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.

 */

#include "sysemu/blockdev.h"

#include "hw/block-common.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) {

        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, drv->format_name);

        } else {

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

                         file, 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)

{

    BlockdevAction action;

    BlockdevActionList 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(BLOCKDEV_ACTION_KIND_BLOCKDEV_SNAPSHOT_SYNC, &snapshot,

                       errp);

}

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

typedef struct BlkTransactionStates {

    BlockDriverState *old_bs;

    BlockDriverState *new_bs;

    QSIMPLEQ_ENTRY(BlkTransactionStates) entry;

} BlkTransactionStates;

/*

 * 'Atomic' group snapshots.  The snapshots are taken as a set, and if any fail

 *  then we do not pivot any of the devices in the group, and abandon the

 *  snapshots

 */

void qmp_transaction(BlockdevActionList *dev_list, Error **errp)

{

    int ret = 0;

    BlockdevActionList *dev_entry = dev_list;

    BlkTransactionStates *states, *next;

    Error *local_err = NULL;

    QSIMPLEQ_HEAD(snap_bdrv_states, BlkTransactionStates) snap_bdrv_states;

    QSIMPLEQ_INIT(&snap_bdrv_states);

    /* drain all i/o before any snapshots */

    bdrv_drain_all();

    /* We don't do anything in this loop that commits us to the snapshot */

    while (NULL != dev_entry) {

        BlockdevAction *dev_info = NULL;

        BlockDriver *proto_drv;

        BlockDriver *drv;

        int flags;

        enum NewImageMode mode;

        const char *new_image_file;

        const char *device;

        const char *format = "qcow2";

        dev_info = dev_entry->value;

        dev_entry = dev_entry->next;

        states = g_malloc0(sizeof(BlkTransactionStates));

        QSIMPLEQ_INSERT_TAIL(&snap_bdrv_states, states, entry);

        switch (dev_info->kind) {

        case BLOCKDEV_ACTION_KIND_BLOCKDEV_SNAPSHOT_SYNC:

            device = dev_info->blockdev_snapshot_sync->device;

            if (!dev_info->blockdev_snapshot_sync->has_mode) {

                dev_info->blockdev_snapshot_sync->mode = NEW_IMAGE_MODE_ABSOLUTE_PATHS;

            }

            new_image_file = dev_info->blockdev_snapshot_sync->snapshot_file;

            if (dev_info->blockdev_snapshot_sync->has_format) {

                format = dev_info->blockdev_snapshot_sync->format;

            }

            mode = dev_info->blockdev_snapshot_sync->mode;

            break;

        default:

            abort();

        }

        drv = bdrv_find_format(format);

        if (!drv) {

            error_set(errp, QERR_INVALID_BLOCK_FORMAT, format);

            goto delete_and_fail;

        }

        states->old_bs = bdrv_find(device);

        if (!states->old_bs) {

            error_set(errp, QERR_DEVICE_NOT_FOUND, device);

            goto delete_and_fail;

        }

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

            error_set(errp, QERR_DEVICE_HAS_NO_MEDIUM, device);

            goto delete_and_fail;

        }

        if (bdrv_in_use(states->old_bs)) {

            error_set(errp, QERR_DEVICE_IN_USE, device);

            goto delete_and_fail;

        }

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

            if (bdrv_flush(states->old_bs)) {

                error_set(errp, QERR_IO_ERROR);

                goto delete_and_fail;

            }

        }

        flags = states->old_bs->open_flags;

        proto_drv = bdrv_find_protocol(new_image_file);

        if (!proto_drv) {

            error_set(errp, QERR_INVALID_BLOCK_FORMAT, format);

            goto delete_and_fail;

        }

        /* create new image w/backing file */

        if (mode != NEW_IMAGE_MODE_EXISTING) {

            bdrv_img_create(new_image_file, format,

                            states->old_bs->filename,

                            states->old_bs->drv->format_name,

                            NULL, -1, flags, &local_err, false);

            if (error_is_set(&local_err)) {

                error_propagate(errp, local_err);

                goto delete_and_fail;

            }

        }

        /* We will manually add the backing_hd field to the bs later */

        states->new_bs = bdrv_new("");

        /* TODO Inherit bs->options or only take explicit options with an

         * extended QMP command? */

        ret = bdrv_open(states->new_bs, new_image_file, NULL,

                        flags | BDRV_O_NO_BACKING, drv);

        if (ret != 0) {

            error_set(errp, QERR_OPEN_FILE_FAILED, new_image_file);

            goto delete_and_fail;

        }

    }

    /* Now we are going to do the actual pivot.  Everything up to this point

     * is reversible, but we are committed at this point */

    QSIMPLEQ_FOREACH(states, &snap_bdrv_states, entry) {

        /* This removes our old bs from the bdrv_states, and adds the new bs */

        bdrv_append(states->new_bs, states->old_bs);

        /* We don't need (or want) to use the transactional

         * bdrv_reopen_multiple() across all the entries at once, because we

         * don't want to abort all of them if one of them fails the reopen */

        bdrv_reopen(states->new_bs, states->new_bs->open_flags & ~BDRV_O_RDWR,

                    NULL);

    }

    /* success */

    goto exit;

delete_and_fail:

    /*

    * failure, and it is all-or-none; abandon each new bs, and keep using

    * the original bs for all images

    */

    QSIMPLEQ_FOREACH(states, &snap_bdrv_states, entry) {

        if (states->new_bs) {

             bdrv_delete(states->new_bs);

        }

    }

exit:

    QSIMPLEQ_FOREACH_SAFE(states, &snap_bdrv_states, entry, next) {

        g_free(states);

    }

}

static void eject_device(BlockDriverState *bs, int force, Error **errp)

{

    if (bdrv_in_use(bs)) {

        error_set(errp, QERR_DEVICE_IN_USE, bdrv_get_device_name(bs));

        return;

    }

    if (!bdrv_dev_has_removable_media(bs)) {

        error_set(errp, QERR_DEVICE_NOT_REMOVABLE, bdrv_get_device_name(bs));

        return;

    }

    if (bdrv_dev_is_medium_locked(bs) && !bdrv_dev_is_tray_open(bs)) {

        bdrv_dev_eject_request(bs, force);

        if (!force) {

            error_set(errp, QERR_DEVICE_LOCKED, bdrv_get_device_name(bs));

            return;

        }

    }

    bdrv_close(bs);

}

void qmp_eject(const char *device, bool has_force, bool force, Error **errp)

{

    BlockDriverState *bs;

    bs = bdrv_find(device);

    if (!bs) {

        error_set(errp, QERR_DEVICE_NOT_FOUND, device);

        return;

    }

    eject_device(bs, force, errp);

}

void qmp_block_passwd(const char *device, const char *password, Error **errp)

{

    BlockDriverState *bs;

    int err;

    bs = bdrv_find(device);

    if (!bs) {

        error_set(errp, QERR_DEVICE_NOT_FOUND, device);

        return;

    }

    err = bdrv_set_key(bs, password);

    if (err == -EINVAL) {

        error_set(errp, QERR_DEVICE_NOT_ENCRYPTED, bdrv_get_device_name(bs));

        return;

    } else if (err < 0) {

        error_set(errp, QERR_INVALID_PASSWORD);

        return;

    }

}

static void qmp_bdrv_open_encrypted(BlockDriverState *bs, const char *filename,

                                    int bdrv_flags, BlockDriver *drv,

                                    const char *password, Error **errp)

{

    if (bdrv_open(bs, filename, NULL, bdrv_flags, drv) < 0) {

        error_set(errp, QERR_OPEN_FILE_FAILED, filename);

        return;

    }

    if (bdrv_key_required(bs)) {

        if (password) {

            if (bdrv_set_key(bs, password) < 0) {

                error_set(errp, QERR_INVALID_PASSWORD);

            }

        } else {

            error_set(errp, QERR_DEVICE_ENCRYPTED, bdrv_get_device_name(bs),

                      bdrv_get_encrypted_filename(bs));

        }

    } else if (password) {

        error_set(errp, QERR_DEVICE_NOT_ENCRYPTED, bdrv_get_device_name(bs));

    }

}

void qmp_change_blockdev(const char *device, const char *filename,

                         bool has_format, const char *format, Error **errp)

{

    BlockDriverState *bs;

    BlockDriver *drv = NULL;

    int bdrv_flags;

    Error *err = NULL;

    bs = bdrv_find(device);

    if (!bs) {

        error_set(errp, QERR_DEVICE_NOT_FOUND, device);

        return;

    }

    if (format) {

        drv = bdrv_find_whitelisted_format(format);

        if (!drv) {

            error_set(errp, QERR_INVALID_BLOCK_FORMAT, format);

            return;

        }

    }

    eject_device(bs, 0, &err);

    if (error_is_set(&err)) {

        error_propagate(errp, err);

        return;

    }

    bdrv_flags = bdrv_is_read_only(bs) ? 0 : BDRV_O_RDWR;

    bdrv_flags |= bdrv_is_snapshot(bs) ? BDRV_O_SNAPSHOT : 0;

    qmp_bdrv_open_encrypted(bs, filename, bdrv_flags, drv, NULL, errp);

}

/* throttling disk I/O limits */

void qmp_block_set_io_throttle(const char *device, int64_t bps, int64_t bps_rd,

                               int64_t bps_wr, int64_t iops, int64_t iops_rd,

                               int64_t iops_wr, Error **errp)

{

    BlockIOLimit io_limits;

    BlockDriverState *bs;

    bs = bdrv_find(device);

    if (!bs) {

        error_set(errp, QERR_DEVICE_NOT_FOUND, device);

        return;

    }

    io_limits.bps[BLOCK_IO_LIMIT_TOTAL] = bps;

    io_limits.bps[BLOCK_IO_LIMIT_READ]  = bps_rd;

    io_limits.bps[BLOCK_IO_LIMIT_WRITE] = bps_wr;

    io_limits.iops[BLOCK_IO_LIMIT_TOTAL]= iops;

    io_limits.iops[BLOCK_IO_LIMIT_READ] = iops_rd;

    io_limits.iops[BLOCK_IO_LIMIT_WRITE]= iops_wr;

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

        return;

    }

    bs->io_limits = io_limits;

    bs->slice_time = BLOCK_IO_SLICE_TIME;

    if (!bs->io_limits_enabled && bdrv_io_limits_enabled(bs)) {

        bdrv_io_limits_enable(bs);

    } else if (bs->io_limits_enabled && !bdrv_io_limits_enabled(bs)) {

        bdrv_io_limits_disable(bs);

    } else {

        if (bs->block_timer) {

            qemu_mod_timer(bs->block_timer, qemu_get_clock_ns(vm_clock));

        }

    }

}

int do_drive_del(Monitor *mon, const QDict *qdict, QObject **ret_data)

{

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

    BlockDriverState *bs;

    bs = bdrv_find(id);

    if (!bs) {

        qerror_report(QERR_DEVICE_NOT_FOUND, id);

        return -1;

    }

    if (bdrv_in_use(bs)) {

        qerror_report(QERR_DEVICE_IN_USE, id);

        return -1;

    }

    /* quiesce block driver; prevent further io */

    bdrv_drain_all();

    bdrv_flush(bs);

    bdrv_close(bs);

    /* if we have a device attached to this BlockDriverState

     * then we need to make the drive anonymous until the device

     * can be removed.  If this is a drive with no device backing

     * then we can just get rid of the block driver state right here.

     */

    if (bdrv_get_attached_dev(bs)) {

        bdrv_make_anon(bs);

    } else {

        drive_uninit(drive_get_by_blockdev(bs));

    }

    return 0;

}

void qmp_block_resize(const char *device, int64_t size, Error **errp)

{

    BlockDriverState *bs;

    bs = bdrv_find(device);

    if (!bs) {

        error_set(errp, QERR_DEVICE_NOT_FOUND, device);

        return;

    }

    if (size < 0) {

        error_set(errp, QERR_INVALID_PARAMETER_VALUE, "size", "a >0 size");

        return;

    }

    switch (bdrv_truncate(bs, size)) {

    case 0:

        break;

    case -ENOMEDIUM:

        error_set(errp, QERR_DEVICE_HAS_NO_MEDIUM, device);

        break;

    case -ENOTSUP:

        error_set(errp, QERR_UNSUPPORTED);

        break;

    case -EACCES:

        error_set(errp, QERR_DEVICE_IS_READ_ONLY, device);

        break;

    case -EBUSY:

        error_set(errp, QERR_DEVICE_IN_USE, device);

        break;

    default:

        error_set(errp, QERR_UNDEFINED_ERROR);

        break;

    }

}

static void block_job_cb(void *opaque, int ret)

{

    BlockDriverState *bs = opaque;

    QObject *obj;

    trace_block_job_cb(bs, bs->job, ret);

    assert(bs->job);

    obj = qobject_from_block_job(bs->job);

    if (ret < 0) {

        QDict *dict = qobject_to_qdict(obj);

        qdict_put(dict, "error", qstring_from_str(strerror(-ret)));

    }

    if (block_job_is_cancelled(bs->job)) {

        monitor_protocol_event(QEVENT_BLOCK_JOB_CANCELLED, obj);

    } else {

        monitor_protocol_event(QEVENT_BLOCK_JOB_COMPLETED, obj);

    }

    qobject_decref(obj);

    drive_put_ref_bh_schedule(drive_get_by_blockdev(bs));

}

void qmp_block_stream(const char *device, bool has_base,

                      const char *base, bool has_speed, int64_t speed,

                      bool has_on_error, BlockdevOnError on_error,

                      Error **errp)

{

    BlockDriverState *bs;

    BlockDriverState *base_bs = NULL;

    Error *local_err = NULL;

    if (!has_on_error) {

        on_error = BLOCKDEV_ON_ERROR_REPORT;

    }

    bs = bdrv_find(device);

    if (!bs) {

        error_set(errp, QERR_DEVICE_NOT_FOUND, device);

        return;

    }

    if (base) {

        base_bs = bdrv_find_backing_image(bs, base);

        if (base_bs == NULL) {

            error_set(errp, QERR_BASE_NOT_FOUND, base);

            return;

        }

    }

    stream_start(bs, base_bs, base, has_speed ? speed : 0,

                 on_error, block_job_cb, bs, &local_err);

    if (error_is_set(&local_err)) {

        error_propagate(errp, local_err);

        return;

    }

    /* Grab a reference so hotplug does not delete the BlockDriverState from

     * underneath us.

     */

    drive_get_ref(drive_get_by_blockdev(bs));

    trace_qmp_block_stream(bs, bs->job);

}

void qmp_block_commit(const char *device,

                      bool has_base, const char *base, const char *top,

                      bool has_speed, int64_t speed,

                      Error **errp)

{

    BlockDriverState *bs;

    BlockDriverState *base_bs, *top_bs;

    Error *local_err = NULL;

    /* This will be part of the QMP command, if/when the

     * BlockdevOnError change for blkmirror makes it in

     */

    BlockdevOnError on_error = BLOCKDEV_ON_ERROR_REPORT;

    /* drain all i/o before commits */

    bdrv_drain_all();

    bs = bdrv_find(device);

    if (!bs) {

        error_set(errp, QERR_DEVICE_NOT_FOUND, device);

        return;

    }

    /* default top_bs is the active layer */

    top_bs = bs;

    if (top) {

        if (strcmp(bs->filename, top) != 0) {

            top_bs = bdrv_find_backing_image(bs, top);

        }

    }

    if (top_bs == NULL) {

        error_setg(errp, "Top image file %s not found", top ? top : "NULL");

        return;

    }

    if (has_base && base) {

        base_bs = bdrv_find_backing_image(top_bs, base);

    } else {

        base_bs = bdrv_find_base(top_bs);

    }

    if (base_bs == NULL) {

        error_set(errp, QERR_BASE_NOT_FOUND, base ? base : "NULL");

        return;

    }

    commit_start(bs, base_bs, top_bs, speed, on_error, block_job_cb, bs,

                &local_err);

    if (local_err != NULL) {

        error_propagate(errp, local_err);

        return;

    }

    /* Grab a reference so hotplug does not delete the BlockDriverState from

     * underneath us.

     */

    drive_get_ref(drive_get_by_blockdev(bs));

}

#define DEFAULT_MIRROR_BUF_SIZE   (10 << 20)

void qmp_drive_mirror(const char *device, const char *target,

                      bool has_format, const char *format,

                      enum MirrorSyncMode sync,

                      bool has_mode, enum NewImageMode mode,

                      bool has_speed, int64_t speed,

                      bool has_granularity, uint32_t granularity,

                      bool has_buf_size, int64_t buf_size,

                      bool has_on_source_error, BlockdevOnError on_source_error,

                      bool has_on_target_error, BlockdevOnError on_target_error,

                      Error **errp)

{

    BlockDriverState *bs;

    BlockDriverState *source, *target_bs;

    BlockDriver *proto_drv;

    BlockDriver *drv = NULL;

    Error *local_err = NULL;

    int flags;

    uint64_t size;

    int ret;

    if (!has_speed) {

        speed = 0;

    }

    if (!has_on_source_error) {

        on_source_error = BLOCKDEV_ON_ERROR_REPORT;

    }

    if (!has_on_target_error) {

        on_target_error = BLOCKDEV_ON_ERROR_REPORT;

    }

    if (!has_mode) {

        mode = NEW_IMAGE_MODE_ABSOLUTE_PATHS;

    }

    if (!has_granularity) {

        granularity = 0;

    }

    if (!has_buf_size) {

        buf_size = DEFAULT_MIRROR_BUF_SIZE;

    }

    if (granularity != 0 && (granularity < 512 || granularity > 1048576 * 64)) {

        error_set(errp, QERR_INVALID_PARAMETER, device);

        return;

    }

    if (granularity & (granularity - 1)) {

        error_set(errp, QERR_INVALID_PARAMETER, device);

        return;

    }

    bs = bdrv_find(device);

    if (!bs) {

        error_set(errp, QERR_DEVICE_NOT_FOUND, device);

        return;

    }

    if (!bdrv_is_inserted(bs)) {

        error_set(errp, QERR_DEVICE_HAS_NO_MEDIUM, device);

        return;

    }

    if (!has_format) {

        format = mode == NEW_IMAGE_MODE_EXISTING ? NULL : bs->drv->format_name;

    }

    if (format) {

        drv = bdrv_find_format(format);

        if (!drv) {

            error_set(errp, QERR_INVALID_BLOCK_FORMAT, format);

            return;

        }

    }

    if (bdrv_in_use(bs)) {

        error_set(errp, QERR_DEVICE_IN_USE, device);

        return;

    }

    flags = bs->open_flags | BDRV_O_RDWR;

    source = bs->backing_hd;

    if (!source && sync == MIRROR_SYNC_MODE_TOP) {

        sync = MIRROR_SYNC_MODE_FULL;

    }