfo->opts = all_opts;

    dinfo->refcount = 1;

    if (serial != NULL) {

        dinfo->serial = g_strdup(serial);

    }

    QTAILQ_INSERT_TAIL(&drives, dinfo, next);

    bdrv_set_on_error(dinfo->bdrv, on_read_error, on_write_error);

    /* disk I/O throttling */

    bdrv_set_io_limits(dinfo->bdrv, &io_limits);

    switch(type) {

    case IF_IDE:

    case IF_SCSI:

    case IF_XEN:

    case IF_NONE:

        dinfo->media_cd = media == MEDIA_CDROM;

        break;

    case IF_SD:

    case IF_FLOPPY:

    case IF_PFLASH:

    case IF_MTD:

        break;

    case IF_VIRTIO:

    {

        /* add virtio block device */

        QemuOpts *devopts;

        devopts = qemu_opts_create_nofail(qemu_find_opts("device"));

        if (arch_type == QEMU_ARCH_S390X) {

            qemu_opt_set(devopts, "driver", "virtio-blk-s390");

        } else {

            qemu_opt_set(devopts, "driver", "virtio-blk-pci");

        }

        qemu_opt_set(devopts, "drive", dinfo->id);

        if (devaddr)

            qemu_opt_set(devopts, "addr", devaddr);

        break;

    }

    default:

        abort();

    }

    if (!file || !*file) {

        if (qdict_size(bs_opts)) {

            file = NULL;

        } else {

            return dinfo;

        }

    }

    if (snapshot) {

        /* always use cache=unsafe with snapshot */

        bdrv_flags &= ~BDRV_O_CACHE_MASK;

        bdrv_flags |= (BDRV_O_SNAPSHOT|BDRV_O_CACHE_WB|BDRV_O_NO_FLUSH);

    }

    if (copy_on_read) {

        bdrv_flags |= BDRV_O_COPY_ON_READ;

    }

    if (runstate_check(RUN_STATE_INMIGRATE)) {

        bdrv_flags |= BDRV_O_INCOMING;

    }

    if (media == MEDIA_CDROM) {

        /* CDROM is fine for any interface, don't check.  */

        ro = 1;

    } else if (ro == 1) {

        if (type != IF_SCSI && type != IF_VIRTIO && type != IF_FLOPPY &&

            type != IF_NONE && type != IF_PFLASH) {

            error_report("readonly not supported by this bus type");

            goto err;

        }

    }

    bdrv_flags |= ro ? 0 : BDRV_O_RDWR;

    if (ro && copy_on_read) {

        error_report("warning: disabling copy_on_read on readonly drive");

    }

    ret = bdrv_open(dinfo->bdrv, file, bs_opts, bdrv_flags, drv);

    bs_opts = NULL;

    if (ret < 0) {

        if (ret == -EMEDIUMTYPE) {

            error_report("could not open disk image %s: not in %s format",

                         file ?: dinfo->id, drv->format_name);

        } else {

            error_report("could not open disk image %s: %s",

                         file ?: dinfo->id, strerror(-ret));

        }

        goto err;

    }

    if (bdrv_key_required(dinfo->bdrv))

        autostart = 0;

    qemu_opts_del(opts);

    return dinfo;

err:

    qemu_opts_del(opts);

    QDECREF(bs_opts);

    bdrv_delete(dinfo->bdrv);

    g_free(dinfo->id);

    QTAILQ_REMOVE(&drives, dinfo, next);

    g_free(dinfo);

    return NULL;

}

void do_commit(Monitor *mon, const QDict *qdict)

{

    const char *device = qdict_get_str(qdict, "device");

    BlockDriverState *bs;

    int ret;

    if (!strcmp(device, "all")) {

        ret = bdrv_commit_all();

    } else {

        bs = bdrv_find(device);

        if (!bs) {

            monitor_printf(mon, "Device '%s' not found\n", device);

            return;

        }

        ret = bdrv_commit(bs);

    }

    if (ret < 0) {

        monitor_printf(mon, "'commit' error for '%s': %s\n", device,

                       strerror(-ret));

    }

}

static void blockdev_do_action(int kind, void *data, Error **errp)

{

    TransactionAction action;

    TransactionActionList list;

    action.kind = kind;

    action.data = data;

    list.value = &action;

    list.next = NULL;

    qmp_transaction(&list, errp);

}

void qmp_blockdev_snapshot_sync(const char *device, const char *snapshot_file,

                                bool has_format, const char *format,

                                bool has_mode, enum NewImageMode mode,

                                Error **errp)

{

    BlockdevSnapshot snapshot = {

        .device = (char *) device,

        .snapshot_file = (char *) snapshot_file,

        .has_format = has_format,

        .format = (char *) format,

        .has_mode = has_mode,

        .mode = mode,

    };

    blockdev_do_action(TRANSACTION_ACTION_KIND_BLOCKDEV_SNAPSHOT_SYNC,

                       &snapshot, errp);

}

/* New and old BlockDriverState structs for group snapshots */

typedef struct BlkTransactionStates BlkTransactionStates;

/* Only prepare() may fail. In a single transaction, only one of commit() or

   abort() will be called, clean() will always be called if it present. */

typedef struct BdrvActionOps {

    /* Size of state struct, in bytes. */

    size_t instance_size;

    /* Prepare the work, must NOT be NULL. */

    void (*prepare)(BlkTransactionStates *common, Error **errp);

    /* Commit the changes, must NOT be NULL. */

    void (*commit)(BlkTransactionStates *common);

    /* Abort the changes on fail, can be NULL. */

    void (*abort)(BlkTransactionStates *common);

    /* Clean up resource in the end, can be NULL. */

    void (*clean)(BlkTransactionStates *common);

} BdrvActionOps;

/*

 * This structure must be arranged as first member in child type, assuming

 * that compiler will also arrange it to the same address with parent instance.

 * Later it will be used in free().

 */

struct BlkTransactionStates {

    TransactionAction *action;

    const BdrvActionOps *ops;

    QSIMPLEQ_ENTRY(BlkTransactionStates) entry;

};

/* external snapshot private data */

typedef struct ExternalSnapshotStates {

    BlkTransactionStates common;

    BlockDriverState *old_bs;

    BlockDriverState *new_bs;

} ExternalSnapshotStates;

static void external_snapshot_prepare(BlkTransactionStates *common,

                                      Error **errp)

{

    BlockDriver *proto_drv;

    BlockDriver *drv;

    int flags, ret;

    Error *local_err = NULL;

    const char *device;

    const char *new_image_file;

    const char *format = "qcow2";

    enum NewImageMode mode = NEW_IMAGE_MODE_ABSOLUTE_PATHS;

    ExternalSnapshotStates *states =

                             DO_UPCAST(ExternalSnapshotStates, common, common);

    TransactionAction *action = common->action;

    /* get parameters */

    g_assert(action->kind == TRANSACTION_ACTION_KIND_BLOCKDEV_SNAPSHOT_SYNC);

    device = action->blockdev_snapshot_sync->device;

    new_image_file = action->blockdev_snapshot_sync->snapshot_file;

    if (action->blockdev_snapshot_sync->has_format) {

        format = action->blockdev_snapshot_sync->format;

    }

    if (action->blockdev_snapshot_sync->has_mode) {

        mode = action->blockdev_snapshot_sync->mode;

    }

    /* start processing */

    drv = bdrv_find_format(format);

    if (!drv) {

        error_set(errp, QERR_INVALID_BLOCK_FORMAT, format);

        return;

    }

    states->old_bs = bdrv_find(device);

    if (!states->old_bs) {

        error_set(errp, QERR_DEVICE_NOT_FOUND, device);

        return;

    }

    if (!bdrv_is_inserted(states->old_bs)) {

        error_set(errp, QERR_DEVICE_HAS_NO_MEDIUM, device);

        return;

    }

    if (bdrv_in_use(states->old_bs)) {

        error_set(errp, QERR_DEVICE_IN_USE, device);

        return;

    }

    if (!bdrv_is_read_only(states->old_bs)) {