blockdev.c 69.9 KB
Newer Older
1 2 3 4 5 6 7
/*
 * 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.
8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
 *
 * 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.
31 32
 */

33
#include "sysemu/blockdev.h"
Paolo Bonzini's avatar
Paolo Bonzini committed
34
#include "hw/block/block.h"
35
#include "block/blockjob.h"
36
#include "monitor/monitor.h"
37
#include "qapi/qmp/qerror.h"
38 39
#include "qemu/option.h"
#include "qemu/config-file.h"
40
#include "qapi/qmp/types.h"
41 42
#include "qapi-visit.h"
#include "qapi/qmp-output-visitor.h"
43
#include "sysemu/sysemu.h"
44
#include "block/block_int.h"
Luiz Capitulino's avatar
Luiz Capitulino committed
45
#include "qmp-commands.h"
46
#include "trace.h"
47
#include "sysemu/arch_init.h"
48

49
static QTAILQ_HEAD(drivelist, DriveInfo) drives = QTAILQ_HEAD_INITIALIZER(drives);
50

51 52 53 54 55 56 57 58 59 60 61 62 63
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] = {
64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79
    /*
     * 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,
80 81
};

82 83 84 85 86 87 88 89 90 91 92
/*
 * 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);

93 94 95 96
    if (dinfo && !dinfo->enable_auto_del) {
        return;
    }

97 98 99
    if (bs->job) {
        block_job_cancel(bs->job);
    }
100 101 102
    if (dinfo) {
        dinfo->auto_del = 1;
    }
103 104 105 106 107 108
}

void blockdev_auto_del(BlockDriverState *bs)
{
    DriveInfo *dinfo = drive_get_by_blockdev(bs);

109
    if (dinfo && dinfo->auto_del) {
110
        drive_put_ref(dinfo);
111 112 113
    }
}

114 115 116 117 118 119 120 121 122 123 124 125
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;
}

126 127 128 129 130 131
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,
132
                    const char *optstr)
133 134
{
    QemuOpts *opts;
135
    char buf[32];
136

137
    opts = drive_def(optstr);
138 139 140
    if (!opts) {
        return NULL;
    }
141 142 143 144 145 146 147
    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);
    }
148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168
    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;
}

169 170 171 172 173 174 175
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));
}

176 177 178 179 180 181 182 183 184 185 186 187 188 189
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;
}

190 191 192 193 194 195 196 197 198 199
/* 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]++);
}

200
DriveInfo *drive_get_by_blockdev(BlockDriverState *bs)
201 202 203 204
{
    DriveInfo *dinfo;

    QTAILQ_FOREACH(dinfo, &drives, next) {
205 206 207
        if (dinfo->bdrv == bs) {
            return dinfo;
        }
208
    }
209
    return NULL;
210 211 212 213
}

static void bdrv_format_print(void *opaque, const char *name)
{
214
    error_printf(" %s", name);
215 216
}

217
static void drive_uninit(DriveInfo *dinfo)
218
{
219 220 221 222
    if (dinfo->opts) {
        qemu_opts_del(dinfo->opts);
    }

Fam Zheng's avatar
Fam Zheng committed
223
    bdrv_unref(dinfo->bdrv);
224
    g_free(dinfo->id);
225
    QTAILQ_REMOVE(&drives, dinfo, next);
226
    g_free(dinfo->serial);
227
    g_free(dinfo);
228 229
}

230 231 232 233 234 235 236 237 238 239 240 241 242
void drive_put_ref(DriveInfo *dinfo)
{
    assert(dinfo->refcount);
    if (--dinfo->refcount == 0) {
        drive_uninit(dinfo);
    }
}

void drive_get_ref(DriveInfo *dinfo)
{
    dinfo->refcount++;
}

243 244
typedef struct {
    QEMUBH *bh;
245 246
    BlockDriverState *bs;
} BDRVPutRefBH;
247

248
static void bdrv_put_ref_bh(void *opaque)
249
{
250
    BDRVPutRefBH *s = opaque;
251

252
    bdrv_unref(s->bs);
253 254 255 256 257
    qemu_bh_delete(s->bh);
    g_free(s);
}

/*
258
 * Release a BDS reference in a BH
259
 *
260 261 262
 * It is not safe to use bdrv_unref() from a callback function when the callers
 * still need the BlockDriverState.  In such cases we schedule a BH to release
 * the reference.
263
 */
264
static void bdrv_put_ref_bh_schedule(BlockDriverState *bs)
265
{
266
    BDRVPutRefBH *s;
267

268 269 270
    s = g_new(BDRVPutRefBH, 1);
    s->bh = qemu_bh_new(bdrv_put_ref_bh, s);
    s->bs = bs;
271 272 273
    qemu_bh_schedule(s->bh);
}

274
static int parse_block_error_action(const char *buf, bool is_read, Error **errp)
275 276
{
    if (!strcmp(buf, "ignore")) {
277
        return BLOCKDEV_ON_ERROR_IGNORE;
278
    } else if (!is_read && !strcmp(buf, "enospc")) {
279
        return BLOCKDEV_ON_ERROR_ENOSPC;
280
    } else if (!strcmp(buf, "stop")) {
281
        return BLOCKDEV_ON_ERROR_STOP;
282
    } else if (!strcmp(buf, "report")) {
283
        return BLOCKDEV_ON_ERROR_REPORT;
284
    } else {
285 286
        error_setg(errp, "'%s' invalid %s error action",
                   buf, is_read ? "read" : "write");
287 288 289 290
        return -1;
    }
}

291
static bool check_throttle_config(ThrottleConfig *cfg, Error **errp)
292
{
293 294 295
    if (throttle_conflicting(cfg)) {
        error_setg(errp, "bps/iops/max total values and read/write values"
                         " cannot be used at the same time");
296 297 298
        return false;
    }

299 300
    if (!throttle_is_valid(cfg)) {
        error_setg(errp, "bps/iops/maxs values must be 0 or greater");
301 302 303
        return false;
    }

304 305 306
    return true;
}

307 308
typedef enum { MEDIA_DISK, MEDIA_CDROM } DriveMediaType;

309
/* Takes the ownership of bs_opts */
310
static DriveInfo *blockdev_init(const char *file, QDict *bs_opts,
311 312
                                BlockInterfaceType type,
                                Error **errp)
313 314 315 316 317 318 319
{
    const char *buf;
    const char *serial;
    int ro = 0;
    int bdrv_flags = 0;
    int on_read_error, on_write_error;
    DriveInfo *dinfo;
320
    ThrottleConfig cfg;
321
    int snapshot = 0;
322
    bool copy_on_read;
323
    int ret;
324
    Error *error = NULL;
325 326
    QemuOpts *opts;
    const char *id;
327
    bool has_driver_specific_opts;
328
    BlockDriver *drv = NULL;
329

330 331 332
    /* Check common options by copying from bs_opts to opts, all other options
     * stay in bs_opts for processing by bdrv_open(). */
    id = qdict_get_try_str(bs_opts, "id");
333 334
    opts = qemu_opts_create(&qemu_common_drive_opts, id, 1, &error);
    if (error_is_set(&error)) {
335
        error_propagate(errp, error);
336 337 338 339 340
        return NULL;
    }

    qemu_opts_absorb_qdict(opts, bs_opts, &error);
    if (error_is_set(&error)) {
341
        error_propagate(errp, error);
342
        goto early_err;
343 344 345 346 347 348
    }

    if (id) {
        qdict_del(bs_opts, "id");
    }

349 350
    has_driver_specific_opts = !!qdict_size(bs_opts);

351 352
    /* extract parameters */
    snapshot = qemu_opt_get_bool(opts, "snapshot", 0);
353
    ro = qemu_opt_get_bool(opts, "read-only", 0);
354
    copy_on_read = qemu_opt_get_bool(opts, "copy-on-read", false);
355 356 357

    serial = qemu_opt_get(opts, "serial");

358 359
    if ((buf = qemu_opt_get(opts, "discard")) != NULL) {
        if (bdrv_parse_discard_flags(buf, &bdrv_flags) != 0) {
360
            error_setg(errp, "invalid discard option");
361
            goto early_err;
362 363 364
        }
    }

365 366 367 368 369 370
    if (qemu_opt_get_bool(opts, "cache.writeback", true)) {
        bdrv_flags |= BDRV_O_CACHE_WB;
    }
    if (qemu_opt_get_bool(opts, "cache.direct", false)) {
        bdrv_flags |= BDRV_O_NOCACHE;
    }
371
    if (qemu_opt_get_bool(opts, "cache.no-flush", false)) {
372
        bdrv_flags |= BDRV_O_NO_FLUSH;
373 374 375 376 377 378 379 380 381
    }

#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 {
382
           error_setg(errp, "invalid aio option");
383
           goto early_err;
384 385 386 387 388
        }
    }
#endif

    if ((buf = qemu_opt_get(opts, "format")) != NULL) {
389 390 391 392
        if (is_help_option(buf)) {
            error_printf("Supported formats:");
            bdrv_iterate_format(bdrv_format_print, NULL);
            error_printf("\n");
393
            goto early_err;
394
        }
395

396
        drv = bdrv_find_format(buf);
397
        if (!drv) {
398
            error_setg(errp, "'%s' invalid format", buf);
399
            goto early_err;
400
        }
401 402
    }

403
    /* disk I/O throttling */
404 405
    memset(&cfg, 0, sizeof(cfg));
    cfg.buckets[THROTTLE_BPS_TOTAL].avg =
406
        qemu_opt_get_number(opts, "throttling.bps-total", 0);
407
    cfg.buckets[THROTTLE_BPS_READ].avg  =
408
        qemu_opt_get_number(opts, "throttling.bps-read", 0);
409
    cfg.buckets[THROTTLE_BPS_WRITE].avg =
410
        qemu_opt_get_number(opts, "throttling.bps-write", 0);
411
    cfg.buckets[THROTTLE_OPS_TOTAL].avg =
412
        qemu_opt_get_number(opts, "throttling.iops-total", 0);
413
    cfg.buckets[THROTTLE_OPS_READ].avg =
414
        qemu_opt_get_number(opts, "throttling.iops-read", 0);
415
    cfg.buckets[THROTTLE_OPS_WRITE].avg =
416
        qemu_opt_get_number(opts, "throttling.iops-write", 0);
417

418 419 420 421 422 423 424 425 426 427 428 429
    cfg.buckets[THROTTLE_BPS_TOTAL].max =
        qemu_opt_get_number(opts, "throttling.bps-total-max", 0);
    cfg.buckets[THROTTLE_BPS_READ].max  =
        qemu_opt_get_number(opts, "throttling.bps-read-max", 0);
    cfg.buckets[THROTTLE_BPS_WRITE].max =
        qemu_opt_get_number(opts, "throttling.bps-write-max", 0);
    cfg.buckets[THROTTLE_OPS_TOTAL].max =
        qemu_opt_get_number(opts, "throttling.iops-total-max", 0);
    cfg.buckets[THROTTLE_OPS_READ].max =
        qemu_opt_get_number(opts, "throttling.iops-read-max", 0);
    cfg.buckets[THROTTLE_OPS_WRITE].max =
        qemu_opt_get_number(opts, "throttling.iops-write-max", 0);
430

431
    cfg.op_size = qemu_opt_get_number(opts, "throttling.iops-size", 0);
432 433

    if (!check_throttle_config(&cfg, &error)) {
434
        error_propagate(errp, error);
435
        goto early_err;
436 437
    }

438
    on_write_error = BLOCKDEV_ON_ERROR_ENOSPC;
439 440
    if ((buf = qemu_opt_get(opts, "werror")) != NULL) {
        if (type != IF_IDE && type != IF_SCSI && type != IF_VIRTIO && type != IF_NONE) {
441
            error_setg(errp, "werror is not supported by this bus type");
442
            goto early_err;
443 444
        }

445 446 447
        on_write_error = parse_block_error_action(buf, 0, &error);
        if (error_is_set(&error)) {
            error_propagate(errp, error);
448
            goto early_err;
449 450 451
        }
    }

452
    on_read_error = BLOCKDEV_ON_ERROR_REPORT;
453
    if ((buf = qemu_opt_get(opts, "rerror")) != NULL) {
454
        if (type != IF_IDE && type != IF_VIRTIO && type != IF_SCSI && type != IF_NONE) {
455
            error_report("rerror is not supported by this bus type");
456
            goto early_err;
457 458
        }

459 460 461
        on_read_error = parse_block_error_action(buf, 1, &error);
        if (error_is_set(&error)) {
            error_propagate(errp, error);
462
            goto early_err;
463 464 465
        }
    }

466 467 468
    /* init */
    dinfo = g_malloc0(sizeof(*dinfo));
    dinfo->id = g_strdup(qemu_opts_id(opts));
469
    dinfo->bdrv = bdrv_new(dinfo->id);
470 471
    dinfo->bdrv->open_flags = snapshot ? BDRV_O_SNAPSHOT : 0;
    dinfo->bdrv->read_only = ro;
472
    dinfo->type = type;
473
    dinfo->refcount = 1;
474 475 476
    if (serial != NULL) {
        dinfo->serial = g_strdup(serial);
    }
477 478
    QTAILQ_INSERT_TAIL(&drives, dinfo, next);

479 480
    bdrv_set_on_error(dinfo->bdrv, on_read_error, on_write_error);

481
    /* disk I/O throttling */
482 483 484 485
    if (throttle_enabled(&cfg)) {
        bdrv_io_limits_enable(dinfo->bdrv);
        bdrv_set_io_limits(dinfo->bdrv, &cfg);
    }
486

487
    if (!file || !*file) {
488
        if (has_driver_specific_opts) {
489 490
            file = NULL;
        } else {
491 492
            QDECREF(bs_opts);
            qemu_opts_del(opts);
493 494
            return dinfo;
        }
495 496 497 498 499 500 501
    }
    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);
    }

502 503 504 505
    if (copy_on_read) {
        bdrv_flags |= BDRV_O_COPY_ON_READ;
    }

506 507 508 509
    if (runstate_check(RUN_STATE_INMIGRATE)) {
        bdrv_flags |= BDRV_O_INCOMING;
    }

510 511
    bdrv_flags |= ro ? 0 : BDRV_O_RDWR;

512
    QINCREF(bs_opts);
513
    ret = bdrv_open(dinfo->bdrv, file, bs_opts, bdrv_flags, drv, &error);
514

515
    if (ret < 0) {
516 517 518
        error_setg(errp, "could not open disk image %s: %s",
                   file ?: dinfo->id, error_get_pretty(error));
        error_free(error);
519
        goto err;
520 521 522 523
    }

    if (bdrv_key_required(dinfo->bdrv))
        autostart = 0;
524

525
    QDECREF(bs_opts);
526 527
    qemu_opts_del(opts);

528
    return dinfo;
529 530

err:
Fam Zheng's avatar
Fam Zheng committed
531
    bdrv_unref(dinfo->bdrv);
532
    g_free(dinfo->id);
533
    QTAILQ_REMOVE(&drives, dinfo, next);
534
    g_free(dinfo);
535 536 537
early_err:
    QDECREF(bs_opts);
    qemu_opts_del(opts);
538
    return NULL;
539 540
}

541 542 543 544 545 546 547 548 549 550 551
static void qemu_opt_rename(QemuOpts *opts, const char *from, const char *to)
{
    const char *value;

    value = qemu_opt_get(opts, from);
    if (value) {
        qemu_opt_set(opts, to, value);
        qemu_opt_unset(opts, from);
    }
}

552 553 554 555 556
QemuOptsList qemu_legacy_drive_opts = {
    .name = "drive",
    .head = QTAILQ_HEAD_INITIALIZER(qemu_legacy_drive_opts.head),
    .desc = {
        {
557 558 559 560 561 562 563 564 565 566 567 568
            .name = "bus",
            .type = QEMU_OPT_NUMBER,
            .help = "bus number",
        },{
            .name = "unit",
            .type = QEMU_OPT_NUMBER,
            .help = "unit number (i.e. lun for scsi)",
        },{
            .name = "index",
            .type = QEMU_OPT_NUMBER,
            .help = "index number",
        },{
569 570 571
            .name = "media",
            .type = QEMU_OPT_STRING,
            .help = "media type (disk, cdrom)",
572 573 574 575
        },{
            .name = "if",
            .type = QEMU_OPT_STRING,
            .help = "interface (ide, scsi, sd, mtd, floppy, pflash, virtio)",
576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591
        },{
            .name = "cyls",
            .type = QEMU_OPT_NUMBER,
            .help = "number of cylinders (ide disk geometry)",
        },{
            .name = "heads",
            .type = QEMU_OPT_NUMBER,
            .help = "number of heads (ide disk geometry)",
        },{
            .name = "secs",
            .type = QEMU_OPT_NUMBER,
            .help = "number of sectors (ide disk geometry)",
        },{
            .name = "trans",
            .type = QEMU_OPT_STRING,
            .help = "chs translation (auto, lba, none)",
592 593 594 595
        },{
            .name = "boot",
            .type = QEMU_OPT_BOOL,
            .help = "(deprecated, ignored)",
596 597 598 599
        },{
            .name = "addr",
            .type = QEMU_OPT_STRING,
            .help = "pci address (virtio only)",
600 601 602 603
        },{
            .name = "file",
            .type = QEMU_OPT_STRING,
            .help = "file name",
604
        },
605 606 607 608 609 610 611 612 613 614 615 616

        /* Options that are passed on, but have special semantics with -drive */
        {
            .name = "read-only",
            .type = QEMU_OPT_BOOL,
            .help = "open drive file as read-only",
        },{
            .name = "copy-on-read",
            .type = QEMU_OPT_BOOL,
            .help = "copy read data from backing file into image file",
        },

617 618 619 620
        { /* end of list */ }
    },
};

621 622
DriveInfo *drive_init(QemuOpts *all_opts, BlockInterfaceType block_default_type)
{
623
    const char *value;
624
    DriveInfo *dinfo = NULL;
625
    QDict *bs_opts;
626 627
    QemuOpts *legacy_opts;
    DriveMediaType media = MEDIA_DISK;
628
    BlockInterfaceType type;
629
    int cyls, heads, secs, translation;
630
    int max_devs, bus_id, unit_id, index;
631
    const char *devaddr;
632 633
    bool read_only = false;
    bool copy_on_read;
634
    const char *filename;
635
    Error *local_err = NULL;
636

637 638 639 640 641 642 643 644 645
    /* Change legacy command line options into QMP ones */
    qemu_opt_rename(all_opts, "iops", "throttling.iops-total");
    qemu_opt_rename(all_opts, "iops_rd", "throttling.iops-read");
    qemu_opt_rename(all_opts, "iops_wr", "throttling.iops-write");

    qemu_opt_rename(all_opts, "bps", "throttling.bps-total");
    qemu_opt_rename(all_opts, "bps_rd", "throttling.bps-read");
    qemu_opt_rename(all_opts, "bps_wr", "throttling.bps-write");

646 647 648 649 650 651 652 653
    qemu_opt_rename(all_opts, "iops_max", "throttling.iops-total-max");
    qemu_opt_rename(all_opts, "iops_rd_max", "throttling.iops-read-max");
    qemu_opt_rename(all_opts, "iops_wr_max", "throttling.iops-write-max");

    qemu_opt_rename(all_opts, "bps_max", "throttling.bps-total-max");
    qemu_opt_rename(all_opts, "bps_rd_max", "throttling.bps-read-max");
    qemu_opt_rename(all_opts, "bps_wr_max", "throttling.bps-write-max");

654 655 656
    qemu_opt_rename(all_opts,
                    "iops_size", "throttling.iops-size");

657 658
    qemu_opt_rename(all_opts, "readonly", "read-only");

659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683
    value = qemu_opt_get(all_opts, "cache");
    if (value) {
        int flags = 0;

        if (bdrv_parse_cache_flags(value, &flags) != 0) {
            error_report("invalid cache option");
            return NULL;
        }

        /* Specific options take precedence */
        if (!qemu_opt_get(all_opts, "cache.writeback")) {
            qemu_opt_set_bool(all_opts, "cache.writeback",
                              !!(flags & BDRV_O_CACHE_WB));
        }
        if (!qemu_opt_get(all_opts, "cache.direct")) {
            qemu_opt_set_bool(all_opts, "cache.direct",
                              !!(flags & BDRV_O_NOCACHE));
        }
        if (!qemu_opt_get(all_opts, "cache.no-flush")) {
            qemu_opt_set_bool(all_opts, "cache.no-flush",
                              !!(flags & BDRV_O_NO_FLUSH));
        }
        qemu_opt_unset(all_opts, "cache");
    }

684 685 686 687
    /* Get a QDict for processing the options */
    bs_opts = qdict_new();
    qemu_opts_to_qdict(all_opts, bs_opts);

688 689
    legacy_opts = qemu_opts_create(&qemu_legacy_drive_opts, NULL, 0,
                                   &error_abort);
690 691 692 693 694 695 696
    qemu_opts_absorb_qdict(legacy_opts, bs_opts, &local_err);
    if (error_is_set(&local_err)) {
        qerror_report_err(local_err);
        error_free(local_err);
        goto fail;
    }

697 698 699 700 701 702 703
    /* Deprecated option boot=[on|off] */
    if (qemu_opt_get(legacy_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");
    }

704 705 706 707 708 709 710
    /* Media type */
    value = qemu_opt_get(legacy_opts, "media");
    if (value) {
        if (!strcmp(value, "disk")) {
            media = MEDIA_DISK;
        } else if (!strcmp(value, "cdrom")) {
            media = MEDIA_CDROM;
711
            read_only = true;
712 713 714 715 716 717
        } else {
            error_report("'%s' invalid media", value);
            goto fail;
        }
    }

718
    /* copy-on-read is disabled with a warning for read-only devices */
719
    read_only |= qemu_opt_get_bool(legacy_opts, "read-only", false);
720 721 722 723 724 725 726 727 728 729 730 731
    copy_on_read = qemu_opt_get_bool(legacy_opts, "copy-on-read", false);

    if (read_only && copy_on_read) {
        error_report("warning: disabling copy-on-read on read-only drive");
        copy_on_read = false;
    }

    qdict_put(bs_opts, "read-only",
              qstring_from_str(read_only ? "on" : "off"));
    qdict_put(bs_opts, "copy-on-read",
              qstring_from_str(copy_on_read ? "on" :"off"));

732 733 734 735 736 737 738 739 740 741 742 743 744 745 746
    /* Controller type */
    value = qemu_opt_get(legacy_opts, "if");
    if (value) {
        for (type = 0;
             type < IF_COUNT && strcmp(value, if_name[type]);
             type++) {
        }
        if (type == IF_COUNT) {
            error_report("unsupported bus type '%s'", value);
            goto fail;
        }
    } else {
        type = block_default_type;
    }

747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793
    /* Geometry */
    cyls  = qemu_opt_get_number(legacy_opts, "cyls", 0);
    heads = qemu_opt_get_number(legacy_opts, "heads", 0);
    secs  = qemu_opt_get_number(legacy_opts, "secs", 0);

    if (cyls || heads || secs) {
        if (cyls < 1) {
            error_report("invalid physical cyls number");
            goto fail;
        }
        if (heads < 1) {
            error_report("invalid physical heads number");
            goto fail;
        }
        if (secs < 1) {
            error_report("invalid physical secs number");
            goto fail;
        }
    }

    translation = BIOS_ATA_TRANSLATION_AUTO;
    value = qemu_opt_get(legacy_opts, "trans");
    if (value != NULL) {
        if (!cyls) {
            error_report("'%s' trans must be used with cyls, heads and secs",
                         value);
            goto fail;
        }
        if (!strcmp(value, "none")) {
            translation = BIOS_ATA_TRANSLATION_NONE;
        } else if (!strcmp(value, "lba")) {
            translation = BIOS_ATA_TRANSLATION_LBA;
        } else if (!strcmp(value, "auto")) {
            translation = BIOS_ATA_TRANSLATION_AUTO;
        } else {
            error_report("'%s' invalid translation type", value);
            goto fail;
        }
    }

    if (media == MEDIA_CDROM) {
        if (cyls || secs || heads) {
            error_report("CHS can't be set with media=cdrom");
            goto fail;
        }
    }

794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850
    /* Device address specified by bus/unit or index.
     * If none was specified, try to find the first free one. */
    bus_id  = qemu_opt_get_number(legacy_opts, "bus", 0);
    unit_id = qemu_opt_get_number(legacy_opts, "unit", -1);
    index   = qemu_opt_get_number(legacy_opts, "index", -1);

    max_devs = if_max_devs[type];

    if (index != -1) {
        if (bus_id != 0 || unit_id != -1) {
            error_report("index cannot be used with bus and unit");
            goto fail;
        }
        bus_id = drive_index_to_bus_id(type, index);
        unit_id = drive_index_to_unit_id(type, index);
    }

    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++;
           }
       }
    }

    if (max_devs && unit_id >= max_devs) {
        error_report("unit %d too big (max is %d)", unit_id, max_devs - 1);
        goto fail;
    }

    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);
        goto fail;
    }

    /* no id supplied -> create one */
    if (qemu_opts_id(all_opts) == NULL) {
        char *new_id;
        const char *mediastr = "";
        if (type == IF_IDE || type == IF_SCSI) {
            mediastr = (media == MEDIA_CDROM) ? "-cd" : "-hd";
        }
        if (max_devs) {
            new_id = g_strdup_printf("%s%i%s%i", if_name[type], bus_id,
                                     mediastr, unit_id);
        } else {
            new_id = g_strdup_printf("%s%s%i", if_name[type],
                                     mediastr, unit_id);
        }
        qdict_put(bs_opts, "id", qstring_from_str(new_id));
        g_free(new_id);
    }

851 852 853 854 855 856 857 858 859
    /* Add virtio block device */
    devaddr = qemu_opt_get(legacy_opts, "addr");
    if (devaddr && type != IF_VIRTIO) {
        error_report("addr is not supported by this bus type");
        goto fail;
    }

    if (type == IF_VIRTIO) {
        QemuOpts *devopts;
860 861
        devopts = qemu_opts_create(qemu_find_opts("device"), NULL, 0,
                                   &error_abort);
862 863 864 865 866 867 868 869 870 871 872
        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", qdict_get_str(bs_opts, "id"));
        if (devaddr) {
            qemu_opt_set(devopts, "addr", devaddr);
        }
    }

873 874
    filename = qemu_opt_get(legacy_opts, "file");

875
    /* Actual block device init: Functionality shared with blockdev-add */
876
    dinfo = blockdev_init(filename, bs_opts, type, &local_err);
877
    if (dinfo == NULL) {
878 879 880 881
        if (error_is_set(&local_err)) {
            qerror_report_err(local_err);
            error_free(local_err);
        }
882
        goto fail;
883 884
    } else {
        assert(!error_is_set(&local_err));
885 886 887 888
    }

    /* Set legacy DriveInfo fields */
    dinfo->enable_auto_del = true;
889
    dinfo->opts = all_opts;
890

891 892 893 894 895
    dinfo->cyls = cyls;
    dinfo->heads = heads;
    dinfo->secs = secs;
    dinfo->trans = translation;

896 897
    dinfo->bus = bus_id;
    dinfo->unit = unit_id;
898
    dinfo->devaddr = devaddr;
899

900 901 902 903 904 905 906 907 908 909 910
    switch(type) {
    case IF_IDE:
    case IF_SCSI:
    case IF_XEN:
    case IF_NONE:
        dinfo->media_cd = media == MEDIA_CDROM;
        break;
    default:
        break;
    }

911
fail:
912
    qemu_opts_del(legacy_opts);
913
    return dinfo;
914 915
}

916 917 918
void do_commit(Monitor *mon, const QDict *qdict)
{
    const char *device = qdict_get_str(qdict, "device");
919
    BlockDriverState *bs;
920
    int ret;
921

922
    if (!strcmp(device, "all")) {
923
        ret = bdrv_commit_all();
924 925
    } else {
        bs = bdrv_find(device);
926
        if (!bs) {
927
            monitor_printf(mon, "Device '%s' not found\n", device);
928
            return;
929
        }
930
        ret = bdrv_commit(bs);
931 932 933 934
    }
    if (ret < 0) {
        monitor_printf(mon, "'commit' error for '%s': %s\n", device,
                       strerror(-ret));
935 936 937
    }
}

938 939
static void blockdev_do_action(int kind, void *data, Error **errp)
{
940 941
    TransactionAction action;
    TransactionActionList list;
942 943 944 945 946 947 948 949

    action.kind = kind;
    action.data = data;
    list.value = &action;
    list.next = NULL;
    qmp_transaction(&list, errp);
}

950 951 952 953 954
void qmp_blockdev_snapshot_sync(bool has_device, const char *device,
                                bool has_node_name, const char *node_name,
                                const char *snapshot_file,
                                bool has_snapshot_node_name,
                                const char *snapshot_node_name,
955
                                bool has_format, const char *format,
956
                                bool has_mode, NewImageMode mode, Error **errp)
957
{
958
    BlockdevSnapshot snapshot = {
959
        .has_device = has_device,
960
        .device = (char *) device,
961 962
        .has_node_name = has_node_name,
        .node_name = (char *) node_name,
963
        .snapshot_file = (char *) snapshot_file,
964 965
        .has_snapshot_node_name = has_snapshot_node_name,
        .snapshot_node_name = (char *) snapshot_node_name,
966 967 968 969 970
        .has_format = has_format,
        .format = (char *) format,
        .has_mode = has_mode,
        .mode = mode,
    };
971 972
    blockdev_do_action(TRANSACTION_ACTION_KIND_BLOCKDEV_SNAPSHOT_SYNC,
                       &snapshot, errp);
973 974
}

975 976 977 978 979 980 981 982 983 984 985 986 987
void qmp_blockdev_snapshot_internal_sync(const char *device,
                                         const char *name,
                                         Error **errp)
{
    BlockdevSnapshotInternal snapshot = {
        .device = (char *) device,
        .name = (char *) name
    };

    blockdev_do_action(TRANSACTION_ACTION_KIND_BLOCKDEV_SNAPSHOT_INTERNAL_SYNC,
                       &snapshot, errp);
}

988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048
SnapshotInfo *qmp_blockdev_snapshot_delete_internal_sync(const char *device,
                                                         bool has_id,
                                                         const char *id,
                                                         bool has_name,
                                                         const char *name,
                                                         Error **errp)
{
    BlockDriverState *bs = bdrv_find(device);
    QEMUSnapshotInfo sn;
    Error *local_err = NULL;
    SnapshotInfo *info = NULL;
    int ret;

    if (!bs) {
        error_set(errp, QERR_DEVICE_NOT_FOUND, device);
        return NULL;
    }

    if (!has_id) {
        id = NULL;
    }

    if (!has_name) {
        name = NULL;
    }

    if (!id && !name) {
        error_setg(errp, "Name or id must be provided");
        return NULL;
    }

    ret = bdrv_snapshot_find_by_id_and_name(bs, id, name, &sn, &local_err);
    if (error_is_set(&local_err)) {
        error_propagate(errp, local_err);
        return NULL;
    }
    if (!ret) {
        error_setg(errp,
                   "Snapshot with id '%s' and name '%s' does not exist on "
                   "device '%s'",
                   STR_OR_NULL(id), STR_OR_NULL(name), device);
        return NULL;
    }

    bdrv_snapshot_delete(bs, id, name, &local_err);
    if (error_is_set(&local_err)) {
        error_propagate(errp, local_err);
        return NULL;
    }

    info = g_malloc0(sizeof(SnapshotInfo));
    info->id = g_strdup(sn.id_str);
    info->name = g_strdup(sn.name);
    info->date_nsec = sn.date_nsec;
    info->date_sec = sn.date_sec;
    info->vm_state_size = sn.vm_state_size;
    info->vm_clock_nsec = sn.vm_clock_nsec % 1000000000;
    info->vm_clock_sec = sn.vm_clock_nsec / 1000000000;

    return info;
}
1049 1050

/* New and old BlockDriverState structs for group snapshots */
1051

1052
typedef struct BlkTransactionState BlkTransactionState;
1053 1054 1055 1056 1057 1058 1059

/* 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. */
1060
    void (*prepare)(BlkTransactionState *common, Error **errp);
1061
    /* Commit the changes, can be NULL. */
1062
    void (*commit)(BlkTransactionState *common);
1063
    /* Abort the changes on fail, can be NULL. */
1064
    void (*abort)(BlkTransactionState *common);
1065
    /* Clean up resource in the end, can be NULL. */
1066
    void (*clean)(BlkTransactionState *common);
1067 1068 1069 1070 1071 1072 1073
} 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().
 */
1074
struct BlkTransactionState {
1075
    TransactionAction *action;
1076
    const BdrvActionOps *ops;
1077
    QSIMPLEQ_ENTRY(BlkTransactionState) entry;
1078 1079
};

1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190
/* internal snapshot private data */
typedef struct InternalSnapshotState {
    BlkTransactionState common;
    BlockDriverState *bs;
    QEMUSnapshotInfo sn;
} InternalSnapshotState;

static void internal_snapshot_prepare(BlkTransactionState *common,
                                      Error **errp)
{
    const char *device;
    const char *name;
    BlockDriverState *bs;
    QEMUSnapshotInfo old_sn, *sn;
    bool ret;
    qemu_timeval tv;
    BlockdevSnapshotInternal *internal;
    InternalSnapshotState *state;
    int ret1;

    g_assert(common->action->kind ==
             TRANSACTION_ACTION_KIND_BLOCKDEV_SNAPSHOT_INTERNAL_SYNC);
    internal = common->action->blockdev_snapshot_internal_sync;
    state = DO_UPCAST(InternalSnapshotState, common, common);

    /* 1. parse input */
    device = internal->device;
    name = internal->name;

    /* 2. check for validation */
    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 (bdrv_is_read_only(bs)) {
        error_set(errp, QERR_DEVICE_IS_READ_ONLY, device);
        return;
    }

    if (!bdrv_can_snapshot(bs)) {
        error_set(errp, QERR_BLOCK_FORMAT_FEATURE_NOT_SUPPORTED,
                  bs->drv->format_name, device, "internal snapshot");
        return;
    }

    if (!strlen(name)) {
        error_setg(errp, "Name is empty");
        return;
    }

    /* check whether a snapshot with name exist */
    ret = bdrv_snapshot_find_by_id_and_name(bs, NULL, name, &old_sn, errp);
    if (error_is_set(errp)) {
        return;
    } else if (ret) {
        error_setg(errp,
                   "Snapshot with name '%s' already exists on device '%s'",
                   name, device);
        return;
    }

    /* 3. take the snapshot */
    sn = &state->sn;
    pstrcpy(sn->name, sizeof(sn->name), name);
    qemu_gettimeofday(&tv);
    sn->date_sec = tv.tv_sec;
    sn->date_nsec = tv.tv_usec * 1000;
    sn->vm_clock_nsec = qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL);

    ret1 = bdrv_snapshot_create(bs, sn);
    if (ret1 < 0) {
        error_setg_errno(errp, -ret1,
                         "Failed to create snapshot '%s' on device '%s'",
                         name, device);
        return;
    }

    /* 4. succeed, mark a snapshot is created */
    state->bs = bs;
}

static void internal_snapshot_abort(BlkTransactionState *common)
{
    InternalSnapshotState *state =
                             DO_UPCAST(InternalSnapshotState, common, common);
    BlockDriverState *bs = state->bs;
    QEMUSnapshotInfo *sn = &state->sn;
    Error *local_error = NULL;

    if (!bs) {
        return;
    }

    if (bdrv_snapshot_delete(bs, sn->id_str, sn->name, &local_error) < 0) {
        error_report("Failed to delete snapshot with id '%s' and name '%s' on "
                     "device '%s' in abort: %s",
                     sn->id_str,
                     sn->name,
                     bdrv_get_device_name(bs),
                     error_get_pretty(local_error));
        error_free(local_error);
    }
}

1191
/* external snapshot private data */
1192 1193
typedef struct ExternalSnapshotState {
    BlkTransactionState common;
1194 1195
    BlockDriverState *old_bs;
    BlockDriverState *new_bs;
1196
} ExternalSnapshotState;
1197

1198
static void external_snapshot_prepare(BlkTransactionState *common,
1199 1200 1201 1202
                                      Error **errp)
{
    BlockDriver *drv;
    int flags, ret;
1203
    QDict *options = NULL;
1204
    Error *local_err = NULL;
1205
    bool has_device = false;
1206
    const char *device;
1207 1208 1209 1210
    bool has_node_name = false;
    const char *node_name;
    bool has_snapshot_node_name = false;
    const char *snapshot_node_name;
1211 1212 1213
    const char *new_image_file;
    const char *format = "qcow2";
    enum NewImageMode mode = NEW_IMAGE_MODE_ABSOLUTE_PATHS;
1214 1215
    ExternalSnapshotState *state =
                             DO_UPCAST(ExternalSnapshotState, common, common);
1216
    TransactionAction *action = common->action;
1217

1218
    /* get parameters */
1219
    g_assert(action->kind == TRANSACTION_ACTION_KIND_BLOCKDEV_SNAPSHOT_SYNC);
1220

1221
    has_device = action->blockdev_snapshot_sync->has_device;
1222
    device = action->blockdev_snapshot_sync->device;
1223 1224 1225 1226 1227 1228
    has_node_name = action->blockdev_snapshot_sync->has_node_name;
    node_name = action->blockdev_snapshot_sync->node_name;
    has_snapshot_node_name =
        action->blockdev_snapshot_sync->has_snapshot_node_name;
    snapshot_node_name = action->blockdev_snapshot_sync->snapshot_node_name;

1229 1230 1231 1232 1233 1234 1235 1236 1237
    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 */
1238 1239 1240 1241 1242 1243
    drv = bdrv_find_format(format);
    if (!drv) {
        error_set(errp, QERR_INVALID_BLOCK_FORMAT, format);
        return;
    }

1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258
    state->old_bs = bdrv_lookup_bs(has_device ? device : NULL,
                                   has_node_name ? node_name : NULL,
                                   &local_err);
    if (error_is_set(&local_err)) {
        error_propagate(errp, local_err);
        return;
    }

    if (has_node_name && !has_snapshot_node_name) {
        error_setg(errp, "New snapshot node name missing");
        return;
    }

    if (has_snapshot_node_name && bdrv_find_node(snapshot_node_name)) {
        error_setg(errp, "New snapshot node name already existing");
1259 1260 1261
        return;
    }

1262
    if (!bdrv_is_inserted(state->old_bs)) {
1263 1264 1265 1266
        error_set(errp, QERR_DEVICE_HAS_NO_MEDIUM, device);
        return;
    }

1267
    if (bdrv_in_use(state->old_bs)) {
1268 1269 1270 1271
        error_set(errp, QERR_DEVICE_IN_USE, device);
        return;
    }

1272 1273
    if (!bdrv_is_read_only(state->old_bs)) {
        if (bdrv_flush(state->old_bs)) {
1274 1275 1276 1277 1278
            error_set(errp, QERR_IO_ERROR);
            return;
        }
    }

1279
    if (!bdrv_is_first_non_filter(state->old_bs)) {
1280 1281 1282 1283
        error_set(errp, QERR_FEATURE_DISABLED, "snapshot");
        return;
    }

1284
    flags = state->old_bs->open_flags;
1285 1286 1287 1288

    /* create new image w/backing file */
    if (mode != NEW_IMAGE_MODE_EXISTING) {
        bdrv_img_create(new_image_file, format,
1289 1290
                        state->old_bs->filename,
                        state->old_bs->drv->format_name,
1291 1292 1293 1294 1295 1296 1297
                        NULL, -1, flags, &local_err, false);
        if (error_is_set(&local_err)) {
            error_propagate(errp, local_err);
            return;
        }
    }

1298 1299 1300 1301 1302 1303
    if (has_snapshot_node_name) {
        options = qdict_new();
        qdict_put(options, "node-name",
                  qstring_from_str(snapshot_node_name));
    }

1304
    /* We will manually add the backing_hd field to the bs later */
1305
    state->new_bs = bdrv_new("");
1306 1307
    /* TODO Inherit bs->options or only take explicit options with an
     * extended QMP command? */
1308
    ret = bdrv_open(state->new_bs, new_image_file, options,
1309
                    flags | BDRV_O_NO_BACKING, drv, &local_err);
1310
    if (ret != 0) {
1311
        error_propagate(errp, local_err);
1312
    }
1313 1314

    QDECREF(options);
1315 1316
}

1317
static void external_snapshot_commit(BlkTransactionState *common)
1318
{
1319 1320
    ExternalSnapshotState *state =
                             DO_UPCAST(ExternalSnapshotState, common, common);
1321

1322 1323
    /* This removes our old bs and adds the new bs */
    bdrv_append(state->new_bs, state->old_bs);
1324 1325 1326
    /* 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 */
1327
    bdrv_reopen(state->new_bs, state->new_bs->open_flags & ~BDRV_O_RDWR,
1328 1329 1330
                NULL);
}

1331
static void external_snapshot_abort(BlkTransactionState *common)
1332
{
1333 1334 1335
    ExternalSnapshotState *state =
                             DO_UPCAST(ExternalSnapshotState, common, common);
    if (state->new_bs) {
Fam Zheng's avatar
Fam Zheng committed
1336
        bdrv_unref(state->new_bs);