vhdx.c 64.7 KB
Newer Older
1 2 3 4 5 6 7 8
/*
 * Block driver for Hyper-V VHDX Images
 *
 * Copyright (c) 2013 Red Hat, Inc.,
 *
 * Authors:
 *  Jeff Cody <jcody@redhat.com>
 *
9
 *  This is based on the "VHDX Format Specification v1.00", published 8/25/2012
10
 *  by Microsoft:
11
 *      https://www.microsoft.com/en-us/download/details.aspx?id=34750
12 13 14 15 16 17
 *
 * This work is licensed under the terms of the GNU LGPL, version 2 or later.
 * See the COPYING.LIB file in the top-level directory.
 *
 */

P
Peter Maydell 已提交
18
#include "qemu/osdep.h"
19 20
#include "qemu-common.h"
#include "block/block_int.h"
21
#include "sysemu/block-backend.h"
22 23 24
#include "qemu/module.h"
#include "qemu/crc32c.h"
#include "block/vhdx.h"
J
Jeff Cody 已提交
25
#include "migration/migration.h"
26

27
#include <uuid/uuid.h>
28 29 30 31 32 33 34 35 36 37 38 39 40
#include <glib.h>

/* Options for VHDX creation */

#define VHDX_BLOCK_OPT_LOG_SIZE   "log_size"
#define VHDX_BLOCK_OPT_BLOCK_SIZE "block_size"
#define VHDX_BLOCK_OPT_ZERO "block_state_zero"

typedef enum VHDXImageType {
    VHDX_TYPE_DYNAMIC = 0,
    VHDX_TYPE_FIXED,
    VHDX_TYPE_DIFFERENCING,   /* Currently unsupported */
} VHDXImageType;
41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103

/* Several metadata and region table data entries are identified by
 * guids in  a MS-specific GUID format. */


/* ------- Known Region Table GUIDs ---------------------- */
static const MSGUID bat_guid =      { .data1 = 0x2dc27766,
                                      .data2 = 0xf623,
                                      .data3 = 0x4200,
                                      .data4 = { 0x9d, 0x64, 0x11, 0x5e,
                                                 0x9b, 0xfd, 0x4a, 0x08} };

static const MSGUID metadata_guid = { .data1 = 0x8b7ca206,
                                      .data2 = 0x4790,
                                      .data3 = 0x4b9a,
                                      .data4 = { 0xb8, 0xfe, 0x57, 0x5f,
                                                 0x05, 0x0f, 0x88, 0x6e} };



/* ------- Known Metadata Entry GUIDs ---------------------- */
static const MSGUID file_param_guid =   { .data1 = 0xcaa16737,
                                          .data2 = 0xfa36,
                                          .data3 = 0x4d43,
                                          .data4 = { 0xb3, 0xb6, 0x33, 0xf0,
                                                     0xaa, 0x44, 0xe7, 0x6b} };

static const MSGUID virtual_size_guid = { .data1 = 0x2FA54224,
                                          .data2 = 0xcd1b,
                                          .data3 = 0x4876,
                                          .data4 = { 0xb2, 0x11, 0x5d, 0xbe,
                                                     0xd8, 0x3b, 0xf4, 0xb8} };

static const MSGUID page83_guid =       { .data1 = 0xbeca12ab,
                                          .data2 = 0xb2e6,
                                          .data3 = 0x4523,
                                          .data4 = { 0x93, 0xef, 0xc3, 0x09,
                                                     0xe0, 0x00, 0xc7, 0x46} };


static const MSGUID phys_sector_guid =  { .data1 = 0xcda348c7,
                                          .data2 = 0x445d,
                                          .data3 = 0x4471,
                                          .data4 = { 0x9c, 0xc9, 0xe9, 0x88,
                                                     0x52, 0x51, 0xc5, 0x56} };

static const MSGUID parent_locator_guid = { .data1 = 0xa8d35f2d,
                                            .data2 = 0xb30b,
                                            .data3 = 0x454d,
                                            .data4 = { 0xab, 0xf7, 0xd3,
                                                       0xd8, 0x48, 0x34,
                                                       0xab, 0x0c} };

static const MSGUID logical_sector_guid = { .data1 = 0x8141bf1d,
                                            .data2 = 0xa96f,
                                            .data3 = 0x4709,
                                            .data4 = { 0xba, 0x47, 0xf2,
                                                       0x33, 0xa8, 0xfa,
                                                       0xab, 0x5f} };

/* Each parent type must have a valid GUID; this is for parent images
 * of type 'VHDX'.  If we were to allow e.g. a QCOW2 parent, we would
 * need to make up our own QCOW2 GUID type */
104 105
static const MSGUID parent_vhdx_guid __attribute__((unused))
                                     = { .data1 = 0xb04aefb7,
106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124
                                         .data2 = 0xd19e,
                                         .data3 = 0x4a81,
                                         .data4 = { 0xb7, 0x89, 0x25, 0xb8,
                                                    0xe9, 0x44, 0x59, 0x13} };


#define META_FILE_PARAMETER_PRESENT      0x01
#define META_VIRTUAL_DISK_SIZE_PRESENT   0x02
#define META_PAGE_83_PRESENT             0x04
#define META_LOGICAL_SECTOR_SIZE_PRESENT 0x08
#define META_PHYS_SECTOR_SIZE_PRESENT    0x10
#define META_PARENT_LOCATOR_PRESENT      0x20

#define META_ALL_PRESENT    \
    (META_FILE_PARAMETER_PRESENT | META_VIRTUAL_DISK_SIZE_PRESENT | \
     META_PAGE_83_PRESENT | META_LOGICAL_SECTOR_SIZE_PRESENT | \
     META_PHYS_SECTOR_SIZE_PRESENT)


125 126 127 128 129 130 131 132 133
typedef struct VHDXSectorInfo {
    uint32_t bat_idx;       /* BAT entry index */
    uint32_t sectors_avail; /* sectors available in payload block */
    uint32_t bytes_left;    /* bytes left in the block after data to r/w */
    uint32_t bytes_avail;   /* bytes available in payload block */
    uint64_t file_offset;   /* absolute offset in bytes, in file */
    uint64_t block_offset;  /* block offset, in bytes */
} VHDXSectorInfo;

134 135 136 137 138 139 140
/* Calculates new checksum.
 *
 * Zero is substituted during crc calculation for the original crc field
 * crc_offset: byte offset in buf of the buffer crc
 * buf: buffer pointer
 * size: size of buffer (must be > crc_offset+4)
 *
J
Jeff Cody 已提交
141 142
 * Note: The buffer should have all multi-byte data in little-endian format,
 *       and the resulting checksum is in little endian format.
143 144 145 146 147 148 149 150 151 152
 */
uint32_t vhdx_update_checksum(uint8_t *buf, size_t size, int crc_offset)
{
    uint32_t crc;

    assert(buf != NULL);
    assert(size > (crc_offset + sizeof(crc)));

    memset(buf + crc_offset, 0, sizeof(crc));
    crc =  crc32c(0xffffffff, buf, size);
J
Jeff Cody 已提交
153
    cpu_to_le32s(&crc);
154 155 156 157 158
    memcpy(buf + crc_offset, &crc, sizeof(crc));

    return crc;
}

159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208
uint32_t vhdx_checksum_calc(uint32_t crc, uint8_t *buf, size_t size,
                            int crc_offset)
{
    uint32_t crc_new;
    uint32_t crc_orig;
    assert(buf != NULL);

    if (crc_offset > 0) {
        memcpy(&crc_orig, buf + crc_offset, sizeof(crc_orig));
        memset(buf + crc_offset, 0, sizeof(crc_orig));
    }

    crc_new = crc32c(crc, buf, size);
    if (crc_offset > 0) {
        memcpy(buf + crc_offset, &crc_orig, sizeof(crc_orig));
    }

    return crc_new;
}

/* Validates the checksum of the buffer, with an in-place CRC.
 *
 * Zero is substituted during crc calculation for the original crc field,
 * and the crc field is restored afterwards.  But the buffer will be modifed
 * during the calculation, so this may not be not suitable for multi-threaded
 * use.
 *
 * crc_offset: byte offset in buf of the buffer crc
 * buf: buffer pointer
 * size: size of buffer (must be > crc_offset+4)
 *
 * returns true if checksum is valid, false otherwise
 */
bool vhdx_checksum_is_valid(uint8_t *buf, size_t size, int crc_offset)
{
    uint32_t crc_orig;
    uint32_t crc;

    assert(buf != NULL);
    assert(size > (crc_offset + 4));

    memcpy(&crc_orig, buf + crc_offset, sizeof(crc_orig));
    crc_orig = le32_to_cpu(crc_orig);

    crc = vhdx_checksum_calc(0xffffffff, buf, size, crc_offset);

    return crc == crc_orig;
}


209 210 211 212 213 214 215 216 217 218 219 220 221
/*
 * This generates a UUID that is compliant with the MS GUIDs used
 * in the VHDX spec (and elsewhere).
 */
void vhdx_guid_generate(MSGUID *guid)
{
    uuid_t uuid;
    assert(guid != NULL);

    uuid_generate(uuid);
    memcpy(guid, uuid, sizeof(MSGUID));
}

222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265
/* Check for region overlaps inside the VHDX image */
static int vhdx_region_check(BDRVVHDXState *s, uint64_t start, uint64_t length)
{
    int ret = 0;
    uint64_t end;
    VHDXRegionEntry *r;

    end = start + length;
    QLIST_FOREACH(r, &s->regions, entries) {
        if (!((start >= r->end) || (end <= r->start))) {
            ret = -EINVAL;
            goto exit;
        }
    }

exit:
    return ret;
}

/* Register a region for future checks */
static void vhdx_region_register(BDRVVHDXState *s,
                                 uint64_t start, uint64_t length)
{
    VHDXRegionEntry *r;

    r = g_malloc0(sizeof(*r));

    r->start = start;
    r->end = start + length;

    QLIST_INSERT_HEAD(&s->regions, r, entries);
}

/* Free all registered regions */
static void vhdx_region_unregister_all(BDRVVHDXState *s)
{
    VHDXRegionEntry *r, *r_next;

    QLIST_FOREACH_SAFE(r, &s->regions, entries, r_next) {
        QLIST_REMOVE(r, entries);
        g_free(r);
    }
}

266 267
static void vhdx_set_shift_bits(BDRVVHDXState *s)
{
M
Max Reitz 已提交
268 269 270 271
    s->logical_sector_size_bits = ctz32(s->logical_sector_size);
    s->sectors_per_block_bits =   ctz32(s->sectors_per_block);
    s->chunk_ratio_bits =         ctz64(s->chunk_ratio);
    s->block_size_bits =          ctz32(s->block_size);
272 273
}

274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292
/*
 * Per the MS VHDX Specification, for every VHDX file:
 *      - The header section is fixed size - 1 MB
 *      - The header section is always the first "object"
 *      - The first 64KB of the header is the File Identifier
 *      - The first uint64 (8 bytes) is the VHDX Signature ("vhdxfile")
 *      - The following 512 bytes constitute a UTF-16 string identifiying the
 *        software that created the file, and is optional and diagnostic only.
 *
 *  Therefore, we probe by looking for the vhdxfile signature "vhdxfile"
 */
static int vhdx_probe(const uint8_t *buf, int buf_size, const char *filename)
{
    if (buf_size >= 8 && !memcmp(buf, "vhdxfile", 8)) {
        return 100;
    }
    return 0;
}

293 294 295 296 297 298 299 300 301 302 303 304
/*
 * Writes the header to the specified offset.
 *
 * This will optionally read in buffer data from disk (otherwise zero-fill),
 * and then update the header checksum.  Header is converted to proper
 * endianness before being written to the specified file offset
 */
static int vhdx_write_header(BlockDriverState *bs_file, VHDXHeader *hdr,
                             uint64_t offset, bool read)
{
    uint8_t *buffer = NULL;
    int ret;
J
Jeff Cody 已提交
305
    VHDXHeader *header_le;
306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325

    assert(bs_file != NULL);
    assert(hdr != NULL);

    /* the header checksum is not over just the packed size of VHDXHeader,
     * but rather over the entire 'reserved' range for the header, which is
     * 4KB (VHDX_HEADER_SIZE). */

    buffer = qemu_blockalign(bs_file, VHDX_HEADER_SIZE);
    if (read) {
        /* if true, we can't assume the extra reserved bytes are 0 */
        ret = bdrv_pread(bs_file, offset, buffer, VHDX_HEADER_SIZE);
        if (ret < 0) {
            goto exit;
        }
    } else {
        memset(buffer, 0, VHDX_HEADER_SIZE);
    }

    /* overwrite the actual VHDXHeader portion */
J
Jeff Cody 已提交
326 327 328 329 330 331
    header_le = (VHDXHeader *)buffer;
    memcpy(header_le, hdr, sizeof(VHDXHeader));
    vhdx_header_le_export(hdr, header_le);
    vhdx_update_checksum(buffer, VHDX_HEADER_SIZE,
                         offsetof(VHDXHeader, checksum));
    ret = bdrv_pwrite_sync(bs_file, offset, header_le, sizeof(VHDXHeader));
332 333 334 335 336 337

exit:
    qemu_vfree(buffer);
    return ret;
}

338 339 340 341 342 343 344
/* Update the VHDX headers
 *
 * This follows the VHDX spec procedures for header updates.
 *
 *  - non-current header is updated with largest sequence number
 */
static int vhdx_update_header(BlockDriverState *bs, BDRVVHDXState *s,
345
                              bool generate_data_write_guid, MSGUID *log_guid)
346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374
{
    int ret = 0;
    int hdr_idx = 0;
    uint64_t header_offset = VHDX_HEADER1_OFFSET;

    VHDXHeader *active_header;
    VHDXHeader *inactive_header;

    /* operate on the non-current header */
    if (s->curr_header == 0) {
        hdr_idx = 1;
        header_offset = VHDX_HEADER2_OFFSET;
    }

    active_header   = s->headers[s->curr_header];
    inactive_header = s->headers[hdr_idx];

    inactive_header->sequence_number = active_header->sequence_number + 1;

    /* a new file guid must be generated before any file write, including
     * headers */
    inactive_header->file_write_guid = s->session_guid;

    /* a new data guid only needs to be generated before any guest-visible
     * writes (i.e. something observable via virtual disk read) */
    if (generate_data_write_guid) {
        vhdx_guid_generate(&inactive_header->data_write_guid);
    }

375 376 377 378 379
    /* update the log guid if present */
    if (log_guid) {
        inactive_header->log_guid = *log_guid;
    }

K
Kevin Wolf 已提交
380
    ret = vhdx_write_header(bs->file->bs, inactive_header, header_offset, true);
381 382 383 384 385 386 387 388 389 390 391 392 393
    if (ret < 0) {
        goto exit;
    }
    s->curr_header = hdr_idx;

exit:
    return ret;
}

/*
 * The VHDX spec calls for header updates to be performed twice, so that both
 * the current and non-current header have valid info
 */
394 395
int vhdx_update_headers(BlockDriverState *bs, BDRVVHDXState *s,
                        bool generate_data_write_guid, MSGUID *log_guid)
396 397 398
{
    int ret;

399
    ret = vhdx_update_header(bs, s, generate_data_write_guid, log_guid);
400 401 402
    if (ret < 0) {
        return ret;
    }
403
    ret = vhdx_update_header(bs, s, generate_data_write_guid, log_guid);
404 405
    return ret;
}
406 407

/* opens the specified header block from the VHDX file header section */
P
Paolo Bonzini 已提交
408 409
static void vhdx_parse_header(BlockDriverState *bs, BDRVVHDXState *s,
                              Error **errp)
410
{
P
Paolo Bonzini 已提交
411
    int ret;
412 413 414 415 416 417 418 419
    VHDXHeader *header1;
    VHDXHeader *header2;
    bool h1_valid = false;
    bool h2_valid = false;
    uint64_t h1_seq = 0;
    uint64_t h2_seq = 0;
    uint8_t *buffer;

420
    /* header1 & header2 are freed in vhdx_close() */
421 422 423 424 425 426 427 428 429 430 431
    header1 = qemu_blockalign(bs, sizeof(VHDXHeader));
    header2 = qemu_blockalign(bs, sizeof(VHDXHeader));

    buffer = qemu_blockalign(bs, VHDX_HEADER_SIZE);

    s->headers[0] = header1;
    s->headers[1] = header2;

    /* We have to read the whole VHDX_HEADER_SIZE instead of
     * sizeof(VHDXHeader), because the checksum is over the whole
     * region */
K
Kevin Wolf 已提交
432 433
    ret = bdrv_pread(bs->file->bs, VHDX_HEADER1_OFFSET, buffer,
                     VHDX_HEADER_SIZE);
434 435 436 437 438 439
    if (ret < 0) {
        goto fail;
    }
    /* copy over just the relevant portion that we need */
    memcpy(header1, buffer, sizeof(VHDXHeader));

J
Jeff Cody 已提交
440 441 442 443 444 445 446
    if (vhdx_checksum_is_valid(buffer, VHDX_HEADER_SIZE, 4)) {
        vhdx_header_le_import(header1);
        if (header1->signature == VHDX_HEADER_SIGNATURE &&
            header1->version == 1) {
            h1_seq = header1->sequence_number;
            h1_valid = true;
        }
447 448
    }

K
Kevin Wolf 已提交
449 450
    ret = bdrv_pread(bs->file->bs, VHDX_HEADER2_OFFSET, buffer,
                     VHDX_HEADER_SIZE);
451 452 453 454 455 456
    if (ret < 0) {
        goto fail;
    }
    /* copy over just the relevant portion that we need */
    memcpy(header2, buffer, sizeof(VHDXHeader));

J
Jeff Cody 已提交
457 458 459 460 461 462 463
    if (vhdx_checksum_is_valid(buffer, VHDX_HEADER_SIZE, 4)) {
        vhdx_header_le_import(header2);
        if (header2->signature == VHDX_HEADER_SIGNATURE &&
            header2->version == 1) {
            h2_seq = header2->sequence_number;
            h2_valid = true;
        }
464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482
    }

    /* If there is only 1 valid header (or no valid headers), we
     * don't care what the sequence numbers are */
    if (h1_valid && !h2_valid) {
        s->curr_header = 0;
    } else if (!h1_valid && h2_valid) {
        s->curr_header = 1;
    } else if (!h1_valid && !h2_valid) {
        goto fail;
    } else {
        /* If both headers are valid, then we choose the active one by the
         * highest sequence number.  If the sequence numbers are equal, that is
         * invalid */
        if (h1_seq > h2_seq) {
            s->curr_header = 0;
        } else if (h2_seq > h1_seq) {
            s->curr_header = 1;
        } else {
483 484 485 486 487 488 489 490
            /* The Microsoft Disk2VHD tool will create 2 identical
             * headers, with identical sequence numbers.  If the headers are
             * identical, don't consider the file corrupt */
            if (!memcmp(header1, header2, sizeof(VHDXHeader))) {
                s->curr_header = 0;
            } else {
                goto fail;
            }
491 492 493
        }
    }

494 495
    vhdx_region_register(s, s->headers[s->curr_header]->log_offset,
                            s->headers[s->curr_header]->log_length);
496 497 498
    goto exit;

fail:
P
Paolo Bonzini 已提交
499
    error_setg_errno(errp, -ret, "No valid VHDX header found");
500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522
    qemu_vfree(header1);
    qemu_vfree(header2);
    s->headers[0] = NULL;
    s->headers[1] = NULL;
exit:
    qemu_vfree(buffer);
}


static int vhdx_open_region_tables(BlockDriverState *bs, BDRVVHDXState *s)
{
    int ret = 0;
    uint8_t *buffer;
    int offset = 0;
    VHDXRegionTableEntry rt_entry;
    uint32_t i;
    bool bat_rt_found = false;
    bool metadata_rt_found = false;

    /* We have to read the whole 64KB block, because the crc32 is over the
     * whole block */
    buffer = qemu_blockalign(bs, VHDX_HEADER_BLOCK_SIZE);

K
Kevin Wolf 已提交
523
    ret = bdrv_pread(bs->file->bs, VHDX_REGION_TABLE_OFFSET, buffer,
524 525 526 527 528 529 530
                     VHDX_HEADER_BLOCK_SIZE);
    if (ret < 0) {
        goto fail;
    }
    memcpy(&s->rt, buffer, sizeof(s->rt));
    offset += sizeof(s->rt);

J
Jeff Cody 已提交
531 532 533 534 535 536 537 538
    if (!vhdx_checksum_is_valid(buffer, VHDX_HEADER_BLOCK_SIZE, 4)) {
        ret = -EINVAL;
        goto fail;
    }

    vhdx_region_header_le_import(&s->rt);

    if (s->rt.signature != VHDX_REGION_SIGNATURE) {
539 540 541 542
        ret = -EINVAL;
        goto fail;
    }

J
Jeff Cody 已提交
543

544 545 546 547 548 549 550 551 552 553
    /* Per spec, maximum region table entry count is 2047 */
    if (s->rt.entry_count > 2047) {
        ret = -EINVAL;
        goto fail;
    }

    for (i = 0; i < s->rt.entry_count; i++) {
        memcpy(&rt_entry, buffer + offset, sizeof(rt_entry));
        offset += sizeof(rt_entry);

554
        vhdx_region_entry_le_import(&rt_entry);
555

556 557 558 559 560 561 562 563 564
        /* check for region overlap between these entries, and any
         * other memory regions in the file */
        ret = vhdx_region_check(s, rt_entry.file_offset, rt_entry.length);
        if (ret < 0) {
            goto fail;
        }

        vhdx_region_register(s, rt_entry.file_offset, rt_entry.length);

565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594
        /* see if we recognize the entry */
        if (guid_eq(rt_entry.guid, bat_guid)) {
            /* must be unique; if we have already found it this is invalid */
            if (bat_rt_found) {
                ret = -EINVAL;
                goto fail;
            }
            bat_rt_found = true;
            s->bat_rt = rt_entry;
            continue;
        }

        if (guid_eq(rt_entry.guid, metadata_guid)) {
            /* must be unique; if we have already found it this is invalid */
            if (metadata_rt_found) {
                ret = -EINVAL;
                goto fail;
            }
            metadata_rt_found = true;
            s->metadata_rt = rt_entry;
            continue;
        }

        if (rt_entry.data_bits & VHDX_REGION_ENTRY_REQUIRED) {
            /* cannot read vhdx file - required region table entry that
             * we do not understand.  per spec, we must fail to open */
            ret = -ENOTSUP;
            goto fail;
        }
    }
595 596 597 598 599 600

    if (!bat_rt_found || !metadata_rt_found) {
        ret = -EINVAL;
        goto fail;
    }

601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635
    ret = 0;

fail:
    qemu_vfree(buffer);
    return ret;
}



/* Metadata initial parser
 *
 * This loads all the metadata entry fields.  This may cause additional
 * fields to be processed (e.g. parent locator, etc..).
 *
 * There are 5 Metadata items that are always required:
 *      - File Parameters (block size, has a parent)
 *      - Virtual Disk Size (size, in bytes, of the virtual drive)
 *      - Page 83 Data (scsi page 83 guid)
 *      - Logical Sector Size (logical sector size in bytes, either 512 or
 *                             4096.  We only support 512 currently)
 *      - Physical Sector Size (512 or 4096)
 *
 * Also, if the File Parameters indicate this is a differencing file,
 * we must also look for the Parent Locator metadata item.
 */
static int vhdx_parse_metadata(BlockDriverState *bs, BDRVVHDXState *s)
{
    int ret = 0;
    uint8_t *buffer;
    int offset = 0;
    uint32_t i = 0;
    VHDXMetadataTableEntry md_entry;

    buffer = qemu_blockalign(bs, VHDX_METADATA_TABLE_MAX_SIZE);

K
Kevin Wolf 已提交
636
    ret = bdrv_pread(bs->file->bs, s->metadata_rt.file_offset, buffer,
637 638 639 640 641 642 643
                     VHDX_METADATA_TABLE_MAX_SIZE);
    if (ret < 0) {
        goto exit;
    }
    memcpy(&s->metadata_hdr, buffer, sizeof(s->metadata_hdr));
    offset += sizeof(s->metadata_hdr);

644
    vhdx_metadata_header_le_import(&s->metadata_hdr);
645

J
Jeff Cody 已提交
646
    if (s->metadata_hdr.signature != VHDX_METADATA_SIGNATURE) {
647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662
        ret = -EINVAL;
        goto exit;
    }

    s->metadata_entries.present = 0;

    if ((s->metadata_hdr.entry_count * sizeof(md_entry)) >
        (VHDX_METADATA_TABLE_MAX_SIZE - offset)) {
        ret = -EINVAL;
        goto exit;
    }

    for (i = 0; i < s->metadata_hdr.entry_count; i++) {
        memcpy(&md_entry, buffer + offset, sizeof(md_entry));
        offset += sizeof(md_entry);

663
        vhdx_metadata_entry_le_import(&md_entry);
664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738

        if (guid_eq(md_entry.item_id, file_param_guid)) {
            if (s->metadata_entries.present & META_FILE_PARAMETER_PRESENT) {
                ret = -EINVAL;
                goto exit;
            }
            s->metadata_entries.file_parameters_entry = md_entry;
            s->metadata_entries.present |= META_FILE_PARAMETER_PRESENT;
            continue;
        }

        if (guid_eq(md_entry.item_id, virtual_size_guid)) {
            if (s->metadata_entries.present & META_VIRTUAL_DISK_SIZE_PRESENT) {
                ret = -EINVAL;
                goto exit;
            }
            s->metadata_entries.virtual_disk_size_entry = md_entry;
            s->metadata_entries.present |= META_VIRTUAL_DISK_SIZE_PRESENT;
            continue;
        }

        if (guid_eq(md_entry.item_id, page83_guid)) {
            if (s->metadata_entries.present & META_PAGE_83_PRESENT) {
                ret = -EINVAL;
                goto exit;
            }
            s->metadata_entries.page83_data_entry = md_entry;
            s->metadata_entries.present |= META_PAGE_83_PRESENT;
            continue;
        }

        if (guid_eq(md_entry.item_id, logical_sector_guid)) {
            if (s->metadata_entries.present &
                META_LOGICAL_SECTOR_SIZE_PRESENT) {
                ret = -EINVAL;
                goto exit;
            }
            s->metadata_entries.logical_sector_size_entry = md_entry;
            s->metadata_entries.present |= META_LOGICAL_SECTOR_SIZE_PRESENT;
            continue;
        }

        if (guid_eq(md_entry.item_id, phys_sector_guid)) {
            if (s->metadata_entries.present & META_PHYS_SECTOR_SIZE_PRESENT) {
                ret = -EINVAL;
                goto exit;
            }
            s->metadata_entries.phys_sector_size_entry = md_entry;
            s->metadata_entries.present |= META_PHYS_SECTOR_SIZE_PRESENT;
            continue;
        }

        if (guid_eq(md_entry.item_id, parent_locator_guid)) {
            if (s->metadata_entries.present & META_PARENT_LOCATOR_PRESENT) {
                ret = -EINVAL;
                goto exit;
            }
            s->metadata_entries.parent_locator_entry = md_entry;
            s->metadata_entries.present |= META_PARENT_LOCATOR_PRESENT;
            continue;
        }

        if (md_entry.data_bits & VHDX_META_FLAGS_IS_REQUIRED) {
            /* cannot read vhdx file - required region table entry that
             * we do not understand.  per spec, we must fail to open */
            ret = -ENOTSUP;
            goto exit;
        }
    }

    if (s->metadata_entries.present != META_ALL_PRESENT) {
        ret = -ENOTSUP;
        goto exit;
    }

K
Kevin Wolf 已提交
739
    ret = bdrv_pread(bs->file->bs,
740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773
                     s->metadata_entries.file_parameters_entry.offset
                                         + s->metadata_rt.file_offset,
                     &s->params,
                     sizeof(s->params));

    if (ret < 0) {
        goto exit;
    }

    le32_to_cpus(&s->params.block_size);
    le32_to_cpus(&s->params.data_bits);


    /* We now have the file parameters, so we can tell if this is a
     * differencing file (i.e.. has_parent), is dynamic or fixed
     * sized (leave_blocks_allocated), and the block size */

    /* The parent locator required iff the file parameters has_parent set */
    if (s->params.data_bits & VHDX_PARAMS_HAS_PARENT) {
        if (s->metadata_entries.present & META_PARENT_LOCATOR_PRESENT) {
            /* TODO: parse  parent locator fields */
            ret = -ENOTSUP; /* temp, until differencing files are supported */
            goto exit;
        } else {
            /* if has_parent is set, but there is not parent locator present,
             * then that is an invalid combination */
            ret = -EINVAL;
            goto exit;
        }
    }

    /* determine virtual disk size, logical sector size,
     * and phys sector size */

K
Kevin Wolf 已提交
774
    ret = bdrv_pread(bs->file->bs,
775 776 777 778 779 780 781
                     s->metadata_entries.virtual_disk_size_entry.offset
                                           + s->metadata_rt.file_offset,
                     &s->virtual_disk_size,
                     sizeof(uint64_t));
    if (ret < 0) {
        goto exit;
    }
K
Kevin Wolf 已提交
782
    ret = bdrv_pread(bs->file->bs,
783 784 785 786 787 788 789
                     s->metadata_entries.logical_sector_size_entry.offset
                                             + s->metadata_rt.file_offset,
                     &s->logical_sector_size,
                     sizeof(uint32_t));
    if (ret < 0) {
        goto exit;
    }
K
Kevin Wolf 已提交
790
    ret = bdrv_pread(bs->file->bs,
791 792 793 794 795 796 797 798 799 800 801 802
                     s->metadata_entries.phys_sector_size_entry.offset
                                          + s->metadata_rt.file_offset,
                     &s->physical_sector_size,
                     sizeof(uint32_t));
    if (ret < 0) {
        goto exit;
    }

    le64_to_cpus(&s->virtual_disk_size);
    le32_to_cpus(&s->logical_sector_size);
    le32_to_cpus(&s->physical_sector_size);

803 804
    if (s->params.block_size < VHDX_BLOCK_SIZE_MIN ||
        s->params.block_size > VHDX_BLOCK_SIZE_MAX) {
805 806 807 808
        ret = -EINVAL;
        goto exit;
    }

809 810 811 812 813 814 815 816
    /* only 2 supported sector sizes */
    if (s->logical_sector_size != 512 && s->logical_sector_size != 4096) {
        ret = -EINVAL;
        goto exit;
    }

    /* Both block_size and sector_size are guaranteed powers of 2, below.
       Due to range checks above, s->sectors_per_block can never be < 256 */
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
    s->sectors_per_block = s->params.block_size / s->logical_sector_size;
    s->chunk_ratio = (VHDX_MAX_SECTORS_PER_BLOCK) *
                     (uint64_t)s->logical_sector_size /
                     (uint64_t)s->params.block_size;

    /* These values are ones we will want to use for division / multiplication
     * later on, and they are all guaranteed (per the spec) to be powers of 2,
     * so we can take advantage of that for shift operations during
     * reads/writes */
    if (s->logical_sector_size & (s->logical_sector_size - 1)) {
        ret = -EINVAL;
        goto exit;
    }
    if (s->sectors_per_block & (s->sectors_per_block - 1)) {
        ret = -EINVAL;
        goto exit;
    }
    if (s->chunk_ratio & (s->chunk_ratio - 1)) {
        ret = -EINVAL;
        goto exit;
    }
    s->block_size = s->params.block_size;
    if (s->block_size & (s->block_size - 1)) {
        ret = -EINVAL;
        goto exit;
    }

844
    vhdx_set_shift_bits(s);
845 846 847 848 849 850 851 852

    ret = 0;

exit:
    qemu_vfree(buffer);
    return ret;
}

853 854 855 856 857 858 859 860
/*
 * Calculate the number of BAT entries, including sector
 * bitmap entries.
 */
static void vhdx_calc_bat_entries(BDRVVHDXState *s)
{
    uint32_t data_blocks_cnt, bitmap_blocks_cnt;

861 862
    data_blocks_cnt = DIV_ROUND_UP(s->virtual_disk_size, s->block_size);
    bitmap_blocks_cnt = DIV_ROUND_UP(data_blocks_cnt, s->chunk_ratio);
863 864 865 866 867 868 869 870 871

    if (s->parent_entries) {
        s->bat_entries = bitmap_blocks_cnt * (s->chunk_ratio + 1);
    } else {
        s->bat_entries = data_blocks_cnt +
                         ((data_blocks_cnt - 1) >> s->chunk_ratio_bits);
    }

}
872

873 874 875 876
static void vhdx_close(BlockDriverState *bs)
{
    BDRVVHDXState *s = bs->opaque;
    qemu_vfree(s->headers[0]);
877
    s->headers[0] = NULL;
878
    qemu_vfree(s->headers[1]);
879
    s->headers[1] = NULL;
880
    qemu_vfree(s->bat);
881
    s->bat = NULL;
882
    qemu_vfree(s->parent_entries);
883
    s->parent_entries = NULL;
884 885
    migrate_del_blocker(s->migration_blocker);
    error_free(s->migration_blocker);
886 887
    qemu_vfree(s->log.hdr);
    s->log.hdr = NULL;
888
    vhdx_region_unregister_all(s);
889 890
}

M
Max Reitz 已提交
891 892
static int vhdx_open(BlockDriverState *bs, QDict *options, int flags,
                     Error **errp)
893 894 895 896 897
{
    BDRVVHDXState *s = bs->opaque;
    int ret = 0;
    uint32_t i;
    uint64_t signature;
P
Paolo Bonzini 已提交
898
    Error *local_err = NULL;
899 900

    s->bat = NULL;
901
    s->first_visible_write = true;
902 903

    qemu_co_mutex_init(&s->lock);
904
    QLIST_INIT(&s->regions);
905 906

    /* validate the file signature */
K
Kevin Wolf 已提交
907
    ret = bdrv_pread(bs->file->bs, 0, &signature, sizeof(uint64_t));
908 909 910 911 912 913 914 915
    if (ret < 0) {
        goto fail;
    }
    if (memcmp(&signature, "vhdxfile", 8)) {
        ret = -EINVAL;
        goto fail;
    }

916 917 918 919 920
    /* This is used for any header updates, for the file_write_guid.
     * The spec dictates that a new value should be used for the first
     * header update */
    vhdx_guid_generate(&s->session_guid);

P
Paolo Bonzini 已提交
921 922 923 924
    vhdx_parse_header(bs, s, &local_err);
    if (local_err != NULL) {
        error_propagate(errp, local_err);
        ret = -EINVAL;
925 926 927
        goto fail;
    }

928
    ret = vhdx_parse_log(bs, s, &s->log_replayed_on_open, errp);
929
    if (ret < 0) {
930 931 932 933
        goto fail;
    }

    ret = vhdx_open_region_tables(bs, s);
934
    if (ret < 0) {
935 936 937 938
        goto fail;
    }

    ret = vhdx_parse_metadata(bs, s);
939
    if (ret < 0) {
940 941
        goto fail;
    }
942

943 944 945 946 947 948
    s->block_size = s->params.block_size;

    /* the VHDX spec dictates that virtual_disk_size is always a multiple of
     * logical_sector_size */
    bs->total_sectors = s->virtual_disk_size >> s->logical_sector_size_bits;

949
    vhdx_calc_bat_entries(s);
950 951 952 953 954 955 956 957 958

    s->bat_offset = s->bat_rt.file_offset;

    if (s->bat_entries > s->bat_rt.length / sizeof(VHDXBatEntry)) {
        /* BAT allocation is not large enough for all entries */
        ret = -EINVAL;
        goto fail;
    }

959
    /* s->bat is freed in vhdx_close() */
K
Kevin Wolf 已提交
960
    s->bat = qemu_try_blockalign(bs->file->bs, s->bat_rt.length);
961 962 963 964 965
    if (s->bat == NULL) {
        ret = -ENOMEM;
        goto fail;
    }

K
Kevin Wolf 已提交
966
    ret = bdrv_pread(bs->file->bs, s->bat_offset, s->bat, s->bat_rt.length);
967 968 969 970
    if (ret < 0) {
        goto fail;
    }

971 972 973
    uint64_t payblocks = s->chunk_ratio;
    /* endian convert, and verify populated BAT field file offsets against
     * region table and log entries */
974 975
    for (i = 0; i < s->bat_entries; i++) {
        le64_to_cpus(&s->bat[i]);
976 977 978
        if (payblocks--) {
            /* payload bat entries */
            if ((s->bat[i] & VHDX_BAT_STATE_BIT_MASK) ==
J
Jeff Cody 已提交
979
                    PAYLOAD_BLOCK_FULLY_PRESENT) {
980 981 982 983 984 985 986 987 988 989 990
                ret = vhdx_region_check(s, s->bat[i] & VHDX_BAT_FILE_OFF_MASK,
                                        s->block_size);
                if (ret < 0) {
                    goto fail;
                }
            }
        } else {
            payblocks = s->chunk_ratio;
            /* Once differencing files are supported, verify sector bitmap
             * blocks here */
        }
991 992 993
    }

    if (flags & BDRV_O_RDWR) {
994
        ret = vhdx_update_headers(bs, s, false, NULL);
995 996 997
        if (ret < 0) {
            goto fail;
        }
998 999
    }

J
Jeff Cody 已提交
1000
    /* TODO: differencing files */
1001

J
Jeff Cody 已提交
1002
    /* Disable migration when VHDX images are used */
1003 1004 1005
    error_setg(&s->migration_blocker, "The vhdx format used by node '%s' "
               "does not support live migration",
               bdrv_get_device_or_node_name(bs));
J
Jeff Cody 已提交
1006 1007
    migrate_add_blocker(s->migration_blocker);

1008 1009
    return 0;
fail:
1010
    vhdx_close(bs);
1011 1012 1013 1014 1015 1016 1017 1018 1019 1020
    return ret;
}

static int vhdx_reopen_prepare(BDRVReopenState *state,
                               BlockReopenQueue *queue, Error **errp)
{
    return 0;
}


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 1049
/*
 * Perform sector to block offset translations, to get various
 * sector and file offsets into the image.  See VHDXSectorInfo
 */
static void vhdx_block_translate(BDRVVHDXState *s, int64_t sector_num,
                                 int nb_sectors, VHDXSectorInfo *sinfo)
{
    uint32_t block_offset;

    sinfo->bat_idx = sector_num >> s->sectors_per_block_bits;
    /* effectively a modulo - this gives us the offset into the block
     * (in sector sizes) for our sector number */
    block_offset = sector_num - (sinfo->bat_idx << s->sectors_per_block_bits);
    /* the chunk ratio gives us the interleaving of the sector
     * bitmaps, so we need to advance our page block index by the
     * sector bitmaps entry number */
    sinfo->bat_idx += sinfo->bat_idx >> s->chunk_ratio_bits;

    /* the number of sectors we can read/write in this cycle */
    sinfo->sectors_avail = s->sectors_per_block - block_offset;

    sinfo->bytes_left = sinfo->sectors_avail << s->logical_sector_size_bits;

    if (sinfo->sectors_avail > nb_sectors) {
        sinfo->sectors_avail = nb_sectors;
    }

    sinfo->bytes_avail = sinfo->sectors_avail << s->logical_sector_size_bits;

1050
    sinfo->file_offset = s->bat[sinfo->bat_idx] & VHDX_BAT_FILE_OFF_MASK;
1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067

    sinfo->block_offset = block_offset << s->logical_sector_size_bits;

    /* The file offset must be past the header section, so must be > 0 */
    if (sinfo->file_offset == 0) {
        return;
    }

    /* block offset is the offset in vhdx logical sectors, in
     * the payload data block. Convert that to a byte offset
     * in the block, and add in the payload data block offset
     * in the file, in bytes, to get the final read address */

    sinfo->file_offset += sinfo->block_offset;
}


P
Paolo Bonzini 已提交
1068 1069 1070 1071 1072 1073
static int vhdx_get_info(BlockDriverState *bs, BlockDriverInfo *bdi)
{
    BDRVVHDXState *s = bs->opaque;

    bdi->cluster_size = s->block_size;

1074 1075 1076
    bdi->unallocated_blocks_are_zero =
        (s->params.data_bits & VHDX_PARAMS_HAS_PARENT) == 0;

P
Paolo Bonzini 已提交
1077 1078 1079
    return 0;
}

1080

1081 1082 1083
static coroutine_fn int vhdx_co_readv(BlockDriverState *bs, int64_t sector_num,
                                      int nb_sectors, QEMUIOVector *qiov)
{
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
    BDRVVHDXState *s = bs->opaque;
    int ret = 0;
    VHDXSectorInfo sinfo;
    uint64_t bytes_done = 0;
    QEMUIOVector hd_qiov;

    qemu_iovec_init(&hd_qiov, qiov->niov);

    qemu_co_mutex_lock(&s->lock);

    while (nb_sectors > 0) {
        /* We are a differencing file, so we need to inspect the sector bitmap
         * to see if we have the data or not */
        if (s->params.data_bits & VHDX_PARAMS_HAS_PARENT) {
            /* not supported yet */
            ret = -ENOTSUP;
            goto exit;
        } else {
            vhdx_block_translate(s, sector_num, nb_sectors, &sinfo);

            qemu_iovec_reset(&hd_qiov);
            qemu_iovec_concat(&hd_qiov, qiov,  bytes_done, sinfo.bytes_avail);

            /* check the payload block state */
            switch (s->bat[sinfo.bat_idx] & VHDX_BAT_STATE_BIT_MASK) {
            case PAYLOAD_BLOCK_NOT_PRESENT: /* fall through */
1110 1111
            case PAYLOAD_BLOCK_UNDEFINED:
            case PAYLOAD_BLOCK_UNMAPPED:
1112
            case PAYLOAD_BLOCK_UNMAPPED_v095:
1113 1114 1115 1116
            case PAYLOAD_BLOCK_ZERO:
                /* return zero */
                qemu_iovec_memset(&hd_qiov, 0, 0, sinfo.bytes_avail);
                break;
J
Jeff Cody 已提交
1117
            case PAYLOAD_BLOCK_FULLY_PRESENT:
1118
                qemu_co_mutex_unlock(&s->lock);
K
Kevin Wolf 已提交
1119
                ret = bdrv_co_readv(bs->file->bs,
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
                                    sinfo.file_offset >> BDRV_SECTOR_BITS,
                                    sinfo.sectors_avail, &hd_qiov);
                qemu_co_mutex_lock(&s->lock);
                if (ret < 0) {
                    goto exit;
                }
                break;
            case PAYLOAD_BLOCK_PARTIALLY_PRESENT:
                /* we don't yet support difference files, fall through
                 * to error */
            default:
                ret = -EIO;
                goto exit;
                break;
            }
            nb_sectors -= sinfo.sectors_avail;
            sector_num += sinfo.sectors_avail;
            bytes_done += sinfo.bytes_avail;
        }
    }
    ret = 0;
exit:
    qemu_co_mutex_unlock(&s->lock);
    qemu_iovec_destroy(&hd_qiov);
    return ret;
1145 1146
}

J
Jeff Cody 已提交
1147 1148 1149 1150 1151 1152 1153 1154 1155 1156
/*
 * Allocate a new payload block at the end of the file.
 *
 * Allocation will happen at 1MB alignment inside the file
 *
 * Returns the file offset start of the new payload block
 */
static int vhdx_allocate_block(BlockDriverState *bs, BDRVVHDXState *s,
                                    uint64_t *new_offset)
{
K
Kevin Wolf 已提交
1157
    *new_offset = bdrv_getlength(bs->file->bs);
J
Jeff Cody 已提交
1158 1159 1160 1161

    /* per the spec, the address for a block is in units of 1MB */
    *new_offset = ROUND_UP(*new_offset, 1024 * 1024);

K
Kevin Wolf 已提交
1162
    return bdrv_truncate(bs->file->bs, *new_offset + s->block_size);
J
Jeff Cody 已提交
1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174
}

/*
 * Update the BAT table entry with the new file offset, and the new entry
 * state */
static void vhdx_update_bat_table_entry(BlockDriverState *bs, BDRVVHDXState *s,
                                       VHDXSectorInfo *sinfo,
                                       uint64_t *bat_entry_le,
                                       uint64_t *bat_offset, int state)
{
    /* The BAT entry is a uint64, with 44 bits for the file offset in units of
     * 1MB, and 3 bits for the block state. */
1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186
    if ((state == PAYLOAD_BLOCK_ZERO)        ||
        (state == PAYLOAD_BLOCK_UNDEFINED)   ||
        (state == PAYLOAD_BLOCK_NOT_PRESENT) ||
        (state == PAYLOAD_BLOCK_UNMAPPED)) {
        s->bat[sinfo->bat_idx]  = 0;  /* For PAYLOAD_BLOCK_ZERO, the
                                         FileOffsetMB field is denoted as
                                         'reserved' in the v1.0 spec.  If it is
                                         non-zero, MS Hyper-V will fail to read
                                         the disk image */
    } else {
        s->bat[sinfo->bat_idx]  = sinfo->file_offset;
    }
1187

J
Jeff Cody 已提交
1188 1189 1190 1191 1192 1193
    s->bat[sinfo->bat_idx] |= state & VHDX_BAT_STATE_BIT_MASK;

    *bat_entry_le = cpu_to_le64(s->bat[sinfo->bat_idx]);
    *bat_offset = s->bat_offset + sinfo->bat_idx * sizeof(VHDXBatEntry);

}
1194

1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206
/* Per the spec, on the first write of guest-visible data to the file the
 * data write guid must be updated in the header */
int vhdx_user_visible_write(BlockDriverState *bs, BDRVVHDXState *s)
{
    int ret = 0;
    if (s->first_visible_write) {
        s->first_visible_write = false;
        ret = vhdx_update_headers(bs, s, true, NULL);
    }
    return ret;
}

1207 1208 1209
static coroutine_fn int vhdx_co_writev(BlockDriverState *bs, int64_t sector_num,
                                      int nb_sectors, QEMUIOVector *qiov)
{
J
Jeff Cody 已提交
1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260
    int ret = -ENOTSUP;
    BDRVVHDXState *s = bs->opaque;
    VHDXSectorInfo sinfo;
    uint64_t bytes_done = 0;
    uint64_t bat_entry = 0;
    uint64_t bat_entry_offset = 0;
    QEMUIOVector hd_qiov;
    struct iovec iov1 = { 0 };
    struct iovec iov2 = { 0 };
    int sectors_to_write;
    int bat_state;
    uint64_t bat_prior_offset = 0;
    bool bat_update = false;

    qemu_iovec_init(&hd_qiov, qiov->niov);

    qemu_co_mutex_lock(&s->lock);

    ret = vhdx_user_visible_write(bs, s);
    if (ret < 0) {
        goto exit;
    }

    while (nb_sectors > 0) {
        bool use_zero_buffers = false;
        bat_update = false;
        if (s->params.data_bits & VHDX_PARAMS_HAS_PARENT) {
            /* not supported yet */
            ret = -ENOTSUP;
            goto exit;
        } else {
            vhdx_block_translate(s, sector_num, nb_sectors, &sinfo);
            sectors_to_write = sinfo.sectors_avail;

            qemu_iovec_reset(&hd_qiov);
            /* check the payload block state */
            bat_state = s->bat[sinfo.bat_idx] & VHDX_BAT_STATE_BIT_MASK;
            switch (bat_state) {
            case PAYLOAD_BLOCK_ZERO:
                /* in this case, we need to preserve zero writes for
                 * data that is not part of this write, so we must pad
                 * the rest of the buffer to zeroes */

                /* if we are on a posix system with ftruncate() that extends
                 * a file, then it is zero-filled for us.  On Win32, the raw
                 * layer uses SetFilePointer and SetFileEnd, which does not
                 * zero fill AFAIK */

                /* Queue another write of zero buffers if the underlying file
                 * does not zero-fill on file extension */

K
Kevin Wolf 已提交
1261
                if (bdrv_has_zero_init(bs->file->bs) == 0) {
J
Jeff Cody 已提交
1262 1263 1264 1265 1266 1267 1268 1269
                    use_zero_buffers = true;

                    /* zero fill the front, if any */
                    if (sinfo.block_offset) {
                        iov1.iov_len = sinfo.block_offset;
                        iov1.iov_base = qemu_blockalign(bs, iov1.iov_len);
                        memset(iov1.iov_base, 0, iov1.iov_len);
                        qemu_iovec_concat_iov(&hd_qiov, &iov1, 1, 0,
K
Kevin Wolf 已提交
1270
                                              iov1.iov_len);
J
Jeff Cody 已提交
1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285
                        sectors_to_write += iov1.iov_len >> BDRV_SECTOR_BITS;
                    }

                    /* our actual data */
                    qemu_iovec_concat(&hd_qiov, qiov,  bytes_done,
                                      sinfo.bytes_avail);

                    /* zero fill the back, if any */
                    if ((sinfo.bytes_avail - sinfo.block_offset) <
                         s->block_size) {
                        iov2.iov_len = s->block_size -
                                      (sinfo.bytes_avail + sinfo.block_offset);
                        iov2.iov_base = qemu_blockalign(bs, iov2.iov_len);
                        memset(iov2.iov_base, 0, iov2.iov_len);
                        qemu_iovec_concat_iov(&hd_qiov, &iov2, 1, 0,
K
Kevin Wolf 已提交
1286
                                              iov2.iov_len);
J
Jeff Cody 已提交
1287 1288 1289 1290 1291
                        sectors_to_write += iov2.iov_len >> BDRV_SECTOR_BITS;
                    }
                }
                /* fall through */
            case PAYLOAD_BLOCK_NOT_PRESENT: /* fall through */
1292
            case PAYLOAD_BLOCK_UNMAPPED:
1293
            case PAYLOAD_BLOCK_UNMAPPED_v095:
1294
            case PAYLOAD_BLOCK_UNDEFINED:
J
Jeff Cody 已提交
1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327
                bat_prior_offset = sinfo.file_offset;
                ret = vhdx_allocate_block(bs, s, &sinfo.file_offset);
                if (ret < 0) {
                    goto exit;
                }
                /* once we support differencing files, this may also be
                 * partially present */
                /* update block state to the newly specified state */
                vhdx_update_bat_table_entry(bs, s, &sinfo, &bat_entry,
                                            &bat_entry_offset,
                                            PAYLOAD_BLOCK_FULLY_PRESENT);
                bat_update = true;
                /* since we just allocated a block, file_offset is the
                 * beginning of the payload block. It needs to be the
                 * write address, which includes the offset into the block */
                if (!use_zero_buffers) {
                    sinfo.file_offset += sinfo.block_offset;
                }
                /* fall through */
            case PAYLOAD_BLOCK_FULLY_PRESENT:
                /* if the file offset address is in the header zone,
                 * there is a problem */
                if (sinfo.file_offset < (1024 * 1024)) {
                    ret = -EFAULT;
                    goto error_bat_restore;
                }

                if (!use_zero_buffers) {
                    qemu_iovec_concat(&hd_qiov, qiov,  bytes_done,
                                      sinfo.bytes_avail);
                }
                /* block exists, so we can just overwrite it */
                qemu_co_mutex_unlock(&s->lock);
K
Kevin Wolf 已提交
1328
                ret = bdrv_co_writev(bs->file->bs,
J
Jeff Cody 已提交
1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378
                                    sinfo.file_offset >> BDRV_SECTOR_BITS,
                                    sectors_to_write, &hd_qiov);
                qemu_co_mutex_lock(&s->lock);
                if (ret < 0) {
                    goto error_bat_restore;
                }
                break;
            case PAYLOAD_BLOCK_PARTIALLY_PRESENT:
                /* we don't yet support difference files, fall through
                 * to error */
            default:
                ret = -EIO;
                goto exit;
                break;
            }

            if (bat_update) {
                /* this will update the BAT entry into the log journal, and
                 * then flush the log journal out to disk */
                ret =  vhdx_log_write_and_flush(bs, s, &bat_entry,
                                                sizeof(VHDXBatEntry),
                                                bat_entry_offset);
                if (ret < 0) {
                    goto exit;
                }
            }

            nb_sectors -= sinfo.sectors_avail;
            sector_num += sinfo.sectors_avail;
            bytes_done += sinfo.bytes_avail;

        }
    }

    goto exit;

error_bat_restore:
    if (bat_update) {
        /* keep metadata in sync, and restore the bat entry state
         * if error. */
        sinfo.file_offset = bat_prior_offset;
        vhdx_update_bat_table_entry(bs, s, &sinfo, &bat_entry,
                                    &bat_entry_offset, bat_state);
    }
exit:
    qemu_vfree(iov1.iov_base);
    qemu_vfree(iov2.iov_base);
    qemu_co_mutex_unlock(&s->lock);
    qemu_iovec_destroy(&hd_qiov);
    return ret;
1379 1380 1381
}


1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394

/*
 * Create VHDX Headers
 *
 * There are 2 headers, and the highest sequence number will represent
 * the active header
 */
static int vhdx_create_new_headers(BlockDriverState *bs, uint64_t image_size,
                                   uint32_t log_size)
{
    int ret = 0;
    VHDXHeader *hdr = NULL;

1395
    hdr = g_new0(VHDXHeader, 1);
1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420

    hdr->signature       = VHDX_HEADER_SIGNATURE;
    hdr->sequence_number = g_random_int();
    hdr->log_version     = 0;
    hdr->version         = 1;
    hdr->log_length      = log_size;
    hdr->log_offset      = VHDX_HEADER_SECTION_END;
    vhdx_guid_generate(&hdr->file_write_guid);
    vhdx_guid_generate(&hdr->data_write_guid);

    ret = vhdx_write_header(bs, hdr, VHDX_HEADER1_OFFSET, false);
    if (ret < 0) {
        goto exit;
    }
    hdr->sequence_number++;
    ret = vhdx_write_header(bs, hdr, VHDX_HEADER2_OFFSET, false);
    if (ret < 0) {
        goto exit;
    }

exit:
    g_free(hdr);
    return ret;
}

1421 1422 1423 1424 1425 1426
#define VHDX_METADATA_ENTRY_BUFFER_SIZE \
                                    (sizeof(VHDXFileParameters)               +\
                                     sizeof(VHDXVirtualDiskSize)              +\
                                     sizeof(VHDXPage83Data)                   +\
                                     sizeof(VHDXVirtualDiskLogicalSectorSize) +\
                                     sizeof(VHDXVirtualDiskPhysicalSectorSize))
1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454

/*
 * Create the Metadata entries.
 *
 * For more details on the entries, see section 3.5 (pg 29) in the
 * VHDX 1.00 specification.
 *
 * We support 5 metadata entries (all required by spec):
 *          File Parameters,
 *          Virtual Disk Size,
 *          Page 83 Data,
 *          Logical Sector Size,
 *          Physical Sector Size
 *
 * The first 64KB of the Metadata section is reserved for the metadata
 * header and entries; beyond that, the metadata items themselves reside.
 */
static int vhdx_create_new_metadata(BlockDriverState *bs,
                                    uint64_t image_size,
                                    uint32_t block_size,
                                    uint32_t sector_size,
                                    uint64_t metadata_offset,
                                    VHDXImageType type)
{
    int ret = 0;
    uint32_t offset = 0;
    void *buffer = NULL;
    void *entry_buffer;
1455
    VHDXMetadataTableHeader *md_table;
1456 1457 1458 1459 1460 1461 1462 1463 1464
    VHDXMetadataTableEntry  *md_table_entry;

    /* Metadata entries */
    VHDXFileParameters     *mt_file_params;
    VHDXVirtualDiskSize    *mt_virtual_size;
    VHDXPage83Data         *mt_page83;
    VHDXVirtualDiskLogicalSectorSize  *mt_log_sector_size;
    VHDXVirtualDiskPhysicalSectorSize *mt_phys_sector_size;

1465
    entry_buffer = g_malloc0(VHDX_METADATA_ENTRY_BUFFER_SIZE);
1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545

    mt_file_params = entry_buffer;
    offset += sizeof(VHDXFileParameters);
    mt_virtual_size = entry_buffer + offset;
    offset += sizeof(VHDXVirtualDiskSize);
    mt_page83 = entry_buffer + offset;
    offset += sizeof(VHDXPage83Data);
    mt_log_sector_size = entry_buffer + offset;
    offset += sizeof(VHDXVirtualDiskLogicalSectorSize);
    mt_phys_sector_size = entry_buffer + offset;

    mt_file_params->block_size = cpu_to_le32(block_size);
    if (type == VHDX_TYPE_FIXED) {
        mt_file_params->data_bits |= VHDX_PARAMS_LEAVE_BLOCKS_ALLOCED;
        cpu_to_le32s(&mt_file_params->data_bits);
    }

    vhdx_guid_generate(&mt_page83->page_83_data);
    cpu_to_leguids(&mt_page83->page_83_data);
    mt_virtual_size->virtual_disk_size        = cpu_to_le64(image_size);
    mt_log_sector_size->logical_sector_size   = cpu_to_le32(sector_size);
    mt_phys_sector_size->physical_sector_size = cpu_to_le32(sector_size);

    buffer = g_malloc0(VHDX_HEADER_BLOCK_SIZE);
    md_table = buffer;

    md_table->signature   = VHDX_METADATA_SIGNATURE;
    md_table->entry_count = 5;
    vhdx_metadata_header_le_export(md_table);


    /* This will reference beyond the reserved table portion */
    offset = 64 * KiB;

    md_table_entry = buffer + sizeof(VHDXMetadataTableHeader);

    md_table_entry[0].item_id = file_param_guid;
    md_table_entry[0].offset  = offset;
    md_table_entry[0].length  = sizeof(VHDXFileParameters);
    md_table_entry[0].data_bits |= VHDX_META_FLAGS_IS_REQUIRED;
    offset += md_table_entry[0].length;
    vhdx_metadata_entry_le_export(&md_table_entry[0]);

    md_table_entry[1].item_id = virtual_size_guid;
    md_table_entry[1].offset  = offset;
    md_table_entry[1].length  = sizeof(VHDXVirtualDiskSize);
    md_table_entry[1].data_bits |= VHDX_META_FLAGS_IS_REQUIRED |
                                   VHDX_META_FLAGS_IS_VIRTUAL_DISK;
    offset += md_table_entry[1].length;
    vhdx_metadata_entry_le_export(&md_table_entry[1]);

    md_table_entry[2].item_id = page83_guid;
    md_table_entry[2].offset  = offset;
    md_table_entry[2].length  = sizeof(VHDXPage83Data);
    md_table_entry[2].data_bits |= VHDX_META_FLAGS_IS_REQUIRED |
                                   VHDX_META_FLAGS_IS_VIRTUAL_DISK;
    offset += md_table_entry[2].length;
    vhdx_metadata_entry_le_export(&md_table_entry[2]);

    md_table_entry[3].item_id = logical_sector_guid;
    md_table_entry[3].offset  = offset;
    md_table_entry[3].length  = sizeof(VHDXVirtualDiskLogicalSectorSize);
    md_table_entry[3].data_bits |= VHDX_META_FLAGS_IS_REQUIRED |
                                   VHDX_META_FLAGS_IS_VIRTUAL_DISK;
    offset += md_table_entry[3].length;
    vhdx_metadata_entry_le_export(&md_table_entry[3]);

    md_table_entry[4].item_id = phys_sector_guid;
    md_table_entry[4].offset  = offset;
    md_table_entry[4].length  = sizeof(VHDXVirtualDiskPhysicalSectorSize);
    md_table_entry[4].data_bits |= VHDX_META_FLAGS_IS_REQUIRED |
                                   VHDX_META_FLAGS_IS_VIRTUAL_DISK;
    vhdx_metadata_entry_le_export(&md_table_entry[4]);

    ret = bdrv_pwrite(bs, metadata_offset, buffer, VHDX_HEADER_BLOCK_SIZE);
    if (ret < 0) {
        goto exit;
    }

    ret = bdrv_pwrite(bs, metadata_offset + (64 * KiB), entry_buffer,
1546
                      VHDX_METADATA_ENTRY_BUFFER_SIZE);
1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567
    if (ret < 0) {
        goto exit;
    }


exit:
    g_free(buffer);
    g_free(entry_buffer);
    return ret;
}

/* This create the actual BAT itself.  We currently only support
 * 'Dynamic' and 'Fixed' image types.
 *
 *  Dynamic images: default state of the BAT is all zeroes.
 *
 *  Fixed images: default state of the BAT is fully populated, with
 *                file offsets and state PAYLOAD_BLOCK_FULLY_PRESENT.
 */
static int vhdx_create_bat(BlockDriverState *bs, BDRVVHDXState *s,
                           uint64_t image_size, VHDXImageType type,
J
Jeff Cody 已提交
1568 1569
                           bool use_zero_blocks, uint64_t file_offset,
                           uint32_t length)
1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583
{
    int ret = 0;
    uint64_t data_file_offset;
    uint64_t total_sectors = 0;
    uint64_t sector_num = 0;
    uint64_t unused;
    int block_state;
    VHDXSectorInfo sinfo;

    assert(s->bat == NULL);

    /* this gives a data start after BAT/bitmap entries, and well
     * past any metadata entries (with a 4 MB buffer for future
     * expansion */
J
Jeff Cody 已提交
1584
    data_file_offset = file_offset + length + 5 * MiB;
1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607
    total_sectors = image_size >> s->logical_sector_size_bits;

    if (type == VHDX_TYPE_DYNAMIC) {
        /* All zeroes, so we can just extend the file - the end of the BAT
         * is the furthest thing we have written yet */
        ret = bdrv_truncate(bs, data_file_offset);
        if (ret < 0) {
            goto exit;
        }
    } else if (type == VHDX_TYPE_FIXED) {
        ret = bdrv_truncate(bs, data_file_offset + image_size);
        if (ret < 0) {
            goto exit;
        }
    } else {
        ret = -ENOTSUP;
        goto exit;
    }

    if (type == VHDX_TYPE_FIXED ||
                use_zero_blocks ||
                bdrv_has_zero_init(bs) == 0) {
        /* for a fixed file, the default BAT entry is not zero */
1608
        s->bat = g_try_malloc0(length);
1609
        if (length && s->bat == NULL) {
1610 1611 1612
            ret = -ENOMEM;
            goto exit;
        }
1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626
        block_state = type == VHDX_TYPE_FIXED ? PAYLOAD_BLOCK_FULLY_PRESENT :
                                                PAYLOAD_BLOCK_NOT_PRESENT;
        block_state = use_zero_blocks ? PAYLOAD_BLOCK_ZERO : block_state;
        /* fill the BAT by emulating sector writes of sectors_per_block size */
        while (sector_num < total_sectors) {
            vhdx_block_translate(s, sector_num, s->sectors_per_block, &sinfo);
            sinfo.file_offset = data_file_offset +
                                (sector_num << s->logical_sector_size_bits);
            sinfo.file_offset = ROUND_UP(sinfo.file_offset, MiB);
            vhdx_update_bat_table_entry(bs, s, &sinfo, &unused, &unused,
                                        block_state);
            cpu_to_le64s(&s->bat[sinfo.bat_idx]);
            sector_num += s->sectors_per_block;
        }
J
Jeff Cody 已提交
1627
        ret = bdrv_pwrite(bs, file_offset, s->bat, length);
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
        if (ret < 0) {
            goto exit;
        }
    }



exit:
    g_free(s->bat);
    return ret;
}

/* Creates the region table header, and region table entries.
 * There are 2 supported region table entries: BAT, and Metadata/
 *
 * As the calculations for the BAT region table are also needed
 * to create the BAT itself, we will also cause the BAT to be
 * created.
 */
static int vhdx_create_new_region_table(BlockDriverState *bs,
                                        uint64_t image_size,
                                        uint32_t block_size,
                                        uint32_t sector_size,
                                        uint32_t log_size,
                                        bool use_zero_blocks,
                                        VHDXImageType type,
                                        uint64_t *metadata_offset)
{
    int ret = 0;
    uint32_t offset = 0;
    void *buffer = NULL;
J
Jeff Cody 已提交
1659 1660
    uint64_t bat_file_offset;
    uint32_t bat_length;
1661 1662 1663 1664 1665 1666 1667 1668 1669
    BDRVVHDXState *s = NULL;
    VHDXRegionTableHeader *region_table;
    VHDXRegionTableEntry *rt_bat;
    VHDXRegionTableEntry *rt_metadata;

    assert(metadata_offset != NULL);

    /* Populate enough of the BDRVVHDXState to be able to use the
     * pre-existing BAT calculation, translation, and update functions */
1670
    s = g_new0(BDRVVHDXState, 1);
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

    s->chunk_ratio = (VHDX_MAX_SECTORS_PER_BLOCK) *
                     (uint64_t) sector_size / (uint64_t) block_size;

    s->sectors_per_block = block_size / sector_size;
    s->virtual_disk_size = image_size;
    s->block_size = block_size;
    s->logical_sector_size = sector_size;

    vhdx_set_shift_bits(s);

    vhdx_calc_bat_entries(s);

    /* At this point the VHDX state is populated enough for creation */

    /* a single buffer is used so we can calculate the checksum over the
     * entire 64KB block */
    buffer = g_malloc0(VHDX_HEADER_BLOCK_SIZE);
    region_table = buffer;
    offset += sizeof(VHDXRegionTableHeader);
    rt_bat = buffer + offset;
    offset += sizeof(VHDXRegionTableEntry);
    rt_metadata  = buffer + offset;

    region_table->signature = VHDX_REGION_SIGNATURE;
    region_table->entry_count = 2;   /* BAT and Metadata */

    rt_bat->guid        = bat_guid;
    rt_bat->length      = ROUND_UP(s->bat_entries * sizeof(VHDXBatEntry), MiB);
    rt_bat->file_offset = ROUND_UP(VHDX_HEADER_SECTION_END + log_size, MiB);
    s->bat_offset = rt_bat->file_offset;

    rt_metadata->guid        = metadata_guid;
    rt_metadata->file_offset = ROUND_UP(rt_bat->file_offset + rt_bat->length,
                                        MiB);
    rt_metadata->length      = 1 * MiB; /* min size, and more than enough */
    *metadata_offset = rt_metadata->file_offset;

J
Jeff Cody 已提交
1709 1710 1711 1712 1713 1714 1715
    bat_file_offset = rt_bat->file_offset;
    bat_length = rt_bat->length;

    vhdx_region_header_le_export(region_table);
    vhdx_region_entry_le_export(rt_bat);
    vhdx_region_entry_le_export(rt_metadata);

1716 1717 1718 1719 1720 1721
    vhdx_update_checksum(buffer, VHDX_HEADER_BLOCK_SIZE,
                         offsetof(VHDXRegionTableHeader, checksum));


    /* The region table gives us the data we need to create the BAT,
     * so do that now */
J
Jeff Cody 已提交
1722 1723 1724 1725 1726
    ret = vhdx_create_bat(bs, s, image_size, type, use_zero_blocks,
                          bat_file_offset, bat_length);
    if (ret < 0) {
        goto exit;
    }
1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 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

    /* Now write out the region headers to disk */
    ret = bdrv_pwrite(bs, VHDX_REGION_TABLE_OFFSET, buffer,
                      VHDX_HEADER_BLOCK_SIZE);
    if (ret < 0) {
        goto exit;
    }

    ret = bdrv_pwrite(bs, VHDX_REGION_TABLE2_OFFSET, buffer,
                      VHDX_HEADER_BLOCK_SIZE);
    if (ret < 0) {
        goto exit;
    }

exit:
    g_free(s);
    g_free(buffer);
    return ret;
}

/* We need to create the following elements:
 *
 *    .-----------------------------------------------------------------.
 *    |   (A)    |   (B)    |    (C)    |     (D)       |     (E)       |
 *    |  File ID |  Header1 |  Header 2 |  Region Tbl 1 |  Region Tbl 2 |
 *    |          |          |           |               |               |
 *    .-----------------------------------------------------------------.
 *    0         64KB      128KB       192KB           256KB           320KB
 *
 *
 *    .---- ~ ----------- ~ ------------ ~ ---------------- ~ -----------.
 *    |     (F)     |     (G)       |    (H)    |                        |
 *    | Journal Log |  BAT / Bitmap |  Metadata |  .... data ......      |
 *    |             |               |           |                        |
 *    .---- ~ ----------- ~ ------------ ~ ---------------- ~ -----------.
 *   1MB
 */
1764
static int vhdx_create(const char *filename, QemuOpts *opts, Error **errp)
1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775
{
    int ret = 0;
    uint64_t image_size = (uint64_t) 2 * GiB;
    uint32_t log_size   = 1 * MiB;
    uint32_t block_size = 0;
    uint64_t signature;
    uint64_t metadata_offset;
    bool use_zero_blocks = false;

    gunichar2 *creator = NULL;
    glong creator_items;
1776
    BlockBackend *blk;
1777
    char *type = NULL;
1778 1779 1780
    VHDXImageType image_type;
    Error *local_err = NULL;

1781 1782
    image_size = ROUND_UP(qemu_opt_get_size_del(opts, BLOCK_OPT_SIZE, 0),
                          BDRV_SECTOR_SIZE);
1783 1784 1785
    log_size = qemu_opt_get_size_del(opts, VHDX_BLOCK_OPT_LOG_SIZE, 0);
    block_size = qemu_opt_get_size_del(opts, VHDX_BLOCK_OPT_BLOCK_SIZE, 0);
    type = qemu_opt_get_del(opts, BLOCK_OPT_SUBFMT);
1786
    use_zero_blocks = qemu_opt_get_bool_del(opts, VHDX_BLOCK_OPT_ZERO, true);
1787 1788 1789 1790 1791 1792 1793 1794

    if (image_size > VHDX_MAX_IMAGE_SIZE) {
        error_setg_errno(errp, EINVAL, "Image size too large; max of 64TB");
        ret = -EINVAL;
        goto exit;
    }

    if (type == NULL) {
1795
        type = g_strdup("dynamic");
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
    }

    if (!strcmp(type, "dynamic")) {
        image_type = VHDX_TYPE_DYNAMIC;
    } else if (!strcmp(type, "fixed")) {
        image_type = VHDX_TYPE_FIXED;
    } else if (!strcmp(type, "differencing")) {
        error_setg_errno(errp, ENOTSUP,
                         "Differencing files not yet supported");
        ret = -ENOTSUP;
        goto exit;
    } else {
        ret = -EINVAL;
        goto exit;
    }

    /* These are pretty arbitrary, and mainly designed to keep the BAT
     * size reasonable to load into RAM */
    if (block_size == 0) {
        if (image_size > 32 * TiB) {
            block_size = 64 * MiB;
        } else if (image_size > (uint64_t) 100 * GiB) {
            block_size = 32 * MiB;
        } else if (image_size > 1 * GiB) {
            block_size = 16 * MiB;
        } else {
            block_size = 8 * MiB;
        }
    }


    /* make the log size close to what was specified, but must be
     * min 1MB, and multiple of 1MB */
    log_size = ROUND_UP(log_size, MiB);

    block_size = ROUND_UP(block_size, MiB);
    block_size = block_size > VHDX_BLOCK_SIZE_MAX ? VHDX_BLOCK_SIZE_MAX :
                                                    block_size;

C
Chunyan Liu 已提交
1835
    ret = bdrv_create_file(filename, opts, &local_err);
1836 1837 1838 1839 1840
    if (ret < 0) {
        error_propagate(errp, local_err);
        goto exit;
    }

1841 1842 1843 1844
    blk = blk_new_open("image", filename, NULL, NULL,
                       BDRV_O_RDWR | BDRV_O_CACHE_WB | BDRV_O_PROTOCOL,
                       &local_err);
    if (blk == NULL) {
1845
        error_propagate(errp, local_err);
1846
        ret = -EIO;
1847 1848 1849
        goto exit;
    }

1850 1851
    blk_set_allow_write_beyond_eof(blk, true);

1852 1853 1854 1855 1856 1857 1858
    /* Create (A) */

    /* The creator field is optional, but may be useful for
     * debugging / diagnostics */
    creator = g_utf8_to_utf16("QEMU v" QEMU_VERSION, -1, NULL,
                              &creator_items, NULL);
    signature = cpu_to_le64(VHDX_FILE_SIGNATURE);
1859
    ret = blk_pwrite(blk, VHDX_FILE_ID_OFFSET, &signature, sizeof(signature));
1860 1861 1862 1863
    if (ret < 0) {
        goto delete_and_exit;
    }
    if (creator) {
1864 1865
        ret = blk_pwrite(blk, VHDX_FILE_ID_OFFSET + sizeof(signature),
                         creator, creator_items * sizeof(gunichar2));
1866 1867 1868 1869 1870 1871 1872
        if (ret < 0) {
            goto delete_and_exit;
        }
    }


    /* Creates (B),(C) */
1873
    ret = vhdx_create_new_headers(blk_bs(blk), image_size, log_size);
1874 1875 1876 1877 1878
    if (ret < 0) {
        goto delete_and_exit;
    }

    /* Creates (D),(E),(G) explicitly. (F) created as by-product */
1879
    ret = vhdx_create_new_region_table(blk_bs(blk), image_size, block_size, 512,
1880 1881 1882 1883 1884 1885 1886
                                       log_size, use_zero_blocks, image_type,
                                       &metadata_offset);
    if (ret < 0) {
        goto delete_and_exit;
    }

    /* Creates (H) */
1887
    ret = vhdx_create_new_metadata(blk_bs(blk), image_size, block_size, 512,
1888 1889 1890 1891 1892 1893 1894
                                   metadata_offset, image_type);
    if (ret < 0) {
        goto delete_and_exit;
    }


delete_and_exit:
1895
    blk_unref(blk);
1896
exit:
1897
    g_free(type);
1898 1899 1900 1901
    g_free(creator);
    return ret;
}

1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919
/* If opened r/w, the VHDX driver will automatically replay the log,
 * if one is present, inside the vhdx_open() call.
 *
 * If qemu-img check -r all is called, the image is automatically opened
 * r/w and any log has already been replayed, so there is nothing (currently)
 * for us to do here
 */
static int vhdx_check(BlockDriverState *bs, BdrvCheckResult *result,
                       BdrvCheckMode fix)
{
    BDRVVHDXState *s = bs->opaque;

    if (s->log_replayed_on_open) {
        result->corruptions_fixed++;
    }
    return 0;
}

1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950
static QemuOptsList vhdx_create_opts = {
    .name = "vhdx-create-opts",
    .head = QTAILQ_HEAD_INITIALIZER(vhdx_create_opts.head),
    .desc = {
        {
           .name = BLOCK_OPT_SIZE,
           .type = QEMU_OPT_SIZE,
           .help = "Virtual disk size; max of 64TB."
       },
       {
           .name = VHDX_BLOCK_OPT_LOG_SIZE,
           .type = QEMU_OPT_SIZE,
           .def_value_str = stringify(DEFAULT_LOG_SIZE),
           .help = "Log size; min 1MB."
       },
       {
           .name = VHDX_BLOCK_OPT_BLOCK_SIZE,
           .type = QEMU_OPT_SIZE,
           .def_value_str = stringify(0),
           .help = "Block Size; min 1MB, max 256MB. " \
                   "0 means auto-calculate based on image size."
       },
       {
           .name = BLOCK_OPT_SUBFMT,
           .type = QEMU_OPT_STRING,
           .help = "VHDX format type, can be either 'dynamic' or 'fixed'. "\
                   "Default is 'dynamic'."
       },
       {
           .name = VHDX_BLOCK_OPT_ZERO,
           .type = QEMU_OPT_BOOL,
1951 1952 1953
           .help = "Force use of payload blocks of type 'ZERO'. "\
                   "Non-standard, but default.  Do not set to 'off' when "\
                   "using 'qemu-img convert' with subformat=dynamic."
1954 1955 1956
       },
       { NULL }
    }
1957 1958
};

1959 1960 1961 1962 1963 1964 1965 1966 1967
static BlockDriver bdrv_vhdx = {
    .format_name            = "vhdx",
    .instance_size          = sizeof(BDRVVHDXState),
    .bdrv_probe             = vhdx_probe,
    .bdrv_open              = vhdx_open,
    .bdrv_close             = vhdx_close,
    .bdrv_reopen_prepare    = vhdx_reopen_prepare,
    .bdrv_co_readv          = vhdx_co_readv,
    .bdrv_co_writev         = vhdx_co_writev,
C
Chunyan Liu 已提交
1968
    .bdrv_create            = vhdx_create,
P
Paolo Bonzini 已提交
1969
    .bdrv_get_info          = vhdx_get_info,
1970
    .bdrv_check             = vhdx_check,
1971
    .bdrv_has_zero_init     = bdrv_has_zero_init_1,
1972

1973
    .create_opts            = &vhdx_create_opts,
1974 1975 1976 1977 1978 1979 1980 1981
};

static void bdrv_vhdx_init(void)
{
    bdrv_register(&bdrv_vhdx);
}

block_init(bdrv_vhdx_init);