block.c

/*
 * QEMU System Emulator block driver
 *
 * Copyright (c) 2003 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 "qemu/osdep.h"
#include "block/trace.h"
#include "block/block_int.h"
#include "block/blockjob.h"
#include "block/nbd.h"
#include "qemu/error-report.h"
#include "module_block.h"
#include "qemu/module.h"
#include "qapi/qmp/qerror.h"
#include "qapi/qmp/qbool.h"
#include "qapi/qmp/qjson.h"
#include "sysemu/block-backend.h"
#include "sysemu/sysemu.h"
#include "qemu/notify.h"
#include "qemu/coroutine.h"
#include "block/qapi.h"
#include "qmp-commands.h"
#include "qemu/timer.h"
#include "qapi-event.h"
#include "qemu/cutils.h"
#include "qemu/id.h"
#include "qapi/util.h"

#ifdef CONFIG_BSD
#include <sys/ioctl.h>
#include <sys/queue.h>
#ifndef __DragonFly__
#include <sys/disk.h>
#endif
#endif

#ifdef _WIN32
#include <windows.h>
#endif

#define NOT_DONE 0x7fffffff /* used while emulated sync operation in progress */

static QTAILQ_HEAD(, BlockDriverState) graph_bdrv_states =
    QTAILQ_HEAD_INITIALIZER(graph_bdrv_states);

static QTAILQ_HEAD(, BlockDriverState) all_bdrv_states =
    QTAILQ_HEAD_INITIALIZER(all_bdrv_states);

static QLIST_HEAD(, BlockDriver) bdrv_drivers =
    QLIST_HEAD_INITIALIZER(bdrv_drivers);

static BlockDriverState *bdrv_open_inherit(const char *filename,
                                           const char *reference,
                                           QDict *options, int flags,
                                           BlockDriverState *parent,
                                           const BdrvChildRole *child_role,
                                           Error **errp);

/* If non-zero, use only whitelisted block drivers */
static int use_bdrv_whitelist;

#ifdef _WIN32
static int is_windows_drive_prefix(const char *filename)
{
    return (((filename[0] >= 'a' && filename[0] <= 'z') ||
             (filename[0] >= 'A' && filename[0] <= 'Z')) &&
            filename[1] == ':');
}

int is_windows_drive(const char *filename)
{
    if (is_windows_drive_prefix(filename) &&
        filename[2] == '\0')
        return 1;
    if (strstart(filename, "\\\\.\\", NULL) ||
        strstart(filename, "//./", NULL))
        return 1;
    return 0;
}
#endif

size_t bdrv_opt_mem_align(BlockDriverState *bs)
{
    if (!bs || !bs->drv) {
        /* page size or 4k (hdd sector size) should be on the safe side */
        return MAX(4096, getpagesize());
    }

    return bs->bl.opt_mem_alignment;
}

size_t bdrv_min_mem_align(BlockDriverState *bs)
{
    if (!bs || !bs->drv) {
        /* page size or 4k (hdd sector size) should be on the safe side */
        return MAX(4096, getpagesize());
    }

    return bs->bl.min_mem_alignment;
}

/* check if the path starts with "<protocol>:" */
int path_has_protocol(const char *path)
{
    const char *p;

#ifdef _WIN32
    if (is_windows_drive(path) ||
        is_windows_drive_prefix(path)) {
        return 0;
    }
    p = path + strcspn(path, ":/\\");
#else
    p = path + strcspn(path, ":/");
#endif

    return *p == ':';
}

int path_is_absolute(const char *path)
{
#ifdef _WIN32
    /* specific case for names like: "\\.\d:" */
    if (is_windows_drive(path) || is_windows_drive_prefix(path)) {
        return 1;
    }
    return (*path == '/' || *path == '\\');
#else
    return (*path == '/');
#endif
}

/* if filename is absolute, just copy it to dest. Otherwise, build a
   path to it by considering it is relative to base_path. URL are
   supported. */
void path_combine(char *dest, int dest_size,
                  const char *base_path,
                  const char *filename)
{
    const char *p, *p1;
    int len;

    if (dest_size <= 0)
        return;
    if (path_is_absolute(filename)) {
        pstrcpy(dest, dest_size, filename);
    } else {
        p = strchr(base_path, ':');
        if (p)
            p++;
        else
            p = base_path;
        p1 = strrchr(base_path, '/');
#ifdef _WIN32
        {
            const char *p2;
            p2 = strrchr(base_path, '\\');
            if (!p1 || p2 > p1)
                p1 = p2;
        }
#endif
        if (p1)
            p1++;
        else
            p1 = base_path;
        if (p1 > p)
            p = p1;
        len = p - base_path;
        if (len > dest_size - 1)
            len = dest_size - 1;
        memcpy(dest, base_path, len);
        dest[len] = '\0';
        pstrcat(dest, dest_size, filename);
    }
}

/* Returns whether the image file is opened as read-only. Note that this can
 * return false and writing to the image file is still not possible because the
 * image is inactivated. */
bool bdrv_is_read_only(BlockDriverState *bs)
{
    return bs->read_only;
}

/* Returns whether the image file can be written to right now */
bool bdrv_is_writable(BlockDriverState *bs)
{
    return !bdrv_is_read_only(bs) && !(bs->open_flags & BDRV_O_INACTIVE);
}

int bdrv_can_set_read_only(BlockDriverState *bs, bool read_only, Error **errp)
{
    /* Do not set read_only if copy_on_read is enabled */
    if (bs->copy_on_read && read_only) {
        error_setg(errp, "Can't set node '%s' to r/o with copy-on-read enabled",
                   bdrv_get_device_or_node_name(bs));
        return -EINVAL;
    }

    /* Do not clear read_only if it is prohibited */
    if (!read_only && !(bs->open_flags & BDRV_O_ALLOW_RDWR)) {
        error_setg(errp, "Node '%s' is read only",
                   bdrv_get_device_or_node_name(bs));
        return -EPERM;
    }

    return 0;
}

int bdrv_set_read_only(BlockDriverState *bs, bool read_only, Error **errp)
{
    int ret = 0;

    ret = bdrv_can_set_read_only(bs, read_only, errp);
    if (ret < 0) {
        return ret;
    }

    bs->read_only = read_only;
    return 0;
}

void bdrv_get_full_backing_filename_from_filename(const char *backed,
                                                  const char *backing,
                                                  char *dest, size_t sz,
                                                  Error **errp)
{
    if (backing[0] == '\0' || path_has_protocol(backing) ||
        path_is_absolute(backing))
    {
        pstrcpy(dest, sz, backing);
    } else if (backed[0] == '\0' || strstart(backed, "json:", NULL)) {
        error_setg(errp, "Cannot use relative backing file names for '%s'",
                   backed);
    } else {
        path_combine(dest, sz, backed, backing);
    }
}

void bdrv_get_full_backing_filename(BlockDriverState *bs, char *dest, size_t sz,
                                    Error **errp)
{
    char *backed = bs->exact_filename[0] ? bs->exact_filename : bs->filename;

    bdrv_get_full_backing_filename_from_filename(backed, bs->backing_file,
                                                 dest, sz, errp);
}

void bdrv_register(BlockDriver *bdrv)
{
    QLIST_INSERT_HEAD(&bdrv_drivers, bdrv, list);
}

BlockDriverState *bdrv_new(void)
{
    BlockDriverState *bs;
    int i;

    bs = g_new0(BlockDriverState, 1);
    QLIST_INIT(&bs->dirty_bitmaps);
    for (i = 0; i < BLOCK_OP_TYPE_MAX; i++) {
        QLIST_INIT(&bs->op_blockers[i]);
    }
    notifier_with_return_list_init(&bs->before_write_notifiers);
    bs->refcnt = 1;
    bs->aio_context = qemu_get_aio_context();

    qemu_co_queue_init(&bs->flush_queue);

    QTAILQ_INSERT_TAIL(&all_bdrv_states, bs, bs_list);

    return bs;
}

static BlockDriver *bdrv_do_find_format(const char *format_name)
{
    BlockDriver *drv1;

    QLIST_FOREACH(drv1, &bdrv_drivers, list) {
        if (!strcmp(drv1->format_name, format_name)) {
            return drv1;
        }
    }

    return NULL;
}

BlockDriver *bdrv_find_format(const char *format_name)
{
    BlockDriver *drv1;
    int i;

    drv1 = bdrv_do_find_format(format_name);
    if (drv1) {
        return drv1;
    }

    /* The driver isn't registered, maybe we need to load a module */
    for (i = 0; i < (int)ARRAY_SIZE(block_driver_modules); ++i) {
        if (!strcmp(block_driver_modules[i].format_name, format_name)) {
            block_module_load_one(block_driver_modules[i].library_name);
            break;
        }
    }

    return bdrv_do_find_format(format_name);
}

static int bdrv_is_whitelisted(BlockDriver *drv, bool read_only)
{
    static const char *whitelist_rw[] = {
        CONFIG_BDRV_RW_WHITELIST
    };
    static const char *whitelist_ro[] = {
        CONFIG_BDRV_RO_WHITELIST
    };
    const char **p;

    if (!whitelist_rw[0] && !whitelist_ro[0]) {
        return 1;               /* no whitelist, anything goes */
    }

    for (p = whitelist_rw; *p; p++) {
        if (!strcmp(drv->format_name, *p)) {
            return 1;
        }
    }
    if (read_only) {
        for (p = whitelist_ro; *p; p++) {
            if (!strcmp(drv->format_name, *p)) {
                return 1;
            }
        }
    }
    return 0;
}

bool bdrv_uses_whitelist(void)
{
    return use_bdrv_whitelist;
}

typedef struct CreateCo {
    BlockDriver *drv;
    char *filename;
    QemuOpts *opts;
    int ret;
    Error *err;
} CreateCo;

static void coroutine_fn bdrv_create_co_entry(void *opaque)
{
    Error *local_err = NULL;
    int ret;

    CreateCo *cco = opaque;
    assert(cco->drv);

    ret = cco->drv->bdrv_create(cco->filename, cco->opts, &local_err);
    error_propagate(&cco->err, local_err);
    cco->ret = ret;
}

int bdrv_create(BlockDriver *drv, const char* filename,
                QemuOpts *opts, Error **errp)
{
    int ret;

    Coroutine *co;
    CreateCo cco = {
        .drv = drv,
        .filename = g_strdup(filename),
        .opts = opts,
        .ret = NOT_DONE,
        .err = NULL,
    };

    if (!drv->bdrv_create) {
        error_setg(errp, "Driver '%s' does not support image creation", drv->format_name);
        ret = -ENOTSUP;
        goto out;
    }

    if (qemu_in_coroutine()) {
        /* Fast-path if already in coroutine context */
        bdrv_create_co_entry(&cco);
    } else {
        co = qemu_coroutine_create(bdrv_create_co_entry, &cco);
        qemu_coroutine_enter(co);
        while (cco.ret == NOT_DONE) {
            aio_poll(qemu_get_aio_context(), true);
        }
    }

    ret = cco.ret;
    if (ret < 0) {
        if (cco.err) {
            error_propagate(errp, cco.err);
        } else {
            error_setg_errno(errp, -ret, "Could not create image");
        }
    }

out:
    g_free(cco.filename);
    return ret;
}

int bdrv_create_file(const char *filename, QemuOpts *opts, Error **errp)
{
    BlockDriver *drv;
    Error *local_err = NULL;
    int ret;

    drv = bdrv_find_protocol(filename, true, errp);
    if (drv == NULL) {
        return -ENOENT;
    }

    ret = bdrv_create(drv, filename, opts, &local_err);
    error_propagate(errp, local_err);
    return ret;
}

/**
 * Try to get @bs's logical and physical block size.
 * On success, store them in @bsz struct and return 0.
 * On failure return -errno.
 * @bs must not be empty.
 */
int bdrv_probe_blocksizes(BlockDriverState *bs, BlockSizes *bsz)
{
    BlockDriver *drv = bs->drv;

    if (drv && drv->bdrv_probe_blocksizes) {
        return drv->bdrv_probe_blocksizes(bs, bsz);
    }

    return -ENOTSUP;
}

/**
 * Try to get @bs's geometry (cyls, heads, sectors).
 * On success, store them in @geo struct and return 0.
 * On failure return -errno.
 * @bs must not be empty.
 */
int bdrv_probe_geometry(BlockDriverState *bs, HDGeometry *geo)
{
    BlockDriver *drv = bs->drv;

    if (drv && drv->bdrv_probe_geometry) {
        return drv->bdrv_probe_geometry(bs, geo);
    }

    return -ENOTSUP;
}

/*
 * Create a uniquely-named empty temporary file.
 * Return 0 upon success, otherwise a negative errno value.
 */
int get_tmp_filename(char *filename, int size)
{
#ifdef _WIN32
    char temp_dir[MAX_PATH];
    /* GetTempFileName requires that its output buffer (4th param)
       have length MAX_PATH or greater.  */
    assert(size >= MAX_PATH);
    return (GetTempPath(MAX_PATH, temp_dir)
            && GetTempFileName(temp_dir, "qem", 0, filename)
            ? 0 : -GetLastError());
#else
    int fd;
    const char *tmpdir;
    tmpdir = getenv("TMPDIR");
    if (!tmpdir) {
        tmpdir = "/var/tmp";
    }
    if (snprintf(filename, size, "%s/vl.XXXXXX", tmpdir) >= size) {
        return -EOVERFLOW;
    }
    fd = mkstemp(filename);
    if (fd < 0) {
        return -errno;
    }
    if (close(fd) != 0) {
        unlink(filename);
        return -errno;
    }
    return 0;
#endif
}

/*
 * Detect host devices. By convention, /dev/cdrom[N] is always
 * recognized as a host CDROM.
 */
static BlockDriver *find_hdev_driver(const char *filename)
{
    int score_max = 0, score;
    BlockDriver *drv = NULL, *d;

    QLIST_FOREACH(d, &bdrv_drivers, list) {
        if (d->bdrv_probe_device) {
            score = d->bdrv_probe_device(filename);
            if (score > score_max) {
                score_max = score;
                drv = d;
            }
        }
    }

    return drv;
}

static BlockDriver *bdrv_do_find_protocol(const char *protocol)
{
    BlockDriver *drv1;

    QLIST_FOREACH(drv1, &bdrv_drivers, list) {
        if (drv1->protocol_name && !strcmp(drv1->protocol_name, protocol)) {
            return drv1;
        }
    }

    return NULL;
}

BlockDriver *bdrv_find_protocol(const char *filename,
                                bool allow_protocol_prefix,
                                Error **errp)
{
    BlockDriver *drv1;
    char protocol[128];
    int len;
    const char *p;
    int i;

    /* TODO Drivers without bdrv_file_open must be specified explicitly */

    /*
     * XXX(hch): we really should not let host device detection
     * override an explicit protocol specification, but moving this
     * later breaks access to device names with colons in them.
     * Thanks to the brain-dead persistent naming schemes on udev-
     * based Linux systems those actually are quite common.
     */
    drv1 = find_hdev_driver(filename);
    if (drv1) {
        return drv1;
    }

    if (!path_has_protocol(filename) || !allow_protocol_prefix) {
        return &bdrv_file;
    }

    p = strchr(filename, ':');
    assert(p != NULL);
    len = p - filename;
    if (len > sizeof(protocol) - 1)
        len = sizeof(protocol) - 1;
    memcpy(protocol, filename, len);
    protocol[len] = '\0';

    drv1 = bdrv_do_find_protocol(protocol);
    if (drv1) {
        return drv1;
    }

    for (i = 0; i < (int)ARRAY_SIZE(block_driver_modules); ++i) {
        if (block_driver_modules[i].protocol_name &&
            !strcmp(block_driver_modules[i].protocol_name, protocol)) {
            block_module_load_one(block_driver_modules[i].library_name);
            break;
        }
    }

    drv1 = bdrv_do_find_protocol(protocol);
    if (!drv1) {
        error_setg(errp, "Unknown protocol '%s'", protocol);
    }
    return drv1;
}

/*
 * Guess image format by probing its contents.
 * This is not a good idea when your image is raw (CVE-2008-2004), but
 * we do it anyway for backward compatibility.
 *
 * @buf         contains the image's first @buf_size bytes.
 * @buf_size    is the buffer size in bytes (generally BLOCK_PROBE_BUF_SIZE,
 *              but can be smaller if the image file is smaller)
 * @filename    is its filename.
 *
 * For all block drivers, call the bdrv_probe() method to get its
 * probing score.
 * Return the first block driver with the highest probing score.
 */
BlockDriver *bdrv_probe_all(const uint8_t *buf, int buf_size,
                            const char *filename)
{
    int score_max = 0, score;
    BlockDriver *drv = NULL, *d;

    QLIST_FOREACH(d, &bdrv_drivers, list) {
        if (d->bdrv_probe) {
            score = d->bdrv_probe(buf, buf_size, filename);
            if (score > score_max) {
                score_max = score;
                drv = d;
            }
        }
    }

    return drv;
}

static int find_image_format(BlockBackend *file, const char *filename,
                             BlockDriver **pdrv, Error **errp)
{
    BlockDriver *drv;
    uint8_t buf[BLOCK_PROBE_BUF_SIZE];
    int ret = 0;

    /* Return the raw BlockDriver * to scsi-generic devices or empty drives */
    if (blk_is_sg(file) || !blk_is_inserted(file) || blk_getlength(file) == 0) {
        *pdrv = &bdrv_raw;
        return ret;
    }

    ret = blk_pread(file, 0, buf, sizeof(buf));
    if (ret < 0) {
        error_setg_errno(errp, -ret, "Could not read image for determining its "
                         "format");
        *pdrv = NULL;
        return ret;
    }

    drv = bdrv_probe_all(buf, ret, filename);
    if (!drv) {
        error_setg(errp, "Could not determine image format: No compatible "
                   "driver found");
        ret = -ENOENT;
    }
    *pdrv = drv;
    return ret;
}

/**
 * Set the current 'total_sectors' value
 * Return 0 on success, -errno on error.
 */
static int refresh_total_sectors(BlockDriverState *bs, int64_t hint)
{
    BlockDriver *drv = bs->drv;

    /* Do not attempt drv->bdrv_getlength() on scsi-generic devices */
    if (bdrv_is_sg(bs))
        return 0;

    /* query actual device if possible, otherwise just trust the hint */
    if (drv->bdrv_getlength) {
        int64_t length = drv->bdrv_getlength(bs);
        if (length < 0) {
            return length;
        }
        hint = DIV_ROUND_UP(length, BDRV_SECTOR_SIZE);
    }

    bs->total_sectors = hint;
    return 0;
}

/**
 * Combines a QDict of new block driver @options with any missing options taken
 * from @old_options, so that leaving out an option defaults to its old value.
 */
static void bdrv_join_options(BlockDriverState *bs, QDict *options,
                              QDict *old_options)
{
    if (bs->drv && bs->drv->bdrv_join_options) {
        bs->drv->bdrv_join_options(options, old_options);
    } else {
        qdict_join(options, old_options, false);
    }
}

/**
 * Set open flags for a given discard mode
 *
 * Return 0 on success, -1 if the discard mode was invalid.
 */
int bdrv_parse_discard_flags(const char *mode, int *flags)
{
    *flags &= ~BDRV_O_UNMAP;

    if (!strcmp(mode, "off") || !strcmp(mode, "ignore")) {
        /* do nothing */
    } else if (!strcmp(mode, "on") || !strcmp(mode, "unmap")) {
        *flags |= BDRV_O_UNMAP;
    } else {
        return -1;
    }

    return 0;
}

/**
 * Set open flags for a given cache mode
 *
 * Return 0 on success, -1 if the cache mode was invalid.
 */
int bdrv_parse_cache_mode(const char *mode, int *flags, bool *writethrough)
{
    *flags &= ~BDRV_O_CACHE_MASK;

    if (!strcmp(mode, "off") || !strcmp(mode, "none")) {
        *writethrough = false;
        *flags |= BDRV_O_NOCACHE;
    } else if (!strcmp(mode, "directsync")) {
        *writethrough = true;
        *flags |= BDRV_O_NOCACHE;
    } else if (!strcmp(mode, "writeback")) {
        *writethrough = false;
    } else if (!strcmp(mode, "unsafe")) {
        *writethrough = false;
        *flags |= BDRV_O_NO_FLUSH;
    } else if (!strcmp(mode, "writethrough")) {
        *writethrough = true;
    } else {
        return -1;
    }

    return 0;
}

static char *bdrv_child_get_parent_desc(BdrvChild *c)
{
    BlockDriverState *parent = c->opaque;
    return g_strdup(bdrv_get_device_or_node_name(parent));
}

static void bdrv_child_cb_drained_begin(BdrvChild *child)
{
    BlockDriverState *bs = child->opaque;
    bdrv_drained_begin(bs);
}

static void bdrv_child_cb_drained_end(BdrvChild *child)
{
    BlockDriverState *bs = child->opaque;
    bdrv_drained_end(bs);
}

static int bdrv_child_cb_inactivate(BdrvChild *child)
{
    BlockDriverState *bs = child->opaque;
    assert(bs->open_flags & BDRV_O_INACTIVE);
    return 0;
}

/*
 * Returns the options and flags that a temporary snapshot should get, based on
 * the originally requested flags (the originally requested image will have
 * flags like a backing file)
 */
static void bdrv_temp_snapshot_options(int *child_flags, QDict *child_options,
                                       int parent_flags, QDict *parent_options)
{
    *child_flags = (parent_flags & ~BDRV_O_SNAPSHOT) | BDRV_O_TEMPORARY;

    /* For temporary files, unconditional cache=unsafe is fine */
    qdict_set_default_str(child_options, BDRV_OPT_CACHE_DIRECT, "off");
    qdict_set_default_str(child_options, BDRV_OPT_CACHE_NO_FLUSH, "on");

    /* Copy the read-only option from the parent */
    qdict_copy_default(child_options, parent_options, BDRV_OPT_READ_ONLY);

    /* aio=native doesn't work for cache.direct=off, so disable it for the
     * temporary snapshot */
    *child_flags &= ~BDRV_O_NATIVE_AIO;
}

/*
 * Returns the options and flags that bs->file should get if a protocol driver
 * is expected, based on the given options and flags for the parent BDS
 */
static void bdrv_inherited_options(int *child_flags, QDict *child_options,
                                   int parent_flags, QDict *parent_options)
{
    int flags = parent_flags;

    /* Enable protocol handling, disable format probing for bs->file */
    flags |= BDRV_O_PROTOCOL;

    /* If the cache mode isn't explicitly set, inherit direct and no-flush from
     * the parent. */
    qdict_copy_default(child_options, parent_options, BDRV_OPT_CACHE_DIRECT);
    qdict_copy_default(child_options, parent_options, BDRV_OPT_CACHE_NO_FLUSH);
    qdict_copy_default(child_options, parent_options, BDRV_OPT_FORCE_SHARE);

    /* Inherit the read-only option from the parent if it's not set */
    qdict_copy_default(child_options, parent_options, BDRV_OPT_READ_ONLY);

    /* Our block drivers take care to send flushes and respect unmap policy,
     * so we can default to enable both on lower layers regardless of the
     * corresponding parent options. */
    qdict_set_default_str(child_options, BDRV_OPT_DISCARD, "unmap");

    /* Clear flags that only apply to the top layer */
    flags &= ~(BDRV_O_SNAPSHOT | BDRV_O_NO_BACKING | BDRV_O_COPY_ON_READ |
               BDRV_O_NO_IO);

    *child_flags = flags;
}

const BdrvChildRole child_file = {
    .get_parent_desc = bdrv_child_get_parent_desc,
    .inherit_options = bdrv_inherited_options,
    .drained_begin   = bdrv_child_cb_drained_begin,
    .drained_end     = bdrv_child_cb_drained_end,
    .inactivate      = bdrv_child_cb_inactivate,
};

/*
 * Returns the options and flags that bs->file should get if the use of formats
 * (and not only protocols) is permitted for it, based on the given options and
 * flags for the parent BDS
 */
static void bdrv_inherited_fmt_options(int *child_flags, QDict *child_options,
                                       int parent_flags, QDict *parent_options)
{
    child_file.inherit_options(child_flags, child_options,
                               parent_flags, parent_options);

    *child_flags &= ~(BDRV_O_PROTOCOL | BDRV_O_NO_IO);
}

const BdrvChildRole child_format = {
    .get_parent_desc = bdrv_child_get_parent_desc,
    .inherit_options = bdrv_inherited_fmt_options,
    .drained_begin   = bdrv_child_cb_drained_begin,
    .drained_end     = bdrv_child_cb_drained_end,
    .inactivate      = bdrv_child_cb_inactivate,
};

static void bdrv_backing_attach(BdrvChild *c)
{
    BlockDriverState *parent = c->opaque;
    BlockDriverState *backing_hd = c->bs;

    assert(!parent->backing_blocker);
    error_setg(&parent->backing_blocker,
               "node is used as backing hd of '%s'",
               bdrv_get_device_or_node_name(parent));

    parent->open_flags &= ~BDRV_O_NO_BACKING;
    pstrcpy(parent->backing_file, sizeof(parent->backing_file),
            backing_hd->filename);
    pstrcpy(parent->backing_format, sizeof(parent->backing_format),
            backing_hd->drv ? backing_hd->drv->format_name : "");

    bdrv_op_block_all(backing_hd, parent->backing_blocker);
    /* Otherwise we won't be able to commit or stream */
    bdrv_op_unblock(backing_hd, BLOCK_OP_TYPE_COMMIT_TARGET,
                    parent->backing_blocker);
    bdrv_op_unblock(backing_hd, BLOCK_OP_TYPE_STREAM,
                    parent->backing_blocker);
    /*
     * We do backup in 3 ways:
     * 1. drive backup
     *    The target bs is new opened, and the source is top BDS
     * 2. blockdev backup
     *    Both the source and the target are top BDSes.
     * 3. internal backup(used for block replication)
     *    Both the source and the target are backing file
     *
     * In case 1 and 2, neither the source nor the target is the backing file.
     * In case 3, we will block the top BDS, so there is only one block job
     * for the top BDS and its backing chain.
     */
    bdrv_op_unblock(backing_hd, BLOCK_OP_TYPE_BACKUP_SOURCE,
                    parent->backing_blocker);
    bdrv_op_unblock(backing_hd, BLOCK_OP_TYPE_BACKUP_TARGET,
                    parent->backing_blocker);
}

static void bdrv_backing_detach(BdrvChild *c)
{
    BlockDriverState *parent = c->opaque;

    assert(parent->backing_blocker);
    bdrv_op_unblock_all(c->bs, parent->backing_blocker);
    error_free(parent->backing_blocker);
    parent->backing_blocker = NULL;
}

/*
 * Returns the options and flags that bs->backing should get, based on the
 * given options and flags for the parent BDS
 */
static void bdrv_backing_options(int *child_flags, QDict *child_options,
                                 int parent_flags, QDict *parent_options)
{
    int flags = parent_flags;

    /* The cache mode is inherited unmodified for backing files; except WCE,
     * which is only applied on the top level (BlockBackend) */
    qdict_copy_default(child_options, parent_options, BDRV_OPT_CACHE_DIRECT);
    qdict_copy_default(child_options, parent_options, BDRV_OPT_CACHE_NO_FLUSH);
    qdict_copy_default(child_options, parent_options, BDRV_OPT_FORCE_SHARE);

    /* backing files always opened read-only */
    qdict_set_default_str(child_options, BDRV_OPT_READ_ONLY, "on");
    flags &= ~BDRV_O_COPY_ON_READ;

    /* snapshot=on is handled on the top layer */
    flags &= ~(BDRV_O_SNAPSHOT | BDRV_O_TEMPORARY);

    *child_flags = flags;
}

const BdrvChildRole child_backing = {
    .get_parent_desc = bdrv_child_get_parent_desc,
    .attach          = bdrv_backing_attach,
    .detach          = bdrv_backing_detach,
    .inherit_options = bdrv_backing_options,
    .drained_begin   = bdrv_child_cb_drained_begin,
    .drained_end     = bdrv_child_cb_drained_end,
    .inactivate      = bdrv_child_cb_inactivate,
};

static int bdrv_open_flags(BlockDriverState *bs, int flags)
{
    int open_flags = flags;

    /*
     * Clear flags that are internal to the block layer before opening the
     * image.
     */
    open_flags &= ~(BDRV_O_SNAPSHOT | BDRV_O_NO_BACKING | BDRV_O_PROTOCOL);

    /*
     * Snapshots should be writable.
     */
    if (flags & BDRV_O_TEMPORARY) {
        open_flags |= BDRV_O_RDWR;
    }

    return open_flags;
}

static void update_flags_from_options(int *flags, QemuOpts *opts)
{
    *flags &= ~BDRV_O_CACHE_MASK;

    assert(qemu_opt_find(opts, BDRV_OPT_CACHE_NO_FLUSH));
    if (qemu_opt_get_bool(opts, BDRV_OPT_CACHE_NO_FLUSH, false)) {
        *flags |= BDRV_O_NO_FLUSH;
    }

    assert(qemu_opt_find(opts, BDRV_OPT_CACHE_DIRECT));
    if (qemu_opt_get_bool(opts, BDRV_OPT_CACHE_DIRECT, false)) {
        *flags |= BDRV_O_NOCACHE;
    }

    *flags &= ~BDRV_O_RDWR;

    assert(qemu_opt_find(opts, BDRV_OPT_READ_ONLY));
    if (!qemu_opt_get_bool(opts, BDRV_OPT_READ_ONLY, false)) {
        *flags |= BDRV_O_RDWR;
    }

}

static void update_options_from_flags(QDict *options, int flags)
{
    if (!qdict_haskey(options, BDRV_OPT_CACHE_DIRECT)) {
        qdict_put_bool(options, BDRV_OPT_CACHE_DIRECT, flags & BDRV_O_NOCACHE);
    }
    if (!qdict_haskey(options, BDRV_OPT_CACHE_NO_FLUSH)) {