blockdev.c 52.2 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"
P
Paolo Bonzini 已提交
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
#include "sysemu/sysemu.h"
42
#include "block/block_int.h"
L
Luiz Capitulino 已提交
43
#include "qmp-commands.h"
S
Stefan Hajnoczi 已提交
44
#include "trace.h"
45
#include "sysemu/arch_init.h"
46

47
static QTAILQ_HEAD(drivelist, DriveInfo) drives = QTAILQ_HEAD_INITIALIZER(drives);
48
extern QemuOptsList qemu_common_drive_opts;
49

50 51 52 53 54 55 56 57 58 59 60 61 62
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] = {
63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78
    /*
     * 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,
79 80
};

81 82 83 84 85 86 87 88 89 90 91
/*
 * 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);

92 93 94
    if (bs->job) {
        block_job_cancel(bs->job);
    }
95 96 97
    if (dinfo) {
        dinfo->auto_del = 1;
    }
98 99 100 101 102 103
}

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

104
    if (dinfo && dinfo->auto_del) {
105
        drive_put_ref(dinfo);
106 107 108
    }
}

109 110 111 112 113 114 115 116 117 118 119 120
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;
}

121 122 123 124 125 126
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,
127
                    const char *optstr)
128 129
{
    QemuOpts *opts;
130
    char buf[32];
131

132
    opts = drive_def(optstr);
133 134 135
    if (!opts) {
        return NULL;
    }
136 137 138 139 140 141 142
    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);
    }
143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163
    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;
}

164 165 166 167 168 169 170
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));
}

171 172 173 174 175 176 177 178 179 180 181 182 183 184
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;
}

185 186 187 188 189 190 191 192 193 194
/* 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]++);
}

195
DriveInfo *drive_get_by_blockdev(BlockDriverState *bs)
196 197 198 199
{
    DriveInfo *dinfo;

    QTAILQ_FOREACH(dinfo, &drives, next) {
200 201 202
        if (dinfo->bdrv == bs) {
            return dinfo;
        }
203
    }
204
    return NULL;
205 206 207 208
}

static void bdrv_format_print(void *opaque, const char *name)
{
209
    error_printf(" %s", name);
210 211
}

212
static void drive_uninit(DriveInfo *dinfo)
213 214 215
{
    qemu_opts_del(dinfo->opts);
    bdrv_delete(dinfo->bdrv);
216
    g_free(dinfo->id);
217
    QTAILQ_REMOVE(&drives, dinfo, next);
218
    g_free(dinfo->serial);
219
    g_free(dinfo);
220 221
}

222 223 224 225 226 227 228 229 230 231 232 233 234
void drive_put_ref(DriveInfo *dinfo)
{
    assert(dinfo->refcount);
    if (--dinfo->refcount == 0) {
        drive_uninit(dinfo);
    }
}

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

235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265
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);
}

266
static int parse_block_error_action(const char *buf, bool is_read)
267 268
{
    if (!strcmp(buf, "ignore")) {
269
        return BLOCKDEV_ON_ERROR_IGNORE;
270
    } else if (!is_read && !strcmp(buf, "enospc")) {
271
        return BLOCKDEV_ON_ERROR_ENOSPC;
272
    } else if (!strcmp(buf, "stop")) {
273
        return BLOCKDEV_ON_ERROR_STOP;
274
    } else if (!strcmp(buf, "report")) {
275
        return BLOCKDEV_ON_ERROR_REPORT;
276
    } else {
277 278
        error_report("'%s' invalid %s error action",
                     buf, is_read ? "read" : "write");
279 280 281 282
        return -1;
    }
}

283
static bool do_check_io_limits(BlockIOLimit *io_limits, Error **errp)
284 285 286 287 288 289 290 291 292 293 294 295 296
{
    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) {
297 298
        error_setg(errp, "bps(iops) and bps_rd/bps_wr(iops_rd/iops_wr) "
                         "cannot be used at the same time");
299 300 301
        return false;
    }

302 303 304 305 306 307 308 309 310 311
    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;
    }

312 313 314
    return true;
}

315
DriveInfo *drive_init(QemuOpts *all_opts, BlockInterfaceType block_default_type)
316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332
{
    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;
333
    BlockIOLimit io_limits;
334
    int snapshot = 0;
335
    bool copy_on_read;
336
    int ret;
337
    Error *error = NULL;
338 339 340
    QemuOpts *opts;
    QDict *bs_opts;
    const char *id;
341 342 343 344

    translation = BIOS_ATA_TRANSLATION_AUTO;
    media = MEDIA_DISK;

345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367
    /* 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");
    }

368 369 370 371 372 373 374 375 376 377 378
    /* 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);
379
    copy_on_read = qemu_opt_get_bool(opts, "copy-on-read", false);
380 381 382 383 384

    file = qemu_opt_get(opts, "file");
    serial = qemu_opt_get(opts, "serial");

    if ((buf = qemu_opt_get(opts, "if")) != NULL) {
385 386 387
        for (type = 0; type < IF_COUNT && strcmp(buf, if_name[type]); type++)
            ;
        if (type == IF_COUNT) {
388
            error_report("unsupported bus type '%s'", buf);
389 390
            return NULL;
	}
391
    } else {
392
        type = block_default_type;
393
    }
394

395
    max_devs = if_max_devs[type];
396 397

    if (cyls || heads || secs) {
398
        if (cyls < 1) {
399
            error_report("invalid physical cyls number");
400 401
	    return NULL;
	}
402
        if (heads < 1) {
403
            error_report("invalid physical heads number");
404 405
	    return NULL;
	}
406
        if (secs < 1) {
407
            error_report("invalid physical secs number");
408 409 410 411 412 413
	    return NULL;
	}
    }

    if ((buf = qemu_opt_get(opts, "trans")) != NULL) {
        if (!cyls) {
414
            error_report("'%s' trans must be used with cyls, heads and secs",
415
                         buf);
416 417 418 419 420 421 422 423 424
            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 {
425
            error_report("'%s' invalid translation type", buf);
426 427 428 429 430 431 432 433 434
	    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) {
435
                error_report("CHS can't be set with media=%s", buf);
436 437 438 439
	        return NULL;
            }
	    media = MEDIA_CDROM;
	} else {
440
	    error_report("'%s' invalid media", buf);
441 442 443 444
	    return NULL;
	}
    }

445 446 447 448 449 450 451
    if ((buf = qemu_opt_get(opts, "discard")) != NULL) {
        if (bdrv_parse_discard_flags(buf, &bdrv_flags) != 0) {
            error_report("invalid discard option");
            return NULL;
        }
    }

452
    bdrv_flags |= BDRV_O_CACHE_WB;
453
    if ((buf = qemu_opt_get(opts, "cache")) != NULL) {
454 455 456
        if (bdrv_parse_cache_flags(buf, &bdrv_flags) != 0) {
            error_report("invalid cache option");
            return NULL;
457 458 459 460 461 462 463 464 465 466
        }
    }

#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 {
467
           error_report("invalid aio option");
468 469 470 471 472 473
           return NULL;
        }
    }
#endif

    if ((buf = qemu_opt_get(opts, "format")) != NULL) {
474 475 476 477 478
        if (is_help_option(buf)) {
            error_printf("Supported formats:");
            bdrv_iterate_format(bdrv_format_print, NULL);
            error_printf("\n");
            return NULL;
479
        }
480
        drv = bdrv_find_whitelisted_format(buf, ro);
481
        if (!drv) {
482
            error_report("'%s' invalid format", buf);
483 484 485 486
            return NULL;
        }
    }

487 488 489 490 491 492 493 494 495 496 497 498 499 500
    /* 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);

501 502 503
    if (!do_check_io_limits(&io_limits, &error)) {
        error_report("%s", error_get_pretty(error));
        error_free(error);
504 505 506
        return NULL;
    }

507 508 509 510 511 512
    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");
    }

513
    on_write_error = BLOCKDEV_ON_ERROR_ENOSPC;
514 515
    if ((buf = qemu_opt_get(opts, "werror")) != NULL) {
        if (type != IF_IDE && type != IF_SCSI && type != IF_VIRTIO && type != IF_NONE) {
516
            error_report("werror is not supported by this bus type");
517 518 519 520 521 522 523 524 525
            return NULL;
        }

        on_write_error = parse_block_error_action(buf, 0);
        if (on_write_error < 0) {
            return NULL;
        }
    }

526
    on_read_error = BLOCKDEV_ON_ERROR_REPORT;
527
    if ((buf = qemu_opt_get(opts, "rerror")) != NULL) {
K
Kevin Wolf 已提交
528
        if (type != IF_IDE && type != IF_VIRTIO && type != IF_SCSI && type != IF_NONE) {
529
            error_report("rerror is not supported by this bus type");
530 531 532 533 534 535 536 537 538 539 540
            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) {
541
            error_report("addr is not supported by this bus type");
542 543 544 545 546 547 548 549
            return NULL;
        }
    }

    /* compute bus and unit according index */

    if (index != -1) {
        if (bus_id != 0 || unit_id != -1) {
550
            error_report("index cannot be used with bus and unit");
551 552
            return NULL;
        }
553 554
        bus_id = drive_index_to_bus_id(type, index);
        unit_id = drive_index_to_unit_id(type, index);
555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574
    }

    /* 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) {
575 576
        error_report("unit %d too big (max is %d)",
                     unit_id, max_devs - 1);
577 578 579 580
        return NULL;
    }

    /*
581
     * catch multiple definitions
582 583 584
     */

    if (drive_get(type, bus_id, unit_id) != NULL) {
585 586
        error_report("drive with bus=%d, unit=%d (index=%d) exists",
                     bus_id, unit_id, index);
587 588 589 590 591
        return NULL;
    }

    /* init */

592
    dinfo = g_malloc0(sizeof(*dinfo));
593
    if ((buf = qemu_opts_id(opts)) != NULL) {
594
        dinfo->id = g_strdup(buf);
595 596
    } else {
        /* no id supplied -> create one */
597
        dinfo->id = g_malloc0(32);
598 599 600 601
        if (type == IF_IDE || type == IF_SCSI)
            mediastr = (media == MEDIA_CDROM) ? "-cd" : "-hd";
        if (max_devs)
            snprintf(dinfo->id, 32, "%s%i%s%i",
602
                     if_name[type], bus_id, mediastr, unit_id);
603 604
        else
            snprintf(dinfo->id, 32, "%s%s%i",
605
                     if_name[type], mediastr, unit_id);
606 607
    }
    dinfo->bdrv = bdrv_new(dinfo->id);
608 609
    dinfo->bdrv->open_flags = snapshot ? BDRV_O_SNAPSHOT : 0;
    dinfo->bdrv->read_only = ro;
610 611 612 613
    dinfo->devaddr = devaddr;
    dinfo->type = type;
    dinfo->bus = bus_id;
    dinfo->unit = unit_id;
614 615 616 617
    dinfo->cyls = cyls;
    dinfo->heads = heads;
    dinfo->secs = secs;
    dinfo->trans = translation;
618
    dinfo->opts = all_opts;
619
    dinfo->refcount = 1;
620 621 622
    if (serial != NULL) {
        dinfo->serial = g_strdup(serial);
    }
623 624
    QTAILQ_INSERT_TAIL(&drives, dinfo, next);

625 626
    bdrv_set_on_error(dinfo->bdrv, on_read_error, on_write_error);

627 628 629
    /* disk I/O throttling */
    bdrv_set_io_limits(dinfo->bdrv, &io_limits);

630 631 632 633 634
    switch(type) {
    case IF_IDE:
    case IF_SCSI:
    case IF_XEN:
    case IF_NONE:
635
        dinfo->media_cd = media == MEDIA_CDROM;
636 637 638 639 640 641 642
        break;
    case IF_SD:
    case IF_FLOPPY:
    case IF_PFLASH:
    case IF_MTD:
        break;
    case IF_VIRTIO:
643
    {
644
        /* add virtio block device */
645 646
        QemuOpts *devopts;
        devopts = qemu_opts_create_nofail(qemu_find_opts("device"));
647
        if (arch_type == QEMU_ARCH_S390X) {
648
            qemu_opt_set(devopts, "driver", "virtio-blk-s390");
649
        } else {
650
            qemu_opt_set(devopts, "driver", "virtio-blk-pci");
651
        }
652
        qemu_opt_set(devopts, "drive", dinfo->id);
653
        if (devaddr)
654
            qemu_opt_set(devopts, "addr", devaddr);
655
        break;
656
    }
657
    default:
658 659
        abort();
    }
660
    if (!file || !*file) {
661 662 663 664 665
        if (qdict_size(bs_opts)) {
            file = NULL;
        } else {
            return dinfo;
        }
666 667 668 669 670 671 672
    }
    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);
    }

673 674 675 676
    if (copy_on_read) {
        bdrv_flags |= BDRV_O_COPY_ON_READ;
    }

677 678 679 680
    if (runstate_check(RUN_STATE_INMIGRATE)) {
        bdrv_flags |= BDRV_O_INCOMING;
    }

681 682 683 684
    if (media == MEDIA_CDROM) {
        /* CDROM is fine for any interface, don't check.  */
        ro = 1;
    } else if (ro == 1) {
685 686
        if (type != IF_SCSI && type != IF_VIRTIO && type != IF_FLOPPY &&
            type != IF_NONE && type != IF_PFLASH) {
687
            error_report("readonly not supported by this bus type");
688
            goto err;
689 690 691 692 693
        }
    }

    bdrv_flags |= ro ? 0 : BDRV_O_RDWR;

694 695 696 697
    if (ro && copy_on_read) {
        error_report("warning: disabling copy_on_read on readonly drive");
    }

698 699 700
    ret = bdrv_open(dinfo->bdrv, file, bs_opts, bdrv_flags, drv);
    bs_opts = NULL;

701
    if (ret < 0) {
702 703
        if (ret == -EMEDIUMTYPE) {
            error_report("could not open disk image %s: not in %s format",
704
                         file ?: dinfo->id, drv->format_name);
705 706
        } else {
            error_report("could not open disk image %s: %s",
707
                         file ?: dinfo->id, strerror(-ret));
708
        }
709
        goto err;
710 711 712 713
    }

    if (bdrv_key_required(dinfo->bdrv))
        autostart = 0;
714 715 716

    qemu_opts_del(opts);

717
    return dinfo;
718 719

err:
720 721
    qemu_opts_del(opts);
    QDECREF(bs_opts);
722
    bdrv_delete(dinfo->bdrv);
723
    g_free(dinfo->id);
724
    QTAILQ_REMOVE(&drives, dinfo, next);
725
    g_free(dinfo);
726
    return NULL;
727 728 729 730 731
}

void do_commit(Monitor *mon, const QDict *qdict)
{
    const char *device = qdict_get_str(qdict, "device");
732
    BlockDriverState *bs;
733
    int ret;
734

735
    if (!strcmp(device, "all")) {
736
        ret = bdrv_commit_all();
737 738
    } else {
        bs = bdrv_find(device);
739
        if (!bs) {
740
            monitor_printf(mon, "Device '%s' not found\n", device);
741
            return;
742
        }
743
        ret = bdrv_commit(bs);
744 745 746 747
    }
    if (ret < 0) {
        monitor_printf(mon, "'commit' error for '%s': %s\n", device,
                       strerror(-ret));
748 749 750
    }
}

751 752
static void blockdev_do_action(int kind, void *data, Error **errp)
{
753 754
    TransactionAction action;
    TransactionActionList list;
755 756 757 758 759 760 761 762

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

763 764
void qmp_blockdev_snapshot_sync(const char *device, const char *snapshot_file,
                                bool has_format, const char *format,
765
                                bool has_mode, enum NewImageMode mode,
766
                                Error **errp)
767
{
768 769 770 771 772 773 774 775
    BlockdevSnapshot snapshot = {
        .device = (char *) device,
        .snapshot_file = (char *) snapshot_file,
        .has_format = has_format,
        .format = (char *) format,
        .has_mode = has_mode,
        .mode = mode,
    };
776 777
    blockdev_do_action(TRANSACTION_ACTION_KIND_BLOCKDEV_SNAPSHOT_SYNC,
                       &snapshot, errp);
778 779
}

780 781

/* New and old BlockDriverState structs for group snapshots */
782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805

typedef struct BlkTransactionStates BlkTransactionStates;

/* Only prepare() may fail. In a single transaction, only one of commit() or
   abort() will be called, clean() will always be called if it present. */
typedef struct BdrvActionOps {
    /* Size of state struct, in bytes. */
    size_t instance_size;
    /* Prepare the work, must NOT be NULL. */
    void (*prepare)(BlkTransactionStates *common, Error **errp);
    /* Commit the changes, must NOT be NULL. */
    void (*commit)(BlkTransactionStates *common);
    /* Abort the changes on fail, can be NULL. */
    void (*abort)(BlkTransactionStates *common);
    /* Clean up resource in the end, can be NULL. */
    void (*clean)(BlkTransactionStates *common);
} BdrvActionOps;

/*
 * This structure must be arranged as first member in child type, assuming
 * that compiler will also arrange it to the same address with parent instance.
 * Later it will be used in free().
 */
struct BlkTransactionStates {
806
    TransactionAction *action;
807 808 809 810 811 812 813
    const BdrvActionOps *ops;
    QSIMPLEQ_ENTRY(BlkTransactionStates) entry;
};

/* external snapshot private data */
typedef struct ExternalSnapshotStates {
    BlkTransactionStates common;
814 815
    BlockDriverState *old_bs;
    BlockDriverState *new_bs;
816
} ExternalSnapshotStates;
817

818
static void external_snapshot_prepare(BlkTransactionStates *common,
819 820 821 822 823 824
                                      Error **errp)
{
    BlockDriver *proto_drv;
    BlockDriver *drv;
    int flags, ret;
    Error *local_err = NULL;
825 826 827 828
    const char *device;
    const char *new_image_file;
    const char *format = "qcow2";
    enum NewImageMode mode = NEW_IMAGE_MODE_ABSOLUTE_PATHS;
829 830
    ExternalSnapshotStates *states =
                             DO_UPCAST(ExternalSnapshotStates, common, common);
831
    TransactionAction *action = common->action;
832

833
    /* get parameters */
834
    g_assert(action->kind == TRANSACTION_ACTION_KIND_BLOCKDEV_SNAPSHOT_SYNC);
835 836 837 838 839 840 841 842 843 844 845

    device = action->blockdev_snapshot_sync->device;
    new_image_file = action->blockdev_snapshot_sync->snapshot_file;
    if (action->blockdev_snapshot_sync->has_format) {
        format = action->blockdev_snapshot_sync->format;
    }
    if (action->blockdev_snapshot_sync->has_mode) {
        mode = action->blockdev_snapshot_sync->mode;
    }

    /* start processing */
846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905
    drv = bdrv_find_format(format);
    if (!drv) {
        error_set(errp, QERR_INVALID_BLOCK_FORMAT, format);
        return;
    }

    states->old_bs = bdrv_find(device);
    if (!states->old_bs) {
        error_set(errp, QERR_DEVICE_NOT_FOUND, device);
        return;
    }

    if (!bdrv_is_inserted(states->old_bs)) {
        error_set(errp, QERR_DEVICE_HAS_NO_MEDIUM, device);
        return;
    }

    if (bdrv_in_use(states->old_bs)) {
        error_set(errp, QERR_DEVICE_IN_USE, device);
        return;
    }

    if (!bdrv_is_read_only(states->old_bs)) {
        if (bdrv_flush(states->old_bs)) {
            error_set(errp, QERR_IO_ERROR);
            return;
        }
    }

    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);
        return;
    }

    /* 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);
            return;
        }
    }

    /* 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);
    }
}

906
static void external_snapshot_commit(BlkTransactionStates *common)
907
{
908 909 910
    ExternalSnapshotStates *states =
                             DO_UPCAST(ExternalSnapshotStates, common, common);

911 912 913 914 915 916 917 918 919
    /* 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);
}

920
static void external_snapshot_abort(BlkTransactionStates *common)
921
{
922 923
    ExternalSnapshotStates *states =
                             DO_UPCAST(ExternalSnapshotStates, common, common);
924 925 926 927 928
    if (states->new_bs) {
        bdrv_delete(states->new_bs);
    }
}

929
static const BdrvActionOps actions[] = {
930
    [TRANSACTION_ACTION_KIND_BLOCKDEV_SNAPSHOT_SYNC] = {
931 932 933 934 935 936 937
        .instance_size = sizeof(ExternalSnapshotStates),
        .prepare  = external_snapshot_prepare,
        .commit   = external_snapshot_commit,
        .abort = external_snapshot_abort,
    },
};

938 939 940 941 942
/*
 * '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
 */
943
void qmp_transaction(TransactionActionList *dev_list, Error **errp)
944
{
945
    TransactionActionList *dev_entry = dev_list;
946
    BlkTransactionStates *states, *next;
947
    Error *local_err = NULL;
948

949
    QSIMPLEQ_HEAD(snap_bdrv_states, BlkTransactionStates) snap_bdrv_states;
950 951 952 953 954 955 956
    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) {
957
        TransactionAction *dev_info = NULL;
958
        const BdrvActionOps *ops;
959

960 961 962
        dev_info = dev_entry->value;
        dev_entry = dev_entry->next;

963 964 965 966 967 968
        assert(dev_info->kind < ARRAY_SIZE(actions));

        ops = &actions[dev_info->kind];
        states = g_malloc0(ops->instance_size);
        states->ops = ops;
        states->action = dev_info;
969 970
        QSIMPLEQ_INSERT_TAIL(&snap_bdrv_states, states, entry);

971 972 973 974
        states->ops->prepare(states, &local_err);
        if (error_is_set(&local_err)) {
            error_propagate(errp, local_err);
            goto delete_and_fail;
975 976 977 978
        }
    }

    QSIMPLEQ_FOREACH(states, &snap_bdrv_states, entry) {
979
        states->ops->commit(states);
980 981 982 983 984 985 986 987 988 989 990
    }

    /* 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) {
991 992 993
        if (states->ops->abort) {
            states->ops->abort(states);
        }
994 995
    }
exit:
996
    QSIMPLEQ_FOREACH_SAFE(states, &snap_bdrv_states, entry, next) {
997 998 999
        if (states->ops->clean) {
            states->ops->clean(states);
        }
1000 1001 1002 1003 1004
        g_free(states);
    }
}


1005
static void eject_device(BlockDriverState *bs, int force, Error **errp)
1006
{
1007 1008 1009 1010
    if (bdrv_in_use(bs)) {
        error_set(errp, QERR_DEVICE_IN_USE, bdrv_get_device_name(bs));
        return;
    }
1011
    if (!bdrv_dev_has_removable_media(bs)) {
1012 1013
        error_set(errp, QERR_DEVICE_NOT_REMOVABLE, bdrv_get_device_name(bs));
        return;
1014
    }
1015

P
Paolo Bonzini 已提交
1016 1017 1018
    if (bdrv_dev_is_medium_locked(bs) && !bdrv_dev_is_tray_open(bs)) {
        bdrv_dev_eject_request(bs, force);
        if (!force) {
1019 1020
            error_set(errp, QERR_DEVICE_LOCKED, bdrv_get_device_name(bs));
            return;
P
Paolo Bonzini 已提交
1021
        }
1022
    }
1023

1024
    bdrv_close(bs);
1025 1026
}

L
Luiz Capitulino 已提交
1027
void qmp_eject(const char *device, bool has_force, bool force, Error **errp)
1028 1029 1030
{
    BlockDriverState *bs;

L
Luiz Capitulino 已提交
1031
    bs = bdrv_find(device);
1032
    if (!bs) {
L
Luiz Capitulino 已提交
1033 1034
        error_set(errp, QERR_DEVICE_NOT_FOUND, device);
        return;
1035 1036
    }

L
Luiz Capitulino 已提交
1037
    eject_device(bs, force, errp);
1038 1039
}

L
Luiz Capitulino 已提交
1040
void qmp_block_passwd(const char *device, const char *password, Error **errp)
1041 1042 1043 1044
{
    BlockDriverState *bs;
    int err;

L
Luiz Capitulino 已提交
1045
    bs = bdrv_find(device);
1046
    if (!bs) {
L
Luiz Capitulino 已提交
1047 1048
        error_set(errp, QERR_DEVICE_NOT_FOUND, device);
        return;
1049 1050
    }

L
Luiz Capitulino 已提交
1051
    err = bdrv_set_key(bs, password);
1052
    if (err == -EINVAL) {
L
Luiz Capitulino 已提交
1053 1054
        error_set(errp, QERR_DEVICE_NOT_ENCRYPTED, bdrv_get_device_name(bs));
        return;
1055
    } else if (err < 0) {
L
Luiz Capitulino 已提交
1056 1057
        error_set(errp, QERR_INVALID_PASSWORD);
        return;
1058 1059 1060
    }
}

L
Luiz Capitulino 已提交
1061 1062 1063 1064
static void qmp_bdrv_open_encrypted(BlockDriverState *bs, const char *filename,
                                    int bdrv_flags, BlockDriver *drv,
                                    const char *password, Error **errp)
{
1065
    if (bdrv_open(bs, filename, NULL, bdrv_flags, drv) < 0) {
L
Luiz Capitulino 已提交
1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085
        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)
1086 1087 1088 1089
{
    BlockDriverState *bs;
    BlockDriver *drv = NULL;
    int bdrv_flags;
1090
    Error *err = NULL;
1091 1092 1093

    bs = bdrv_find(device);
    if (!bs) {
L
Luiz Capitulino 已提交
1094 1095
        error_set(errp, QERR_DEVICE_NOT_FOUND, device);
        return;
1096
    }
L
Luiz Capitulino 已提交
1097 1098

    if (format) {
1099
        drv = bdrv_find_whitelisted_format(format, bs->read_only);
1100
        if (!drv) {
L
Luiz Capitulino 已提交
1101 1102
            error_set(errp, QERR_INVALID_BLOCK_FORMAT, format);
            return;
1103 1104
        }
    }
L
Luiz Capitulino 已提交
1105

1106 1107
    eject_device(bs, 0, &err);
    if (error_is_set(&err)) {
L
Luiz Capitulino 已提交
1108 1109
        error_propagate(errp, err);
        return;
1110
    }
L
Luiz Capitulino 已提交
1111

1112
    bdrv_flags = bdrv_is_read_only(bs) ? 0 : BDRV_O_RDWR;
1113
    bdrv_flags |= bdrv_is_snapshot(bs) ? BDRV_O_SNAPSHOT : 0;
L
Luiz Capitulino 已提交
1114 1115

    qmp_bdrv_open_encrypted(bs, filename, bdrv_flags, drv, NULL, errp);
1116
}
1117

Z
Zhi Yong Wu 已提交
1118
/* throttling disk I/O limits */
1119 1120 1121
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)
Z
Zhi Yong Wu 已提交
1122 1123 1124 1125
{
    BlockIOLimit io_limits;
    BlockDriverState *bs;

1126
    bs = bdrv_find(device);
Z
Zhi Yong Wu 已提交
1127
    if (!bs) {
1128 1129
        error_set(errp, QERR_DEVICE_NOT_FOUND, device);
        return;
Z
Zhi Yong Wu 已提交
1130 1131
    }

1132 1133 1134 1135 1136 1137
    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;
Z
Zhi Yong Wu 已提交
1138

1139
    if (!do_check_io_limits(&io_limits, errp)) {
1140
        return;
Z
Zhi Yong Wu 已提交
1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155
    }

    bs->io_limits = io_limits;

    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));
        }
    }
}

1156 1157 1158 1159 1160 1161 1162 1163 1164 1165
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;
    }
M
Marcelo Tosatti 已提交
1166 1167 1168 1169
    if (bdrv_in_use(bs)) {
        qerror_report(QERR_DEVICE_IN_USE, id);
        return -1;
    }
1170 1171

    /* quiesce block driver; prevent further io */
1172
    bdrv_drain_all();
1173 1174 1175
    bdrv_flush(bs);
    bdrv_close(bs);

1176
    /* if we have a device attached to this BlockDriverState
1177 1178 1179 1180
     * 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.
     */
1181
    if (bdrv_get_attached_dev(bs)) {
1182
        bdrv_make_anon(bs);
1183 1184 1185 1186

        /* Further I/O must not pause the guest */
        bdrv_set_on_error(bs, BLOCKDEV_ON_ERROR_REPORT,
                          BLOCKDEV_ON_ERROR_REPORT);
1187 1188
    } else {
        drive_uninit(drive_get_by_blockdev(bs));
1189 1190 1191 1192
    }

    return 0;
}
1193

L
Luiz Capitulino 已提交
1194
void qmp_block_resize(const char *device, int64_t size, Error **errp)
1195 1196
{
    BlockDriverState *bs;
1197
    int ret;
1198 1199 1200

    bs = bdrv_find(device);
    if (!bs) {
L
Luiz Capitulino 已提交
1201 1202
        error_set(errp, QERR_DEVICE_NOT_FOUND, device);
        return;
1203 1204 1205
    }

    if (size < 0) {
1206
        error_set(errp, QERR_INVALID_PARAMETER_VALUE, "size", "a >0 size");
L
Luiz Capitulino 已提交
1207
        return;
1208 1209
    }

1210 1211 1212
    /* complete all in-flight operations before resizing the device */
    bdrv_drain_all();

1213 1214
    ret = bdrv_truncate(bs, size);
    switch (ret) {
1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229
    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:
1230
        error_setg_errno(errp, -ret, "Could not resize");
1231
        break;
1232 1233
    }
}
S
Stefan Hajnoczi 已提交
1234

1235
static void block_job_cb(void *opaque, int ret)
S
Stefan Hajnoczi 已提交
1236 1237 1238 1239
{
    BlockDriverState *bs = opaque;
    QObject *obj;

1240
    trace_block_job_cb(bs, bs->job, ret);
S
Stefan Hajnoczi 已提交
1241 1242 1243 1244 1245 1246 1247 1248

    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)));
    }

1249 1250 1251 1252 1253
    if (block_job_is_cancelled(bs->job)) {
        monitor_protocol_event(QEVENT_BLOCK_JOB_CANCELLED, obj);
    } else {
        monitor_protocol_event(QEVENT_BLOCK_JOB_COMPLETED, obj);
    }
S
Stefan Hajnoczi 已提交
1254
    qobject_decref(obj);
1255 1256

    drive_put_ref_bh_schedule(drive_get_by_blockdev(bs));
S
Stefan Hajnoczi 已提交
1257 1258 1259
}

void qmp_block_stream(const char *device, bool has_base,
P
Paolo Bonzini 已提交
1260 1261 1262
                      const char *base, bool has_speed, int64_t speed,
                      bool has_on_error, BlockdevOnError on_error,
                      Error **errp)
S
Stefan Hajnoczi 已提交
1263 1264
{
    BlockDriverState *bs;
1265
    BlockDriverState *base_bs = NULL;
1266
    Error *local_err = NULL;
S
Stefan Hajnoczi 已提交
1267

P
Paolo Bonzini 已提交
1268 1269 1270 1271
    if (!has_on_error) {
        on_error = BLOCKDEV_ON_ERROR_REPORT;
    }

S
Stefan Hajnoczi 已提交
1272 1273 1274 1275 1276 1277 1278
    bs = bdrv_find(device);
    if (!bs) {
        error_set(errp, QERR_DEVICE_NOT_FOUND, device);
        return;
    }

    if (base) {
1279 1280 1281 1282 1283
        base_bs = bdrv_find_backing_image(bs, base);
        if (base_bs == NULL) {
            error_set(errp, QERR_BASE_NOT_FOUND, base);
            return;
        }
S
Stefan Hajnoczi 已提交
1284 1285
    }

1286
    stream_start(bs, base_bs, base, has_speed ? speed : 0,
P
Paolo Bonzini 已提交
1287
                 on_error, block_job_cb, bs, &local_err);
1288 1289 1290
    if (error_is_set(&local_err)) {
        error_propagate(errp, local_err);
        return;
S
Stefan Hajnoczi 已提交
1291 1292
    }

1293 1294 1295 1296 1297
    /* Grab a reference so hotplug does not delete the BlockDriverState from
     * underneath us.
     */
    drive_get_ref(drive_get_by_blockdev(bs));

S
Stefan Hajnoczi 已提交
1298 1299
    trace_qmp_block_stream(bs, bs->job);
}
1300

1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311
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
     */
1312
    BlockdevOnError on_error = BLOCKDEV_ON_ERROR_REPORT;
1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336

    /* 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;
    }

1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347
    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;
    }

1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359
    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));
}

1360 1361
#define DEFAULT_MIRROR_BUF_SIZE   (10 << 20)

P
Paolo Bonzini 已提交
1362 1363 1364 1365
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,
1366
                      bool has_speed, int64_t speed,
1367
                      bool has_granularity, uint32_t granularity,
1368
                      bool has_buf_size, int64_t buf_size,
1369 1370 1371
                      bool has_on_source_error, BlockdevOnError on_source_error,
                      bool has_on_target_error, BlockdevOnError on_target_error,
                      Error **errp)
P
Paolo Bonzini 已提交
1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384
{
    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;
    }
1385 1386 1387 1388 1389 1390
    if (!has_on_source_error) {
        on_source_error = BLOCKDEV_ON_ERROR_REPORT;
    }
    if (!has_on_target_error) {
        on_target_error = BLOCKDEV_ON_ERROR_REPORT;
    }
P
Paolo Bonzini 已提交
1391 1392 1393
    if (!has_mode) {
        mode = NEW_IMAGE_MODE_ABSOLUTE_PATHS;
    }
1394 1395 1396
    if (!has_granularity) {
        granularity = 0;
    }
1397 1398 1399 1400
    if (!has_buf_size) {
        buf_size = DEFAULT_MIRROR_BUF_SIZE;
    }

1401 1402 1403 1404 1405 1406 1407 1408
    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;
    }
P
Paolo Bonzini 已提交
1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448

    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;
    }

    proto_drv = bdrv_find_protocol(target);
    if (!proto_drv) {
        error_set(errp, QERR_INVALID_BLOCK_FORMAT, format);
        return;
    }

1449 1450
    bdrv_get_geometry(bs, &size);
    size *= 512;
P
Paolo Bonzini 已提交
1451 1452 1453
    if (sync == MIRROR_SYNC_MODE_FULL && mode != NEW_IMAGE_MODE_EXISTING) {
        /* create new image w/o backing file */
        assert(format && drv);
1454
        bdrv_img_create(target, format,
1455
                        NULL, NULL, NULL, size, flags, &local_err, false);
P
Paolo Bonzini 已提交
1456 1457 1458 1459 1460 1461 1462
    } else {
        switch (mode) {
        case NEW_IMAGE_MODE_EXISTING:
            ret = 0;
            break;
        case NEW_IMAGE_MODE_ABSOLUTE_PATHS:
            /* create new image with backing file */
1463 1464 1465
            bdrv_img_create(target, format,
                            source->filename,
                            source->drv->format_name,
1466
                            NULL, size, flags, &local_err, false);
P
Paolo Bonzini 已提交
1467 1468 1469 1470 1471 1472
            break;
        default:
            abort();
        }
    }

1473 1474
    if (error_is_set(&local_err)) {
        error_propagate(errp, local_err);
P
Paolo Bonzini 已提交
1475 1476 1477
        return;
    }

1478 1479 1480
    /* Mirroring takes care of copy-on-write using the source's backing
     * file.
     */
P
Paolo Bonzini 已提交
1481
    target_bs = bdrv_new("");
1482
    ret = bdrv_open(target_bs, target, NULL, flags | BDRV_O_NO_BACKING, drv);
P
Paolo Bonzini 已提交
1483 1484 1485 1486 1487 1488 1489

    if (ret < 0) {
        bdrv_delete(target_bs);
        error_set(errp, QERR_OPEN_FILE_FAILED, target);
        return;
    }

1490
    mirror_start(bs, target_bs, speed, granularity, buf_size, sync,
1491
                 on_source_error, on_target_error,
1492
                 block_job_cb, bs, &local_err);
P
Paolo Bonzini 已提交
1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504
    if (local_err != NULL) {
        bdrv_delete(target_bs);
        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));
}

1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515
static BlockJob *find_block_job(const char *device)
{
    BlockDriverState *bs;

    bs = bdrv_find(device);
    if (!bs || !bs->job) {
        return NULL;
    }
    return bs->job;
}

1516
void qmp_block_job_set_speed(const char *device, int64_t speed, Error **errp)
1517 1518 1519 1520
{
    BlockJob *job = find_block_job(device);

    if (!job) {
1521
        error_set(errp, QERR_BLOCK_JOB_NOT_ACTIVE, device);
1522 1523 1524
        return;
    }

1525
    block_job_set_speed(job, speed, errp);
1526
}
1527

1528 1529
void qmp_block_job_cancel(const char *device,
                          bool has_force, bool force, Error **errp)
1530 1531 1532
{
    BlockJob *job = find_block_job(device);

1533 1534 1535 1536
    if (!has_force) {
        force = false;
    }

1537
    if (!job) {
1538
        error_set(errp, QERR_BLOCK_JOB_NOT_ACTIVE, device);
1539 1540
        return;
    }
1541
    if (job->paused && !force) {
1542 1543 1544
        error_set(errp, QERR_BLOCK_JOB_PAUSED, device);
        return;
    }
1545 1546 1547 1548

    trace_qmp_block_job_cancel(job);
    block_job_cancel(job);
}
S
Stefan Hajnoczi 已提交
1549

1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575
void qmp_block_job_pause(const char *device, Error **errp)
{
    BlockJob *job = find_block_job(device);

    if (!job) {
        error_set(errp, QERR_BLOCK_JOB_NOT_ACTIVE, device);
        return;
    }

    trace_qmp_block_job_pause(job);
    block_job_pause(job);
}

void qmp_block_job_resume(const char *device, Error **errp)
{
    BlockJob *job = find_block_job(device);

    if (!job) {
        error_set(errp, QERR_BLOCK_JOB_NOT_ACTIVE, device);
        return;
    }

    trace_qmp_block_job_resume(job);
    block_job_resume(job);
}

P
Paolo Bonzini 已提交
1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588
void qmp_block_job_complete(const char *device, Error **errp)
{
    BlockJob *job = find_block_job(device);

    if (!job) {
        error_set(errp, QERR_BLOCK_JOB_NOT_ACTIVE, device);
        return;
    }

    trace_qmp_block_job_complete(job);
    block_job_complete(job, errp);
}

S
Stefan Hajnoczi 已提交
1589 1590 1591 1592 1593 1594
static void do_qmp_query_block_jobs_one(void *opaque, BlockDriverState *bs)
{
    BlockJobInfoList **prev = opaque;
    BlockJob *job = bs->job;

    if (job) {
P
Paolo Bonzini 已提交
1595 1596
        BlockJobInfoList *elem = g_new0(BlockJobInfoList, 1);
        elem->value = block_job_query(bs->job);
S
Stefan Hajnoczi 已提交
1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609
        (*prev)->next = elem;
        *prev = elem;
    }
}

BlockJobInfoList *qmp_query_block_jobs(Error **errp)
{
    /* Dummy is a fake list element for holding the head pointer */
    BlockJobInfoList dummy = {};
    BlockJobInfoList *prev = &dummy;
    bdrv_iterate(do_qmp_query_block_jobs_one, &prev);
    return dummy.next;
}
1610

1611
QemuOptsList qemu_common_drive_opts = {
1612
    .name = "drive",
1613
    .head = QTAILQ_HEAD_INITIALIZER(qemu_common_drive_opts.head),
1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658
    .desc = {
        {
            .name = "bus",
            .type = QEMU_OPT_NUMBER,
            .help = "bus number",
        },{
            .name = "unit",
            .type = QEMU_OPT_NUMBER,
            .help = "unit number (i.e. lun for scsi)",
        },{
            .name = "if",
            .type = QEMU_OPT_STRING,
            .help = "interface (ide, scsi, sd, mtd, floppy, pflash, virtio)",
        },{
            .name = "index",
            .type = QEMU_OPT_NUMBER,
            .help = "index number",
        },{
            .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)",
        },{
            .name = "media",
            .type = QEMU_OPT_STRING,
            .help = "media type (disk, cdrom)",
        },{
            .name = "snapshot",
            .type = QEMU_OPT_BOOL,
            .help = "enable/disable snapshot mode",
        },{
            .name = "file",
            .type = QEMU_OPT_STRING,
            .help = "disk image",
1659 1660 1661 1662
        },{
            .name = "discard",
            .type = QEMU_OPT_STRING,
            .help = "discard operation (ignore/off, unmap/on)",
1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731
        },{
            .name = "cache",
            .type = QEMU_OPT_STRING,
            .help = "host cache usage (none, writeback, writethrough, "
                    "directsync, unsafe)",
        },{
            .name = "aio",
            .type = QEMU_OPT_STRING,
            .help = "host AIO implementation (threads, native)",
        },{
            .name = "format",
            .type = QEMU_OPT_STRING,
            .help = "disk format (raw, qcow2, ...)",
        },{
            .name = "serial",
            .type = QEMU_OPT_STRING,
            .help = "disk serial number",
        },{
            .name = "rerror",
            .type = QEMU_OPT_STRING,
            .help = "read error action",
        },{
            .name = "werror",
            .type = QEMU_OPT_STRING,
            .help = "write error action",
        },{
            .name = "addr",
            .type = QEMU_OPT_STRING,
            .help = "pci address (virtio only)",
        },{
            .name = "readonly",
            .type = QEMU_OPT_BOOL,
            .help = "open drive file as read-only",
        },{
            .name = "iops",
            .type = QEMU_OPT_NUMBER,
            .help = "limit total I/O operations per second",
        },{
            .name = "iops_rd",
            .type = QEMU_OPT_NUMBER,
            .help = "limit read operations per second",
        },{
            .name = "iops_wr",
            .type = QEMU_OPT_NUMBER,
            .help = "limit write operations per second",
        },{
            .name = "bps",
            .type = QEMU_OPT_NUMBER,
            .help = "limit total bytes per second",
        },{
            .name = "bps_rd",
            .type = QEMU_OPT_NUMBER,
            .help = "limit read bytes per second",
        },{
            .name = "bps_wr",
            .type = QEMU_OPT_NUMBER,
            .help = "limit write bytes per second",
        },{
            .name = "copy-on-read",
            .type = QEMU_OPT_BOOL,
            .help = "copy read data from backing file into image file",
        },{
            .name = "boot",
            .type = QEMU_OPT_BOOL,
            .help = "(deprecated, ignored)",
        },
        { /* end of list */ }
    },
};
1732 1733 1734 1735 1736

QemuOptsList qemu_drive_opts = {
    .name = "drive",
    .head = QTAILQ_HEAD_INITIALIZER(qemu_drive_opts.head),
    .desc = {
1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850
        {
            .name = "bus",
            .type = QEMU_OPT_NUMBER,
            .help = "bus number",
        },{
            .name = "unit",
            .type = QEMU_OPT_NUMBER,
            .help = "unit number (i.e. lun for scsi)",
        },{
            .name = "if",
            .type = QEMU_OPT_STRING,
            .help = "interface (ide, scsi, sd, mtd, floppy, pflash, virtio)",
        },{
            .name = "index",
            .type = QEMU_OPT_NUMBER,
            .help = "index number",
        },{
            .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)",
        },{
            .name = "media",
            .type = QEMU_OPT_STRING,
            .help = "media type (disk, cdrom)",
        },{
            .name = "snapshot",
            .type = QEMU_OPT_BOOL,
            .help = "enable/disable snapshot mode",
        },{
            .name = "file",
            .type = QEMU_OPT_STRING,
            .help = "disk image",
        },{
            .name = "discard",
            .type = QEMU_OPT_STRING,
            .help = "discard operation (ignore/off, unmap/on)",
        },{
            .name = "cache",
            .type = QEMU_OPT_STRING,
            .help = "host cache usage (none, writeback, writethrough, "
                    "directsync, unsafe)",
        },{
            .name = "aio",
            .type = QEMU_OPT_STRING,
            .help = "host AIO implementation (threads, native)",
        },{
            .name = "format",
            .type = QEMU_OPT_STRING,
            .help = "disk format (raw, qcow2, ...)",
        },{
            .name = "serial",
            .type = QEMU_OPT_STRING,
            .help = "disk serial number",
        },{
            .name = "rerror",
            .type = QEMU_OPT_STRING,
            .help = "read error action",
        },{
            .name = "werror",
            .type = QEMU_OPT_STRING,
            .help = "write error action",
        },{
            .name = "addr",
            .type = QEMU_OPT_STRING,
            .help = "pci address (virtio only)",
        },{
            .name = "readonly",
            .type = QEMU_OPT_BOOL,
            .help = "open drive file as read-only",
        },{
            .name = "iops",
            .type = QEMU_OPT_NUMBER,
            .help = "limit total I/O operations per second",
        },{
            .name = "iops_rd",
            .type = QEMU_OPT_NUMBER,
            .help = "limit read operations per second",
        },{
            .name = "iops_wr",
            .type = QEMU_OPT_NUMBER,
            .help = "limit write operations per second",
        },{
            .name = "bps",
            .type = QEMU_OPT_NUMBER,
            .help = "limit total bytes per second",
        },{
            .name = "bps_rd",
            .type = QEMU_OPT_NUMBER,
            .help = "limit read bytes per second",
        },{
            .name = "bps_wr",
            .type = QEMU_OPT_NUMBER,
            .help = "limit write bytes per second",
        },{
            .name = "copy-on-read",
            .type = QEMU_OPT_BOOL,
            .help = "copy read data from backing file into image file",
        },{
            .name = "boot",
            .type = QEMU_OPT_BOOL,
            .help = "(deprecated, ignored)",
        },
1851 1852 1853
        { /* end of list */ }
    },
};