summaryrefslogtreecommitdiff
blob: f5e38dbd28b816a372430a370e6c9b4f513c2757 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
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
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
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
209
210
211
212
213
214
215
216
217
218
219
220
221
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
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
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
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
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
595
596
597
598
599
600
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
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
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
739
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
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
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
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
/*
 * Block driver for the Virtual Disk Image (VDI) format
 *
 * Copyright (c) 2009 Stefan Weil
 *
 * This program is free software: you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation, either version 2 of the License, or
 * (at your option) version 3 or any later version.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with this program.  If not, see <http://www.gnu.org/licenses/>.
 *
 * Reference:
 * http://forums.virtualbox.org/viewtopic.php?t=8046
 *
 * This driver supports create / read / write operations on VDI images.
 *
 * Todo (see also TODO in code):
 *
 * Some features like snapshots are still missing.
 *
 * Deallocation of zero-filled blocks and shrinking images are missing, too
 * (might be added to common block layer).
 *
 * Allocation of blocks could be optimized (less writes to block map and
 * header).
 *
 * Read and write of adjacents blocks could be done in one operation
 * (current code uses one operation per block (1 MiB).
 *
 * The code is not thread safe (missing locks for changes in header and
 * block table, no problem with current QEMU).
 *
 * Hints:
 *
 * Blocks (VDI documentation) correspond to clusters (QEMU).
 * QEMU's backing files could be implemented using VDI snapshot files (TODO).
 * VDI snapshot files may also contain the complete machine state.
 * Maybe this machine state can be converted to QEMU PC machine snapshot data.
 *
 * The driver keeps a block cache (little endian entries) in memory.
 * For the standard block size (1 MiB), a 1 TiB disk will use 4 MiB RAM,
 * so this seems to be reasonable.
 */

#include "qemu-common.h"
#include "block_int.h"
#include "module.h"

#if defined(HAVE_UUID_H)
#include <uuid/uuid.h>
#else
/* TODO: move uuid emulation to some central place in QEMU. */
#include "sysemu.h"     /* UUID_FMT */
typedef unsigned char uuid_t[16];
void uuid_generate(uuid_t out);
int uuid_is_null(const uuid_t uu);
void uuid_unparse(const uuid_t uu, char *out);
#endif

/* Code configuration options. */

/* Enable debug messages. */
//~ #define CONFIG_VDI_DEBUG

/* Support write operations on VDI images. */
#define CONFIG_VDI_WRITE

/* Support non-standard block (cluster) size. This is untested.
 * Maybe it will be needed for very large images.
 */
//~ #define CONFIG_VDI_BLOCK_SIZE

/* Support static (fixed, pre-allocated) images. */
#define CONFIG_VDI_STATIC_IMAGE

/* Command line option for static images. */
#define BLOCK_OPT_STATIC "static"

#define KiB     1024
#define MiB     (KiB * KiB)

#define SECTOR_SIZE 512

#if defined(CONFIG_VDI_DEBUG)
#define logout(fmt, ...) \
                fprintf(stderr, "vdi\t%-24s" fmt, __func__, ##__VA_ARGS__)
#else
#define logout(fmt, ...) ((void)0)
#endif

/* Image signature. */
#define VDI_SIGNATURE 0xbeda107f

/* Image version. */
#define VDI_VERSION_1_1 0x00010001

/* Image type. */
#define VDI_TYPE_DYNAMIC 1
#define VDI_TYPE_STATIC  2

/* Innotek / SUN images use these strings in header.text:
 * "<<< innotek VirtualBox Disk Image >>>\n"
 * "<<< Sun xVM VirtualBox Disk Image >>>\n"
 * "<<< Sun VirtualBox Disk Image >>>\n"
 * The value does not matter, so QEMU created images use a different text.
 */
#define VDI_TEXT "<<< QEMU VM Virtual Disk Image >>>\n"

/* Unallocated blocks use this index (no need to convert endianess). */
#define VDI_UNALLOCATED UINT32_MAX

#if !defined(HAVE_UUID_H)
void uuid_generate(uuid_t out)
{
    memset(out, 0, sizeof(out));
}

int uuid_is_null(const uuid_t uu)
{
    uuid_t null_uuid = { 0 };
    return memcmp(uu, null_uuid, sizeof(uu)) == 0;
}

void uuid_unparse(const uuid_t uu, char *out)
{
    snprintf(out, 37, UUID_FMT,
            uu[0], uu[1], uu[2], uu[3], uu[4], uu[5], uu[6], uu[7],
            uu[8], uu[9], uu[10], uu[11], uu[12], uu[13], uu[14], uu[15]);
}
#endif

typedef struct {
    BlockDriverAIOCB common;
    int64_t sector_num;
    QEMUIOVector *qiov;
    uint8_t *buf;
    /* Total number of sectors. */
    int nb_sectors;
    /* Number of sectors for current AIO. */
    int n_sectors;
    /* New allocated block map entry. */
    uint32_t bmap_first;
    uint32_t bmap_last;
    /* Buffer for new allocated block. */
    void *block_buffer;
    void *orig_buf;
    int header_modified;
    BlockDriverAIOCB *hd_aiocb;
    struct iovec hd_iov;
    QEMUIOVector hd_qiov;
    QEMUBH *bh;
} VdiAIOCB;

typedef struct {
    char text[0x40];
    uint32_t signature;
    uint32_t version;
    uint32_t header_size;
    uint32_t image_type;
    uint32_t image_flags;
    char description[256];
    uint32_t offset_bmap;
    uint32_t offset_data;
    uint32_t cylinders;         /* disk geometry, unused here */
    uint32_t heads;             /* disk geometry, unused here */
    uint32_t sectors;           /* disk geometry, unused here */
    uint32_t sector_size;
    uint32_t unused1;
    uint64_t disk_size;
    uint32_t block_size;
    uint32_t block_extra;       /* unused here */
    uint32_t blocks_in_image;
    uint32_t blocks_allocated;
    uuid_t uuid_image;
    uuid_t uuid_last_snap;
    uuid_t uuid_link;
    uuid_t uuid_parent;
    uint64_t unused2[7];
} VdiHeader;

typedef struct {
    BlockDriverState *hd;
    /* The block map entries are little endian (even in memory). */
    uint32_t *bmap;
    /* Size of block (bytes). */
    uint32_t block_size;
    /* Size of block (sectors). */
    uint32_t block_sectors;
    /* First sector of block map. */
    uint32_t bmap_sector;
    /* VDI header (converted to host endianess). */
    VdiHeader header;
} BDRVVdiState;

/* Change UUID from little endian (IPRT = VirtualBox format) to big endian
 * format (network byte order, standard, see RFC 4122) and vice versa.
 */
static void uuid_convert(uuid_t uuid)
{
    bswap32s((uint32_t *)&uuid[0]);
    bswap16s((uint16_t *)&uuid[4]);
    bswap16s((uint16_t *)&uuid[6]);
}

static void vdi_header_to_cpu(VdiHeader *header)
{
    le32_to_cpus(&header->signature);
    le32_to_cpus(&header->version);
    le32_to_cpus(&header->header_size);
    le32_to_cpus(&header->image_type);
    le32_to_cpus(&header->image_flags);
    le32_to_cpus(&header->offset_bmap);
    le32_to_cpus(&header->offset_data);
    le32_to_cpus(&header->cylinders);
    le32_to_cpus(&header->heads);
    le32_to_cpus(&header->sectors);
    le32_to_cpus(&header->sector_size);
    le64_to_cpus(&header->disk_size);
    le32_to_cpus(&header->block_size);
    le32_to_cpus(&header->block_extra);
    le32_to_cpus(&header->blocks_in_image);
    le32_to_cpus(&header->blocks_allocated);
    uuid_convert(header->uuid_image);
    uuid_convert(header->uuid_last_snap);
    uuid_convert(header->uuid_link);
    uuid_convert(header->uuid_parent);
}

static void vdi_header_to_le(VdiHeader *header)
{
    cpu_to_le32s(&header->signature);
    cpu_to_le32s(&header->version);
    cpu_to_le32s(&header->header_size);
    cpu_to_le32s(&header->image_type);
    cpu_to_le32s(&header->image_flags);
    cpu_to_le32s(&header->offset_bmap);
    cpu_to_le32s(&header->offset_data);
    cpu_to_le32s(&header->cylinders);
    cpu_to_le32s(&header->heads);
    cpu_to_le32s(&header->sectors);
    cpu_to_le32s(&header->sector_size);
    cpu_to_le64s(&header->disk_size);
    cpu_to_le32s(&header->block_size);
    cpu_to_le32s(&header->block_extra);
    cpu_to_le32s(&header->blocks_in_image);
    cpu_to_le32s(&header->blocks_allocated);
    cpu_to_le32s(&header->blocks_allocated);
    uuid_convert(header->uuid_image);
    uuid_convert(header->uuid_last_snap);
    uuid_convert(header->uuid_link);
    uuid_convert(header->uuid_parent);
}

#if defined(CONFIG_VDI_DEBUG)
static void vdi_header_print(VdiHeader *header)
{
    char uuid[37];
    logout("text        %s", header->text);
    logout("signature   0x%04x\n", header->signature);
    logout("header size 0x%04x\n", header->header_size);
    logout("image type  0x%04x\n", header->image_type);
    logout("image flags 0x%04x\n", header->image_flags);
    logout("description %s\n", header->description);
    logout("offset bmap 0x%04x\n", header->offset_bmap);
    logout("offset data 0x%04x\n", header->offset_data);
    logout("cylinders   0x%04x\n", header->cylinders);
    logout("heads       0x%04x\n", header->heads);
    logout("sectors     0x%04x\n", header->sectors);
    logout("sector size 0x%04x\n", header->sector_size);
    logout("image size  0x%" PRIx64 " B (%" PRIu64 " MiB)\n",
           header->disk_size, header->disk_size / MiB);
    logout("block size  0x%04x\n", header->block_size);
    logout("block extra 0x%04x\n", header->block_extra);
    logout("blocks tot. 0x%04x\n", header->blocks_in_image);
    logout("blocks all. 0x%04x\n", header->blocks_allocated);
    uuid_unparse(header->uuid_image, uuid);
    logout("uuid image  %s\n", uuid);
    uuid_unparse(header->uuid_last_snap, uuid);
    logout("uuid snap   %s\n", uuid);
    uuid_unparse(header->uuid_link, uuid);
    logout("uuid link   %s\n", uuid);
    uuid_unparse(header->uuid_parent, uuid);
    logout("uuid parent %s\n", uuid);
}
#endif

static int vdi_check(BlockDriverState *bs)
{
    /* TODO: additional checks possible. */
    BDRVVdiState *s = (BDRVVdiState *)bs->opaque;
    int n_errors = 0;
    uint32_t blocks_allocated = 0;
    uint32_t block;
    uint32_t *bmap;
    logout("\n");

    bmap = qemu_malloc(s->header.blocks_in_image * sizeof(uint32_t));
    memset(bmap, 0xff, s->header.blocks_in_image * sizeof(uint32_t));

    /* Check block map and value of blocks_allocated. */
    for (block = 0; block < s->header.blocks_in_image; block++) {
        uint32_t bmap_entry = le32_to_cpu(s->bmap[block]);
        if (bmap_entry != VDI_UNALLOCATED) {
            if (bmap_entry < s->header.blocks_in_image) {
                blocks_allocated++;
                if (bmap[bmap_entry] == VDI_UNALLOCATED) {
                    bmap[bmap_entry] = bmap_entry;
                } else {
                    fprintf(stderr, "ERROR: block index %" PRIu32
                            " also used by %" PRIu32 "\n", bmap[bmap_entry], bmap_entry);
                }
            } else {
                fprintf(stderr, "ERROR: block index %" PRIu32
                        " too large, is %" PRIu32 "\n", block, bmap_entry);
                n_errors++;
            }
        }
    }
    if (blocks_allocated != s->header.blocks_allocated) {
        fprintf(stderr, "ERROR: allocated blocks mismatch, is %" PRIu32
               ", should be %" PRIu32 "\n",
               blocks_allocated, s->header.blocks_allocated);
        n_errors++;
    }

    qemu_free(bmap);

    return n_errors;
}

static int vdi_get_info(BlockDriverState *bs, BlockDriverInfo *bdi)
{
    /* TODO: vdi_get_info would be needed for machine snapshots.
       vm_state_offset is still missing. */
    BDRVVdiState *s = (BDRVVdiState *)bs->opaque;
    logout("\n");
    bdi->cluster_size = s->block_size;
    bdi->vm_state_offset = 0;
    return 0;
}

static int vdi_make_empty(BlockDriverState *bs)
{
    /* TODO: missing code. */
    logout("\n");
    /* The return value for missing code must be 0, see block.c. */
    return 0;
}

static int vdi_probe(const uint8_t *buf, int buf_size, const char *filename)
{
    const VdiHeader *header = (const VdiHeader *)buf;
    int result = 0;

    logout("\n");

    if (buf_size < sizeof(*header)) {
        /* Header too small, no VDI. */
    } else if (le32_to_cpu(header->signature) == VDI_SIGNATURE) {
        result = 100;
    }

    if (result == 0) {
        logout("no vdi image\n");
    } else {
        logout("%s", header->text);
    }

    return result;
}

static int vdi_open(BlockDriverState *bs, const char *filename, int flags)
{
    BDRVVdiState *s = bs->opaque;
    VdiHeader header;
    size_t bmap_size;
    int ret;

    logout("\n");

    ret = bdrv_file_open(&s->hd, filename, flags);
    if (ret < 0) {
        return ret;
    }

    if (bdrv_read(s->hd, 0, (uint8_t *)&header, 1) < 0) {
        goto fail;
    }

    vdi_header_to_cpu(&header);
#if defined(CONFIG_VDI_DEBUG)
    vdi_header_print(&header);
#endif

    if (header.version != VDI_VERSION_1_1) {
        logout("unsupported version %u.%u\n",
               header.version >> 16, header.version & 0xffff);
        goto fail;
    } else if (header.offset_bmap % SECTOR_SIZE != 0) {
        /* We only support block maps which start on a sector boundary. */
        logout("unsupported block map offset 0x%x B\n", header.offset_bmap);
        goto fail;
    } else if (header.offset_data % SECTOR_SIZE != 0) {
        /* We only support data blocks which start on a sector boundary. */
        logout("unsupported data offset 0x%x B\n", header.offset_data);
        goto fail;
    } else if (header.sector_size != SECTOR_SIZE) {
        logout("unsupported sector size %u B\n", header.sector_size);
        goto fail;
    } else if (header.block_size != 1 * MiB) {
        logout("unsupported block size %u B\n", header.block_size);
        goto fail;
    } else if (header.disk_size !=
               (uint64_t)header.blocks_in_image * header.block_size) {
        logout("unexpected block number %u B\n", header.blocks_in_image);
        goto fail;
    } else if (!uuid_is_null(header.uuid_link)) {
        logout("link uuid != 0, unsupported\n");
        goto fail;
    } else if (!uuid_is_null(header.uuid_parent)) {
        logout("parent uuid != 0, unsupported\n");
        goto fail;
    }

    bs->total_sectors = header.disk_size / SECTOR_SIZE;

    s->block_size = header.block_size;
    s->block_sectors = header.block_size / SECTOR_SIZE;
    s->bmap_sector = header.offset_bmap / SECTOR_SIZE;
    s->header = header;

    bmap_size = header.blocks_in_image * sizeof(uint32_t);
    bmap_size = (bmap_size + SECTOR_SIZE - 1) / SECTOR_SIZE;
    s->bmap = qemu_malloc(bmap_size * SECTOR_SIZE);
    if (bdrv_read(s->hd, s->bmap_sector, (uint8_t *)s->bmap, bmap_size) < 0) {
        goto fail_free_bmap;
    }

    return 0;

 fail_free_bmap:
    qemu_free(s->bmap);

 fail:
    bdrv_delete(s->hd);
    return -1;
}

static int vdi_is_allocated(BlockDriverState *bs, int64_t sector_num,
                             int nb_sectors, int *pnum)
{
    /* TODO: Check for too large sector_num (in bdrv_is_allocated or here). */
    BDRVVdiState *s = (BDRVVdiState *)bs->opaque;
    size_t bmap_index = sector_num / s->block_sectors;
    size_t sector_in_block = sector_num % s->block_sectors;
    int n_sectors = s->block_sectors - sector_in_block;
    uint32_t bmap_entry = le32_to_cpu(s->bmap[bmap_index]);
    logout("%p, %" PRId64 ", %d, %p\n", bs, sector_num, nb_sectors, pnum);
    if (n_sectors > nb_sectors) {
        n_sectors = nb_sectors;
    }
    *pnum = n_sectors;
    return bmap_entry != VDI_UNALLOCATED;
}

static void vdi_aio_cancel(BlockDriverAIOCB *blockacb)
{
    /* TODO: This code is untested. How can I get it executed? */
    VdiAIOCB *acb = (VdiAIOCB *)blockacb;
    logout("\n");
    if (acb->hd_aiocb) {
        bdrv_aio_cancel(acb->hd_aiocb);
    }
    qemu_aio_release(acb);
}

static AIOPool vdi_aio_pool = {
    .aiocb_size = sizeof(VdiAIOCB),
    .cancel = vdi_aio_cancel,
};

static VdiAIOCB *vdi_aio_setup(BlockDriverState *bs, int64_t sector_num,
        QEMUIOVector *qiov, int nb_sectors,
        BlockDriverCompletionFunc *cb, void *opaque, int is_write)
{
    VdiAIOCB *acb;

    logout("%p, %" PRId64 ", %p, %d, %p, %p, %d\n",
           bs, sector_num, qiov, nb_sectors, cb, opaque, is_write);

    acb = qemu_aio_get(&vdi_aio_pool, bs, cb, opaque);
    if (acb) {
        acb->hd_aiocb = NULL;
        acb->sector_num = sector_num;
        acb->qiov = qiov;
        if (qiov->niov > 1) {
            acb->buf = qemu_blockalign(bs, qiov->size);
            acb->orig_buf = acb->buf;
            if (is_write) {
                qemu_iovec_to_buffer(qiov, acb->buf);
            }
        } else {
            acb->buf = (uint8_t *)qiov->iov->iov_base;
        }
        acb->nb_sectors = nb_sectors;
        acb->n_sectors = 0;
        acb->bmap_first = VDI_UNALLOCATED;
        acb->bmap_last = VDI_UNALLOCATED;
        acb->block_buffer = NULL;
        acb->header_modified = 0;
    }
    return acb;
}

static int vdi_schedule_bh(QEMUBHFunc *cb, VdiAIOCB *acb)
{
    logout("\n");

    if (acb->bh) {
        return -EIO;
    }

    acb->bh = qemu_bh_new(cb, acb);
    if (!acb->bh) {
        return -EIO;
    }

    qemu_bh_schedule(acb->bh);

    return 0;
}

static void vdi_aio_read_cb(void *opaque, int ret);

static void vdi_aio_read_bh(void *opaque)
{
    VdiAIOCB *acb = opaque;
    logout("\n");
    qemu_bh_delete(acb->bh);
    acb->bh = NULL;
    vdi_aio_read_cb(opaque, 0);
}

static void vdi_aio_read_cb(void *opaque, int ret)
{
    VdiAIOCB *acb = opaque;
    BlockDriverState *bs = acb->common.bs;
    BDRVVdiState *s = bs->opaque;
    uint32_t bmap_entry;
    uint32_t block_index;
    uint32_t sector_in_block;
    uint32_t n_sectors;

    logout("%u sectors read\n", acb->n_sectors);

    acb->hd_aiocb = NULL;

    if (ret < 0) {
        goto done;
    }

    acb->nb_sectors -= acb->n_sectors;

    if (acb->nb_sectors == 0) {
        /* request completed */
        ret = 0;
        goto done;
    }

    acb->sector_num += acb->n_sectors;
    acb->buf += acb->n_sectors * SECTOR_SIZE;

    block_index = acb->sector_num / s->block_sectors;
    sector_in_block = acb->sector_num % s->block_sectors;
    n_sectors = s->block_sectors - sector_in_block;
    if (n_sectors > acb->nb_sectors) {
        n_sectors = acb->nb_sectors;
    }

    logout("will read %u sectors starting at sector %" PRIu64 "\n",
           n_sectors, acb->sector_num);

    /* prepare next AIO request */
    acb->n_sectors = n_sectors;
    bmap_entry = le32_to_cpu(s->bmap[block_index]);
    if (bmap_entry == VDI_UNALLOCATED) {
        /* Block not allocated, return zeros, no need to wait. */
        memset(acb->buf, 0, n_sectors * SECTOR_SIZE);
        ret = vdi_schedule_bh(vdi_aio_read_bh, acb);
        if (ret < 0) {
            goto done;
        }
    } else {
        uint64_t offset = s->header.offset_data / SECTOR_SIZE +
                          (uint64_t)bmap_entry * s->block_sectors +
                          sector_in_block;
        acb->hd_iov.iov_base = (void *)acb->buf;
        acb->hd_iov.iov_len = n_sectors * SECTOR_SIZE;
        qemu_iovec_init_external(&acb->hd_qiov, &acb->hd_iov, 1);
        acb->hd_aiocb = bdrv_aio_readv(s->hd, offset, &acb->hd_qiov,
                                       n_sectors, vdi_aio_read_cb, acb);
        if (acb->hd_aiocb == NULL) {
            goto done;
        }
    }
    return;
done:
    if (acb->qiov->niov > 1) {
        qemu_iovec_from_buffer(acb->qiov, acb->orig_buf, acb->qiov->size);
        qemu_vfree(acb->orig_buf);
    }
    acb->common.cb(acb->common.opaque, ret);
    qemu_aio_release(acb);
}

static BlockDriverAIOCB *vdi_aio_readv(BlockDriverState *bs,
        int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
        BlockDriverCompletionFunc *cb, void *opaque)
{
    VdiAIOCB *acb;
    logout("\n");
    acb = vdi_aio_setup(bs, sector_num, qiov, nb_sectors, cb, opaque, 0);
    if (!acb) {
        return NULL;
    }
    vdi_aio_read_cb(acb, 0);
    return &acb->common;
}

static void vdi_aio_write_cb(void *opaque, int ret)
{
    VdiAIOCB *acb = opaque;
    BlockDriverState *bs = acb->common.bs;
    BDRVVdiState *s = bs->opaque;
    uint32_t bmap_entry;
    uint32_t block_index;
    uint32_t sector_in_block;
    uint32_t n_sectors;

    acb->hd_aiocb = NULL;

    if (ret < 0) {
        goto done;
    }

    acb->nb_sectors -= acb->n_sectors;
    acb->sector_num += acb->n_sectors;
    acb->buf += acb->n_sectors * SECTOR_SIZE;

    if (acb->nb_sectors == 0) {
        logout("finished data write\n");
        acb->n_sectors = 0;
        if (acb->header_modified) {
            VdiHeader *header = acb->block_buffer;
            logout("now writing modified header\n");
            assert(acb->bmap_first != VDI_UNALLOCATED);
            *header = s->header;
            vdi_header_to_le(header);
            acb->header_modified = 0;
            acb->hd_iov.iov_base = acb->block_buffer;
            acb->hd_iov.iov_len = SECTOR_SIZE;
            qemu_iovec_init_external(&acb->hd_qiov, &acb->hd_iov, 1);
            acb->hd_aiocb = bdrv_aio_writev(s->hd, 0, &acb->hd_qiov, 1,
                                            vdi_aio_write_cb, acb);
            if (acb->hd_aiocb == NULL) {
                goto done;
            }
            return;
        } else if (acb->bmap_first != VDI_UNALLOCATED) {
            /* One or more new blocks were allocated. */
            uint64_t offset;
            uint32_t bmap_first;
            uint32_t bmap_last;
            qemu_free(acb->block_buffer);
            acb->block_buffer = NULL;
            bmap_first = acb->bmap_first;
            bmap_last = acb->bmap_last;
            logout("now writing modified block map entry %u...%u\n",
                   bmap_first, bmap_last);
            /* Write modified sectors from block map. */
            bmap_first /= (SECTOR_SIZE / sizeof(uint32_t));
            bmap_last /= (SECTOR_SIZE / sizeof(uint32_t));
            n_sectors = bmap_last - bmap_first + 1;
            offset = s->bmap_sector + bmap_first;
            acb->bmap_first = VDI_UNALLOCATED;
            acb->hd_iov.iov_base = (void *)((uint8_t *)&s->bmap[0] +
                                            bmap_first * SECTOR_SIZE);
            acb->hd_iov.iov_len = n_sectors * SECTOR_SIZE;
            qemu_iovec_init_external(&acb->hd_qiov, &acb->hd_iov, 1);
            logout("will write %u block map sectors starting from entry %u\n",
                   n_sectors, bmap_first);
            acb->hd_aiocb = bdrv_aio_writev(s->hd, offset, &acb->hd_qiov,
                                            n_sectors, vdi_aio_write_cb, acb);
            if (acb->hd_aiocb == NULL) {
                goto done;
            }
            return;
        }
        ret = 0;
        goto done;
    }

    logout("%u sectors written\n", acb->n_sectors);

    block_index = acb->sector_num / s->block_sectors;
    sector_in_block = acb->sector_num % s->block_sectors;
    n_sectors = s->block_sectors - sector_in_block;
    if (n_sectors > acb->nb_sectors) {
        n_sectors = acb->nb_sectors;
    }

    logout("will write %u sectors starting at sector %" PRIu64 "\n",
           n_sectors, acb->sector_num);

    /* prepare next AIO request */
    acb->n_sectors = n_sectors;
    bmap_entry = le32_to_cpu(s->bmap[block_index]);
    if (bmap_entry == VDI_UNALLOCATED) {
        /* Allocate new block and write to it. */
        uint64_t offset;
        uint8_t *block;
        bmap_entry = s->header.blocks_allocated;
        s->bmap[block_index] = cpu_to_le32(bmap_entry);
        s->header.blocks_allocated++;
        offset = s->header.offset_data / SECTOR_SIZE +
                 (uint64_t)bmap_entry * s->block_sectors;
        block = acb->block_buffer;
        if (block == NULL) {
            block = qemu_mallocz(s->block_size);
            acb->block_buffer = block;
            acb->bmap_first = block_index;
            assert(!acb->header_modified);
            acb->header_modified = 1;
        }
        acb->bmap_last = block_index;
        memcpy(block + sector_in_block * SECTOR_SIZE,
               acb->buf, n_sectors * SECTOR_SIZE);
        acb->hd_iov.iov_base = (void *)block;
        acb->hd_iov.iov_len = s->block_size;
        qemu_iovec_init_external(&acb->hd_qiov, &acb->hd_iov, 1);
        acb->hd_aiocb = bdrv_aio_writev(s->hd, offset,
                                        &acb->hd_qiov, s->block_sectors,
                                        vdi_aio_write_cb, acb);
        if (acb->hd_aiocb == NULL) {
            goto done;
        }
    } else {
        uint64_t offset = s->header.offset_data / SECTOR_SIZE +
                          (uint64_t)bmap_entry * s->block_sectors +
                          sector_in_block;
        acb->hd_iov.iov_base = (void *)acb->buf;
        acb->hd_iov.iov_len = n_sectors * SECTOR_SIZE;
        qemu_iovec_init_external(&acb->hd_qiov, &acb->hd_iov, 1);
        acb->hd_aiocb = bdrv_aio_writev(s->hd, offset, &acb->hd_qiov,
                                        n_sectors, vdi_aio_write_cb, acb);
        if (acb->hd_aiocb == NULL) {
            goto done;
        }
    }

    return;

done:
    if (acb->qiov->niov > 1) {
        qemu_vfree(acb->orig_buf);
    }
    acb->common.cb(acb->common.opaque, ret);
    qemu_aio_release(acb);
}

static BlockDriverAIOCB *vdi_aio_writev(BlockDriverState *bs,
        int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
        BlockDriverCompletionFunc *cb, void *opaque)
{
    VdiAIOCB *acb;
    logout("\n");
    acb = vdi_aio_setup(bs, sector_num, qiov, nb_sectors, cb, opaque, 1);
    if (!acb) {
        return NULL;
    }
    vdi_aio_write_cb(acb, 0);
    return &acb->common;
}

static int vdi_create(const char *filename, QEMUOptionParameter *options)
{
    int fd;
    int result = 0;
    uint64_t bytes = 0;
    uint32_t blocks;
    size_t block_size = 1 * MiB;
    uint32_t image_type = VDI_TYPE_DYNAMIC;
    VdiHeader header;
    size_t i;
    size_t bmap_size;
    uint32_t *bmap;

    logout("\n");

    /* Read out options. */
    while (options && options->name) {
        if (!strcmp(options->name, BLOCK_OPT_SIZE)) {
            bytes = options->value.n;
#if defined(CONFIG_VDI_BLOCK_SIZE)
        } else if (!strcmp(options->name, BLOCK_OPT_CLUSTER_SIZE)) {
            if (options->value.n) {
                /* TODO: Additional checks (SECTOR_SIZE * 2^n, ...). */
                block_size = options->value.n;
            }
#endif
#if defined(CONFIG_VDI_STATIC_IMAGE)
        } else if (!strcmp(options->name, BLOCK_OPT_STATIC)) {
            if (options->value.n) {
                image_type = VDI_TYPE_STATIC;
            }
#endif
        }
        options++;
    }

    fd = open(filename, O_WRONLY | O_CREAT | O_TRUNC | O_BINARY | O_LARGEFILE,
              0644);
    if (fd < 0) {
        return -errno;
    }

    blocks = bytes / block_size;
    bmap_size = blocks * sizeof(uint32_t);
    bmap_size = ((bmap_size + SECTOR_SIZE - 1) & ~(SECTOR_SIZE -1));

    memset(&header, 0, sizeof(header));
    pstrcpy(header.text, sizeof(header.text), VDI_TEXT);
    header.signature = VDI_SIGNATURE;
    header.version = VDI_VERSION_1_1;
    header.header_size = 0x180;
    header.image_type = image_type;
    header.offset_bmap = 0x200;
    header.offset_data = 0x200 + bmap_size;
    header.sector_size = SECTOR_SIZE;
    header.disk_size = bytes;
    header.block_size = block_size;
    header.blocks_in_image = blocks;
    if (image_type == VDI_TYPE_STATIC) {
        header.blocks_allocated = blocks;
    }
    uuid_generate(header.uuid_image);
    uuid_generate(header.uuid_last_snap);
    /* There is no need to set header.uuid_link or header.uuid_parent here. */
#if defined(CONFIG_VDI_DEBUG)
    vdi_header_print(&header);
#endif
    vdi_header_to_le(&header);
    if (write(fd, &header, sizeof(header)) < 0) {
        result = -errno;
    }

    bmap = (uint32_t *)qemu_mallocz(bmap_size);
    for (i = 0; i < blocks; i++) {
        if (image_type == VDI_TYPE_STATIC) {
            bmap[i] = i;
        } else {
            bmap[i] = VDI_UNALLOCATED;
        }
    }
    if (write(fd, bmap, bmap_size) < 0) {
        result = -errno;
    }
    qemu_free(bmap);
    if (image_type == VDI_TYPE_STATIC) {
        if (ftruncate(fd, sizeof(header) + bmap_size + blocks * block_size)) {
            result = -errno;
        }
    }

    if (close(fd) < 0) {
        result = -errno;
    }

    return result;
}

static void vdi_close(BlockDriverState *bs)
{
    BDRVVdiState *s = bs->opaque;
    logout("\n");
    bdrv_delete(s->hd);
}

static void vdi_flush(BlockDriverState *bs)
{
    BDRVVdiState *s = bs->opaque;
    logout("\n");
    bdrv_flush(s->hd);
}


static QEMUOptionParameter vdi_create_options[] = {
    {
        .name = BLOCK_OPT_SIZE,
        .type = OPT_SIZE,
        .help = "Virtual disk size"
    },
#if defined(CONFIG_VDI_BLOCK_SIZE)
    {
        .name = BLOCK_OPT_CLUSTER_SIZE,
        .type = OPT_SIZE,
        .help = "VDI cluster (block) size"
    },
#endif
#if defined(CONFIG_VDI_STATIC_IMAGE)
    {
        .name = BLOCK_OPT_STATIC,
        .type = OPT_FLAG,
        .help = "VDI static (pre-allocated) image"
    },
#endif
    /* TODO: An additional option to set UUID values might be useful. */
    { NULL }
};

static BlockDriver bdrv_vdi = {
    .format_name = "vdi",
    .instance_size = sizeof(BDRVVdiState),
    .bdrv_probe = vdi_probe,
    .bdrv_open = vdi_open,
    .bdrv_close = vdi_close,
    .bdrv_create = vdi_create,
    .bdrv_flush = vdi_flush,
    .bdrv_is_allocated = vdi_is_allocated,
    .bdrv_make_empty = vdi_make_empty,

    .bdrv_aio_readv = vdi_aio_readv,
#if defined(CONFIG_VDI_WRITE)
    .bdrv_aio_writev = vdi_aio_writev,
#endif

    .bdrv_get_info = vdi_get_info,

    .create_options = vdi_create_options,
    .bdrv_check = vdi_check,
};

static void bdrv_vdi_init(void)
{
    logout("\n");
    bdrv_register(&bdrv_vdi);
}

block_init(bdrv_vdi_init);