MIPS: Fix tlbwi/tlbwr
[qemu/qemu-JZ.git] / block-qcow2.c
blobb9f168866378bb41618572bfea362c7446b4bef6
1 /*
2 * Block driver for the QCOW version 2 format
4 * Copyright (c) 2004-2006 Fabrice Bellard
6 * Permission is hereby granted, free of charge, to any person obtaining a copy
7 * of this software and associated documentation files (the "Software"), to deal
8 * in the Software without restriction, including without limitation the rights
9 * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
10 * copies of the Software, and to permit persons to whom the Software is
11 * furnished to do so, subject to the following conditions:
13 * The above copyright notice and this permission notice shall be included in
14 * all copies or substantial portions of the Software.
16 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
17 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
18 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
19 * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
20 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
21 * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
22 * THE SOFTWARE.
24 #include "qemu-common.h"
25 #include "block_int.h"
26 #include <zlib.h>
27 #include "aes.h"
28 #include <assert.h>
31 Differences with QCOW:
33 - Support for multiple incremental snapshots.
34 - Memory management by reference counts.
35 - Clusters which have a reference count of one have the bit
36 QCOW_OFLAG_COPIED to optimize write performance.
37 - Size of compressed clusters is stored in sectors to reduce bit usage
38 in the cluster offsets.
39 - Support for storing additional data (such as the VM state) in the
40 snapshots.
41 - If a backing store is used, the cluster size is not constrained
42 (could be backported to QCOW).
43 - L2 tables have always a size of one cluster.
46 //#define DEBUG_ALLOC
47 //#define DEBUG_ALLOC2
49 #define QCOW_MAGIC (('Q' << 24) | ('F' << 16) | ('I' << 8) | 0xfb)
50 #define QCOW_VERSION 2
52 #define QCOW_CRYPT_NONE 0
53 #define QCOW_CRYPT_AES 1
55 #define QCOW_MAX_CRYPT_CLUSTERS 32
57 /* indicate that the refcount of the referenced cluster is exactly one. */
58 #define QCOW_OFLAG_COPIED (1LL << 63)
59 /* indicate that the cluster is compressed (they never have the copied flag) */
60 #define QCOW_OFLAG_COMPRESSED (1LL << 62)
62 #define REFCOUNT_SHIFT 1 /* refcount size is 2 bytes */
64 #ifndef offsetof
65 #define offsetof(type, field) ((size_t) &((type *)0)->field)
66 #endif
68 typedef struct QCowHeader {
69 uint32_t magic;
70 uint32_t version;
71 uint64_t backing_file_offset;
72 uint32_t backing_file_size;
73 uint32_t cluster_bits;
74 uint64_t size; /* in bytes */
75 uint32_t crypt_method;
76 uint32_t l1_size; /* XXX: save number of clusters instead ? */
77 uint64_t l1_table_offset;
78 uint64_t refcount_table_offset;
79 uint32_t refcount_table_clusters;
80 uint32_t nb_snapshots;
81 uint64_t snapshots_offset;
82 } QCowHeader;
84 typedef struct __attribute__((packed)) QCowSnapshotHeader {
85 /* header is 8 byte aligned */
86 uint64_t l1_table_offset;
88 uint32_t l1_size;
89 uint16_t id_str_size;
90 uint16_t name_size;
92 uint32_t date_sec;
93 uint32_t date_nsec;
95 uint64_t vm_clock_nsec;
97 uint32_t vm_state_size;
98 uint32_t extra_data_size; /* for extension */
99 /* extra data follows */
100 /* id_str follows */
101 /* name follows */
102 } QCowSnapshotHeader;
104 #define L2_CACHE_SIZE 16
106 typedef struct QCowSnapshot {
107 uint64_t l1_table_offset;
108 uint32_t l1_size;
109 char *id_str;
110 char *name;
111 uint32_t vm_state_size;
112 uint32_t date_sec;
113 uint32_t date_nsec;
114 uint64_t vm_clock_nsec;
115 } QCowSnapshot;
117 typedef struct BDRVQcowState {
118 BlockDriverState *hd;
119 int cluster_bits;
120 int cluster_size;
121 int cluster_sectors;
122 int l2_bits;
123 int l2_size;
124 int l1_size;
125 int l1_vm_state_index;
126 int csize_shift;
127 int csize_mask;
128 uint64_t cluster_offset_mask;
129 uint64_t l1_table_offset;
130 uint64_t *l1_table;
131 uint64_t *l2_cache;
132 uint64_t l2_cache_offsets[L2_CACHE_SIZE];
133 uint32_t l2_cache_counts[L2_CACHE_SIZE];
134 uint8_t *cluster_cache;
135 uint8_t *cluster_data;
136 uint64_t cluster_cache_offset;
138 uint64_t *refcount_table;
139 uint64_t refcount_table_offset;
140 uint32_t refcount_table_size;
141 uint64_t refcount_block_cache_offset;
142 uint16_t *refcount_block_cache;
143 int64_t free_cluster_index;
144 int64_t free_byte_offset;
146 uint32_t crypt_method; /* current crypt method, 0 if no key yet */
147 uint32_t crypt_method_header;
148 AES_KEY aes_encrypt_key;
149 AES_KEY aes_decrypt_key;
150 uint64_t snapshots_offset;
151 int snapshots_size;
152 int nb_snapshots;
153 QCowSnapshot *snapshots;
154 } BDRVQcowState;
156 static int decompress_cluster(BDRVQcowState *s, uint64_t cluster_offset);
157 static int qcow_read(BlockDriverState *bs, int64_t sector_num,
158 uint8_t *buf, int nb_sectors);
159 static int qcow_read_snapshots(BlockDriverState *bs);
160 static void qcow_free_snapshots(BlockDriverState *bs);
161 static int refcount_init(BlockDriverState *bs);
162 static void refcount_close(BlockDriverState *bs);
163 static int get_refcount(BlockDriverState *bs, int64_t cluster_index);
164 static int update_cluster_refcount(BlockDriverState *bs,
165 int64_t cluster_index,
166 int addend);
167 static void update_refcount(BlockDriverState *bs,
168 int64_t offset, int64_t length,
169 int addend);
170 static int64_t alloc_clusters(BlockDriverState *bs, int64_t size);
171 static int64_t alloc_bytes(BlockDriverState *bs, int size);
172 static void free_clusters(BlockDriverState *bs,
173 int64_t offset, int64_t size);
174 #ifdef DEBUG_ALLOC
175 static void check_refcounts(BlockDriverState *bs);
176 #endif
178 static int qcow_probe(const uint8_t *buf, int buf_size, const char *filename)
180 const QCowHeader *cow_header = (const void *)buf;
182 if (buf_size >= sizeof(QCowHeader) &&
183 be32_to_cpu(cow_header->magic) == QCOW_MAGIC &&
184 be32_to_cpu(cow_header->version) == QCOW_VERSION)
185 return 100;
186 else
187 return 0;
190 static int qcow_open(BlockDriverState *bs, const char *filename, int flags)
192 BDRVQcowState *s = bs->opaque;
193 int len, i, shift, ret;
194 QCowHeader header;
196 ret = bdrv_file_open(&s->hd, filename, flags);
197 if (ret < 0)
198 return ret;
199 if (bdrv_pread(s->hd, 0, &header, sizeof(header)) != sizeof(header))
200 goto fail;
201 be32_to_cpus(&header.magic);
202 be32_to_cpus(&header.version);
203 be64_to_cpus(&header.backing_file_offset);
204 be32_to_cpus(&header.backing_file_size);
205 be64_to_cpus(&header.size);
206 be32_to_cpus(&header.cluster_bits);
207 be32_to_cpus(&header.crypt_method);
208 be64_to_cpus(&header.l1_table_offset);
209 be32_to_cpus(&header.l1_size);
210 be64_to_cpus(&header.refcount_table_offset);
211 be32_to_cpus(&header.refcount_table_clusters);
212 be64_to_cpus(&header.snapshots_offset);
213 be32_to_cpus(&header.nb_snapshots);
215 if (header.magic != QCOW_MAGIC || header.version != QCOW_VERSION)
216 goto fail;
217 if (header.size <= 1 ||
218 header.cluster_bits < 9 ||
219 header.cluster_bits > 16)
220 goto fail;
221 if (header.crypt_method > QCOW_CRYPT_AES)
222 goto fail;
223 s->crypt_method_header = header.crypt_method;
224 if (s->crypt_method_header)
225 bs->encrypted = 1;
226 s->cluster_bits = header.cluster_bits;
227 s->cluster_size = 1 << s->cluster_bits;
228 s->cluster_sectors = 1 << (s->cluster_bits - 9);
229 s->l2_bits = s->cluster_bits - 3; /* L2 is always one cluster */
230 s->l2_size = 1 << s->l2_bits;
231 bs->total_sectors = header.size / 512;
232 s->csize_shift = (62 - (s->cluster_bits - 8));
233 s->csize_mask = (1 << (s->cluster_bits - 8)) - 1;
234 s->cluster_offset_mask = (1LL << s->csize_shift) - 1;
235 s->refcount_table_offset = header.refcount_table_offset;
236 s->refcount_table_size =
237 header.refcount_table_clusters << (s->cluster_bits - 3);
239 s->snapshots_offset = header.snapshots_offset;
240 s->nb_snapshots = header.nb_snapshots;
242 /* read the level 1 table */
243 s->l1_size = header.l1_size;
244 shift = s->cluster_bits + s->l2_bits;
245 s->l1_vm_state_index = (header.size + (1LL << shift) - 1) >> shift;
246 /* the L1 table must contain at least enough entries to put
247 header.size bytes */
248 if (s->l1_size < s->l1_vm_state_index)
249 goto fail;
250 s->l1_table_offset = header.l1_table_offset;
251 s->l1_table = qemu_malloc(s->l1_size * sizeof(uint64_t));
252 if (!s->l1_table)
253 goto fail;
254 if (bdrv_pread(s->hd, s->l1_table_offset, s->l1_table, s->l1_size * sizeof(uint64_t)) !=
255 s->l1_size * sizeof(uint64_t))
256 goto fail;
257 for(i = 0;i < s->l1_size; i++) {
258 be64_to_cpus(&s->l1_table[i]);
260 /* alloc L2 cache */
261 s->l2_cache = qemu_malloc(s->l2_size * L2_CACHE_SIZE * sizeof(uint64_t));
262 if (!s->l2_cache)
263 goto fail;
264 s->cluster_cache = qemu_malloc(s->cluster_size);
265 if (!s->cluster_cache)
266 goto fail;
267 /* one more sector for decompressed data alignment */
268 s->cluster_data = qemu_malloc(QCOW_MAX_CRYPT_CLUSTERS * s->cluster_size
269 + 512);
270 if (!s->cluster_data)
271 goto fail;
272 s->cluster_cache_offset = -1;
274 if (refcount_init(bs) < 0)
275 goto fail;
277 /* read the backing file name */
278 if (header.backing_file_offset != 0) {
279 len = header.backing_file_size;
280 if (len > 1023)
281 len = 1023;
282 if (bdrv_pread(s->hd, header.backing_file_offset, bs->backing_file, len) != len)
283 goto fail;
284 bs->backing_file[len] = '\0';
286 if (qcow_read_snapshots(bs) < 0)
287 goto fail;
289 #ifdef DEBUG_ALLOC
290 check_refcounts(bs);
291 #endif
292 return 0;
294 fail:
295 qcow_free_snapshots(bs);
296 refcount_close(bs);
297 qemu_free(s->l1_table);
298 qemu_free(s->l2_cache);
299 qemu_free(s->cluster_cache);
300 qemu_free(s->cluster_data);
301 bdrv_delete(s->hd);
302 return -1;
305 static int qcow_set_key(BlockDriverState *bs, const char *key)
307 BDRVQcowState *s = bs->opaque;
308 uint8_t keybuf[16];
309 int len, i;
311 memset(keybuf, 0, 16);
312 len = strlen(key);
313 if (len > 16)
314 len = 16;
315 /* XXX: we could compress the chars to 7 bits to increase
316 entropy */
317 for(i = 0;i < len;i++) {
318 keybuf[i] = key[i];
320 s->crypt_method = s->crypt_method_header;
322 if (AES_set_encrypt_key(keybuf, 128, &s->aes_encrypt_key) != 0)
323 return -1;
324 if (AES_set_decrypt_key(keybuf, 128, &s->aes_decrypt_key) != 0)
325 return -1;
326 #if 0
327 /* test */
329 uint8_t in[16];
330 uint8_t out[16];
331 uint8_t tmp[16];
332 for(i=0;i<16;i++)
333 in[i] = i;
334 AES_encrypt(in, tmp, &s->aes_encrypt_key);
335 AES_decrypt(tmp, out, &s->aes_decrypt_key);
336 for(i = 0; i < 16; i++)
337 printf(" %02x", tmp[i]);
338 printf("\n");
339 for(i = 0; i < 16; i++)
340 printf(" %02x", out[i]);
341 printf("\n");
343 #endif
344 return 0;
347 /* The crypt function is compatible with the linux cryptoloop
348 algorithm for < 4 GB images. NOTE: out_buf == in_buf is
349 supported */
350 static void encrypt_sectors(BDRVQcowState *s, int64_t sector_num,
351 uint8_t *out_buf, const uint8_t *in_buf,
352 int nb_sectors, int enc,
353 const AES_KEY *key)
355 union {
356 uint64_t ll[2];
357 uint8_t b[16];
358 } ivec;
359 int i;
361 for(i = 0; i < nb_sectors; i++) {
362 ivec.ll[0] = cpu_to_le64(sector_num);
363 ivec.ll[1] = 0;
364 AES_cbc_encrypt(in_buf, out_buf, 512, key,
365 ivec.b, enc);
366 sector_num++;
367 in_buf += 512;
368 out_buf += 512;
372 static int copy_sectors(BlockDriverState *bs, uint64_t start_sect,
373 uint64_t cluster_offset, int n_start, int n_end)
375 BDRVQcowState *s = bs->opaque;
376 int n, ret;
378 n = n_end - n_start;
379 if (n <= 0)
380 return 0;
381 ret = qcow_read(bs, start_sect + n_start, s->cluster_data, n);
382 if (ret < 0)
383 return ret;
384 if (s->crypt_method) {
385 encrypt_sectors(s, start_sect + n_start,
386 s->cluster_data,
387 s->cluster_data, n, 1,
388 &s->aes_encrypt_key);
390 ret = bdrv_write(s->hd, (cluster_offset >> 9) + n_start,
391 s->cluster_data, n);
392 if (ret < 0)
393 return ret;
394 return 0;
397 static void l2_cache_reset(BlockDriverState *bs)
399 BDRVQcowState *s = bs->opaque;
401 memset(s->l2_cache, 0, s->l2_size * L2_CACHE_SIZE * sizeof(uint64_t));
402 memset(s->l2_cache_offsets, 0, L2_CACHE_SIZE * sizeof(uint64_t));
403 memset(s->l2_cache_counts, 0, L2_CACHE_SIZE * sizeof(uint32_t));
406 static inline int l2_cache_new_entry(BlockDriverState *bs)
408 BDRVQcowState *s = bs->opaque;
409 uint32_t min_count;
410 int min_index, i;
412 /* find a new entry in the least used one */
413 min_index = 0;
414 min_count = 0xffffffff;
415 for(i = 0; i < L2_CACHE_SIZE; i++) {
416 if (s->l2_cache_counts[i] < min_count) {
417 min_count = s->l2_cache_counts[i];
418 min_index = i;
421 return min_index;
424 static int64_t align_offset(int64_t offset, int n)
426 offset = (offset + n - 1) & ~(n - 1);
427 return offset;
430 static int grow_l1_table(BlockDriverState *bs, int min_size)
432 BDRVQcowState *s = bs->opaque;
433 int new_l1_size, new_l1_size2, ret, i;
434 uint64_t *new_l1_table;
435 uint64_t new_l1_table_offset;
436 uint64_t data64;
437 uint32_t data32;
439 new_l1_size = s->l1_size;
440 if (min_size <= new_l1_size)
441 return 0;
442 while (min_size > new_l1_size) {
443 new_l1_size = (new_l1_size * 3 + 1) / 2;
445 #ifdef DEBUG_ALLOC2
446 printf("grow l1_table from %d to %d\n", s->l1_size, new_l1_size);
447 #endif
449 new_l1_size2 = sizeof(uint64_t) * new_l1_size;
450 new_l1_table = qemu_mallocz(new_l1_size2);
451 if (!new_l1_table)
452 return -ENOMEM;
453 memcpy(new_l1_table, s->l1_table, s->l1_size * sizeof(uint64_t));
455 /* write new table (align to cluster) */
456 new_l1_table_offset = alloc_clusters(bs, new_l1_size2);
458 for(i = 0; i < s->l1_size; i++)
459 new_l1_table[i] = cpu_to_be64(new_l1_table[i]);
460 ret = bdrv_pwrite(s->hd, new_l1_table_offset, new_l1_table, new_l1_size2);
461 if (ret != new_l1_size2)
462 goto fail;
463 for(i = 0; i < s->l1_size; i++)
464 new_l1_table[i] = be64_to_cpu(new_l1_table[i]);
466 /* set new table */
467 data64 = cpu_to_be64(new_l1_table_offset);
468 if (bdrv_pwrite(s->hd, offsetof(QCowHeader, l1_table_offset),
469 &data64, sizeof(data64)) != sizeof(data64))
470 goto fail;
471 data32 = cpu_to_be32(new_l1_size);
472 if (bdrv_pwrite(s->hd, offsetof(QCowHeader, l1_size),
473 &data32, sizeof(data32)) != sizeof(data32))
474 goto fail;
475 qemu_free(s->l1_table);
476 free_clusters(bs, s->l1_table_offset, s->l1_size * sizeof(uint64_t));
477 s->l1_table_offset = new_l1_table_offset;
478 s->l1_table = new_l1_table;
479 s->l1_size = new_l1_size;
480 return 0;
481 fail:
482 qemu_free(s->l1_table);
483 return -EIO;
487 * seek_l2_table
489 * seek l2_offset in the l2_cache table
490 * if not found, return NULL,
491 * if found,
492 * increments the l2 cache hit count of the entry,
493 * if counter overflow, divide by two all counters
494 * return the pointer to the l2 cache entry
498 static uint64_t *seek_l2_table(BDRVQcowState *s, uint64_t l2_offset)
500 int i, j;
502 for(i = 0; i < L2_CACHE_SIZE; i++) {
503 if (l2_offset == s->l2_cache_offsets[i]) {
504 /* increment the hit count */
505 if (++s->l2_cache_counts[i] == 0xffffffff) {
506 for(j = 0; j < L2_CACHE_SIZE; j++) {
507 s->l2_cache_counts[j] >>= 1;
510 return s->l2_cache + (i << s->l2_bits);
513 return NULL;
517 * l2_load
519 * Loads a L2 table into memory. If the table is in the cache, the cache
520 * is used; otherwise the L2 table is loaded from the image file.
522 * Returns a pointer to the L2 table on success, or NULL if the read from
523 * the image file failed.
526 static uint64_t *l2_load(BlockDriverState *bs, uint64_t l2_offset)
528 BDRVQcowState *s = bs->opaque;
529 int min_index;
530 uint64_t *l2_table;
532 /* seek if the table for the given offset is in the cache */
534 l2_table = seek_l2_table(s, l2_offset);
535 if (l2_table != NULL)
536 return l2_table;
538 /* not found: load a new entry in the least used one */
540 min_index = l2_cache_new_entry(bs);
541 l2_table = s->l2_cache + (min_index << s->l2_bits);
542 if (bdrv_pread(s->hd, l2_offset, l2_table, s->l2_size * sizeof(uint64_t)) !=
543 s->l2_size * sizeof(uint64_t))
544 return NULL;
545 s->l2_cache_offsets[min_index] = l2_offset;
546 s->l2_cache_counts[min_index] = 1;
548 return l2_table;
552 * l2_allocate
554 * Allocate a new l2 entry in the file. If l1_index points to an already
555 * used entry in the L2 table (i.e. we are doing a copy on write for the L2
556 * table) copy the contents of the old L2 table into the newly allocated one.
557 * Otherwise the new table is initialized with zeros.
561 static uint64_t *l2_allocate(BlockDriverState *bs, int l1_index)
563 BDRVQcowState *s = bs->opaque;
564 int min_index;
565 uint64_t old_l2_offset, tmp;
566 uint64_t *l2_table, l2_offset;
568 old_l2_offset = s->l1_table[l1_index];
570 /* allocate a new l2 entry */
572 l2_offset = alloc_clusters(bs, s->l2_size * sizeof(uint64_t));
574 /* update the L1 entry */
576 s->l1_table[l1_index] = l2_offset | QCOW_OFLAG_COPIED;
578 tmp = cpu_to_be64(l2_offset | QCOW_OFLAG_COPIED);
579 if (bdrv_pwrite(s->hd, s->l1_table_offset + l1_index * sizeof(tmp),
580 &tmp, sizeof(tmp)) != sizeof(tmp))
581 return NULL;
583 /* allocate a new entry in the l2 cache */
585 min_index = l2_cache_new_entry(bs);
586 l2_table = s->l2_cache + (min_index << s->l2_bits);
588 if (old_l2_offset == 0) {
589 /* if there was no old l2 table, clear the new table */
590 memset(l2_table, 0, s->l2_size * sizeof(uint64_t));
591 } else {
592 /* if there was an old l2 table, read it from the disk */
593 if (bdrv_pread(s->hd, old_l2_offset,
594 l2_table, s->l2_size * sizeof(uint64_t)) !=
595 s->l2_size * sizeof(uint64_t))
596 return NULL;
598 /* write the l2 table to the file */
599 if (bdrv_pwrite(s->hd, l2_offset,
600 l2_table, s->l2_size * sizeof(uint64_t)) !=
601 s->l2_size * sizeof(uint64_t))
602 return NULL;
604 /* update the l2 cache entry */
606 s->l2_cache_offsets[min_index] = l2_offset;
607 s->l2_cache_counts[min_index] = 1;
609 return l2_table;
613 * get_cluster_offset
615 * For a given offset of the disk image, return cluster offset in
616 * qcow2 file.
618 * on entry, *num is the number of contiguous clusters we'd like to
619 * access following offset.
621 * on exit, *num is the number of contiguous clusters we can read.
623 * Return 1, if the offset is found
624 * Return 0, otherwise.
628 static uint64_t get_cluster_offset(BlockDriverState *bs,
629 uint64_t offset, int *num)
631 BDRVQcowState *s = bs->opaque;
632 int l1_index, l2_index;
633 uint64_t l2_offset, *l2_table, cluster_offset, next;
634 int l1_bits;
635 int index_in_cluster, nb_available, nb_needed;
637 index_in_cluster = (offset >> 9) & (s->cluster_sectors - 1);
638 nb_needed = *num + index_in_cluster;
640 l1_bits = s->l2_bits + s->cluster_bits;
642 /* compute how many bytes there are between the offset and
643 * and the end of the l1 entry
646 nb_available = (1 << l1_bits) - (offset & ((1 << l1_bits) - 1));
648 /* compute the number of available sectors */
650 nb_available = (nb_available >> 9) + index_in_cluster;
652 cluster_offset = 0;
654 /* seek the the l2 offset in the l1 table */
656 l1_index = offset >> l1_bits;
657 if (l1_index >= s->l1_size)
658 goto out;
660 l2_offset = s->l1_table[l1_index];
662 /* seek the l2 table of the given l2 offset */
664 if (!l2_offset)
665 goto out;
667 /* load the l2 table in memory */
669 l2_offset &= ~QCOW_OFLAG_COPIED;
670 l2_table = l2_load(bs, l2_offset);
671 if (l2_table == NULL)
672 return 0;
674 /* find the cluster offset for the given disk offset */
676 l2_index = (offset >> s->cluster_bits) & (s->l2_size - 1);
677 cluster_offset = be64_to_cpu(l2_table[l2_index]);
678 nb_available = s->cluster_sectors;
679 l2_index++;
681 if (!cluster_offset) {
683 /* how many empty clusters ? */
685 while (nb_available < nb_needed && !l2_table[l2_index]) {
686 l2_index++;
687 nb_available += s->cluster_sectors;
689 } else {
691 /* how many allocated clusters ? */
693 cluster_offset &= ~QCOW_OFLAG_COPIED;
694 while (nb_available < nb_needed) {
695 next = be64_to_cpu(l2_table[l2_index]) & ~QCOW_OFLAG_COPIED;
696 if (next != cluster_offset + (nb_available << 9))
697 break;
698 l2_index++;
699 nb_available += s->cluster_sectors;
703 out:
704 if (nb_available > nb_needed)
705 nb_available = nb_needed;
707 *num = nb_available - index_in_cluster;
709 return cluster_offset;
713 * free_any_clusters
715 * free clusters according to its type: compressed or not
719 static void free_any_clusters(BlockDriverState *bs,
720 uint64_t cluster_offset, int nb_clusters)
722 BDRVQcowState *s = bs->opaque;
724 /* free the cluster */
726 if (cluster_offset & QCOW_OFLAG_COMPRESSED) {
727 int nb_csectors;
728 nb_csectors = ((cluster_offset >> s->csize_shift) &
729 s->csize_mask) + 1;
730 free_clusters(bs, (cluster_offset & s->cluster_offset_mask) & ~511,
731 nb_csectors * 512);
732 return;
735 free_clusters(bs, cluster_offset, nb_clusters << s->cluster_bits);
737 return;
741 * get_cluster_table
743 * for a given disk offset, load (and allocate if needed)
744 * the l2 table.
746 * the l2 table offset in the qcow2 file and the cluster index
747 * in the l2 table are given to the caller.
751 static int get_cluster_table(BlockDriverState *bs, uint64_t offset,
752 uint64_t **new_l2_table,
753 uint64_t *new_l2_offset,
754 int *new_l2_index)
756 BDRVQcowState *s = bs->opaque;
757 int l1_index, l2_index, ret;
758 uint64_t l2_offset, *l2_table;
760 /* seek the the l2 offset in the l1 table */
762 l1_index = offset >> (s->l2_bits + s->cluster_bits);
763 if (l1_index >= s->l1_size) {
764 ret = grow_l1_table(bs, l1_index + 1);
765 if (ret < 0)
766 return 0;
768 l2_offset = s->l1_table[l1_index];
770 /* seek the l2 table of the given l2 offset */
772 if (l2_offset & QCOW_OFLAG_COPIED) {
773 /* load the l2 table in memory */
774 l2_offset &= ~QCOW_OFLAG_COPIED;
775 l2_table = l2_load(bs, l2_offset);
776 if (l2_table == NULL)
777 return 0;
778 } else {
779 if (l2_offset)
780 free_clusters(bs, l2_offset, s->l2_size * sizeof(uint64_t));
781 l2_table = l2_allocate(bs, l1_index);
782 if (l2_table == NULL)
783 return 0;
784 l2_offset = s->l1_table[l1_index] & ~QCOW_OFLAG_COPIED;
787 /* find the cluster offset for the given disk offset */
789 l2_index = (offset >> s->cluster_bits) & (s->l2_size - 1);
791 *new_l2_table = l2_table;
792 *new_l2_offset = l2_offset;
793 *new_l2_index = l2_index;
795 return 1;
799 * alloc_compressed_cluster_offset
801 * For a given offset of the disk image, return cluster offset in
802 * qcow2 file.
804 * If the offset is not found, allocate a new compressed cluster.
806 * Return the cluster offset if successful,
807 * Return 0, otherwise.
811 static uint64_t alloc_compressed_cluster_offset(BlockDriverState *bs,
812 uint64_t offset,
813 int compressed_size)
815 BDRVQcowState *s = bs->opaque;
816 int l2_index, ret;
817 uint64_t l2_offset, *l2_table, cluster_offset;
818 int nb_csectors;
820 ret = get_cluster_table(bs, offset, &l2_table, &l2_offset, &l2_index);
821 if (ret == 0)
822 return 0;
824 cluster_offset = be64_to_cpu(l2_table[l2_index]);
825 if (cluster_offset & QCOW_OFLAG_COPIED)
826 return cluster_offset & ~QCOW_OFLAG_COPIED;
828 if (cluster_offset)
829 free_any_clusters(bs, cluster_offset, 1);
831 cluster_offset = alloc_bytes(bs, compressed_size);
832 nb_csectors = ((cluster_offset + compressed_size - 1) >> 9) -
833 (cluster_offset >> 9);
835 cluster_offset |= QCOW_OFLAG_COMPRESSED |
836 ((uint64_t)nb_csectors << s->csize_shift);
838 /* update L2 table */
840 /* compressed clusters never have the copied flag */
842 l2_table[l2_index] = cpu_to_be64(cluster_offset);
843 if (bdrv_pwrite(s->hd,
844 l2_offset + l2_index * sizeof(uint64_t),
845 l2_table + l2_index,
846 sizeof(uint64_t)) != sizeof(uint64_t))
847 return 0;
849 return cluster_offset;
853 * alloc_cluster_offset
855 * For a given offset of the disk image, return cluster offset in
856 * qcow2 file.
858 * If the offset is not found, allocate a new cluster.
860 * Return the cluster offset if successful,
861 * Return 0, otherwise.
865 static uint64_t alloc_cluster_offset(BlockDriverState *bs,
866 uint64_t offset,
867 int n_start, int n_end,
868 int *num)
870 BDRVQcowState *s = bs->opaque;
871 int l2_index, ret;
872 uint64_t l2_offset, *l2_table, cluster_offset;
873 int nb_available, nb_clusters, i, j;
874 uint64_t start_sect, current;
876 ret = get_cluster_table(bs, offset, &l2_table, &l2_offset, &l2_index);
877 if (ret == 0)
878 return 0;
880 nb_clusters = ((n_end << 9) + s->cluster_size - 1) >>
881 s->cluster_bits;
882 if (nb_clusters > s->l2_size - l2_index)
883 nb_clusters = s->l2_size - l2_index;
885 cluster_offset = be64_to_cpu(l2_table[l2_index]);
887 /* We keep all QCOW_OFLAG_COPIED clusters */
889 if (cluster_offset & QCOW_OFLAG_COPIED) {
891 for (i = 1; i < nb_clusters; i++) {
892 current = be64_to_cpu(l2_table[l2_index + i]);
893 if (cluster_offset + (i << s->cluster_bits) != current)
894 break;
896 nb_clusters = i;
898 nb_available = nb_clusters << (s->cluster_bits - 9);
899 if (nb_available > n_end)
900 nb_available = n_end;
902 cluster_offset &= ~QCOW_OFLAG_COPIED;
904 goto out;
907 /* for the moment, multiple compressed clusters are not managed */
909 if (cluster_offset & QCOW_OFLAG_COMPRESSED)
910 nb_clusters = 1;
912 /* how many available clusters ? */
914 i = 0;
915 while (i < nb_clusters) {
917 i++;
919 if (!cluster_offset) {
921 /* how many free clusters ? */
923 while (i < nb_clusters) {
924 cluster_offset = l2_table[l2_index + i];
925 if (cluster_offset != 0)
926 break;
927 i++;
930 if ((cluster_offset & QCOW_OFLAG_COPIED) ||
931 (cluster_offset & QCOW_OFLAG_COMPRESSED))
932 break;
934 } else {
936 /* how many contiguous clusters ? */
938 j = 1;
939 current = 0;
940 while (i < nb_clusters) {
941 current = be64_to_cpu(l2_table[l2_index + i]);
942 if (cluster_offset + (j << s->cluster_bits) != current)
943 break;
945 i++;
946 j++;
949 free_any_clusters(bs, cluster_offset, j);
950 if (current)
951 break;
952 cluster_offset = current;
955 nb_clusters = i;
957 /* allocate a new cluster */
959 cluster_offset = alloc_clusters(bs, nb_clusters * s->cluster_size);
961 /* we must initialize the cluster content which won't be
962 written */
964 nb_available = nb_clusters << (s->cluster_bits - 9);
965 if (nb_available > n_end)
966 nb_available = n_end;
968 /* copy content of unmodified sectors */
970 start_sect = (offset & ~(s->cluster_size - 1)) >> 9;
971 if (n_start) {
972 ret = copy_sectors(bs, start_sect, cluster_offset, 0, n_start);
973 if (ret < 0)
974 return 0;
977 if (nb_available & (s->cluster_sectors - 1)) {
978 uint64_t end = nb_available & ~(uint64_t)(s->cluster_sectors - 1);
979 ret = copy_sectors(bs, start_sect + end,
980 cluster_offset + (end << 9),
981 nb_available - end,
982 s->cluster_sectors);
983 if (ret < 0)
984 return 0;
987 /* update L2 table */
989 for (i = 0; i < nb_clusters; i++)
990 l2_table[l2_index + i] = cpu_to_be64((cluster_offset +
991 (i << s->cluster_bits)) |
992 QCOW_OFLAG_COPIED);
994 if (bdrv_pwrite(s->hd,
995 l2_offset + l2_index * sizeof(uint64_t),
996 l2_table + l2_index,
997 nb_clusters * sizeof(uint64_t)) !=
998 nb_clusters * sizeof(uint64_t))
999 return 0;
1001 out:
1002 *num = nb_available - n_start;
1004 return cluster_offset;
1007 static int qcow_is_allocated(BlockDriverState *bs, int64_t sector_num,
1008 int nb_sectors, int *pnum)
1010 uint64_t cluster_offset;
1012 *pnum = nb_sectors;
1013 cluster_offset = get_cluster_offset(bs, sector_num << 9, pnum);
1015 return (cluster_offset != 0);
1018 static int decompress_buffer(uint8_t *out_buf, int out_buf_size,
1019 const uint8_t *buf, int buf_size)
1021 z_stream strm1, *strm = &strm1;
1022 int ret, out_len;
1024 memset(strm, 0, sizeof(*strm));
1026 strm->next_in = (uint8_t *)buf;
1027 strm->avail_in = buf_size;
1028 strm->next_out = out_buf;
1029 strm->avail_out = out_buf_size;
1031 ret = inflateInit2(strm, -12);
1032 if (ret != Z_OK)
1033 return -1;
1034 ret = inflate(strm, Z_FINISH);
1035 out_len = strm->next_out - out_buf;
1036 if ((ret != Z_STREAM_END && ret != Z_BUF_ERROR) ||
1037 out_len != out_buf_size) {
1038 inflateEnd(strm);
1039 return -1;
1041 inflateEnd(strm);
1042 return 0;
1045 static int decompress_cluster(BDRVQcowState *s, uint64_t cluster_offset)
1047 int ret, csize, nb_csectors, sector_offset;
1048 uint64_t coffset;
1050 coffset = cluster_offset & s->cluster_offset_mask;
1051 if (s->cluster_cache_offset != coffset) {
1052 nb_csectors = ((cluster_offset >> s->csize_shift) & s->csize_mask) + 1;
1053 sector_offset = coffset & 511;
1054 csize = nb_csectors * 512 - sector_offset;
1055 ret = bdrv_read(s->hd, coffset >> 9, s->cluster_data, nb_csectors);
1056 if (ret < 0) {
1057 return -1;
1059 if (decompress_buffer(s->cluster_cache, s->cluster_size,
1060 s->cluster_data + sector_offset, csize) < 0) {
1061 return -1;
1063 s->cluster_cache_offset = coffset;
1065 return 0;
1068 /* handle reading after the end of the backing file */
1069 static int backing_read1(BlockDriverState *bs,
1070 int64_t sector_num, uint8_t *buf, int nb_sectors)
1072 int n1;
1073 if ((sector_num + nb_sectors) <= bs->total_sectors)
1074 return nb_sectors;
1075 if (sector_num >= bs->total_sectors)
1076 n1 = 0;
1077 else
1078 n1 = bs->total_sectors - sector_num;
1079 memset(buf + n1 * 512, 0, 512 * (nb_sectors - n1));
1080 return n1;
1083 static int qcow_read(BlockDriverState *bs, int64_t sector_num,
1084 uint8_t *buf, int nb_sectors)
1086 BDRVQcowState *s = bs->opaque;
1087 int ret, index_in_cluster, n, n1;
1088 uint64_t cluster_offset;
1090 while (nb_sectors > 0) {
1091 n = nb_sectors;
1092 cluster_offset = get_cluster_offset(bs, sector_num << 9, &n);
1093 index_in_cluster = sector_num & (s->cluster_sectors - 1);
1094 if (!cluster_offset) {
1095 if (bs->backing_hd) {
1096 /* read from the base image */
1097 n1 = backing_read1(bs->backing_hd, sector_num, buf, n);
1098 if (n1 > 0) {
1099 ret = bdrv_read(bs->backing_hd, sector_num, buf, n1);
1100 if (ret < 0)
1101 return -1;
1103 } else {
1104 memset(buf, 0, 512 * n);
1106 } else if (cluster_offset & QCOW_OFLAG_COMPRESSED) {
1107 if (decompress_cluster(s, cluster_offset) < 0)
1108 return -1;
1109 memcpy(buf, s->cluster_cache + index_in_cluster * 512, 512 * n);
1110 } else {
1111 ret = bdrv_pread(s->hd, cluster_offset + index_in_cluster * 512, buf, n * 512);
1112 if (ret != n * 512)
1113 return -1;
1114 if (s->crypt_method) {
1115 encrypt_sectors(s, sector_num, buf, buf, n, 0,
1116 &s->aes_decrypt_key);
1119 nb_sectors -= n;
1120 sector_num += n;
1121 buf += n * 512;
1123 return 0;
1126 static int qcow_write(BlockDriverState *bs, int64_t sector_num,
1127 const uint8_t *buf, int nb_sectors)
1129 BDRVQcowState *s = bs->opaque;
1130 int ret, index_in_cluster, n;
1131 uint64_t cluster_offset;
1132 int n_end;
1134 while (nb_sectors > 0) {
1135 index_in_cluster = sector_num & (s->cluster_sectors - 1);
1136 n_end = index_in_cluster + nb_sectors;
1137 if (s->crypt_method &&
1138 n_end > QCOW_MAX_CRYPT_CLUSTERS * s->cluster_sectors)
1139 n_end = QCOW_MAX_CRYPT_CLUSTERS * s->cluster_sectors;
1140 cluster_offset = alloc_cluster_offset(bs, sector_num << 9,
1141 index_in_cluster,
1142 n_end, &n);
1143 if (!cluster_offset)
1144 return -1;
1145 if (s->crypt_method) {
1146 encrypt_sectors(s, sector_num, s->cluster_data, buf, n, 1,
1147 &s->aes_encrypt_key);
1148 ret = bdrv_pwrite(s->hd, cluster_offset + index_in_cluster * 512,
1149 s->cluster_data, n * 512);
1150 } else {
1151 ret = bdrv_pwrite(s->hd, cluster_offset + index_in_cluster * 512, buf, n * 512);
1153 if (ret != n * 512)
1154 return -1;
1155 nb_sectors -= n;
1156 sector_num += n;
1157 buf += n * 512;
1159 s->cluster_cache_offset = -1; /* disable compressed cache */
1160 return 0;
1163 typedef struct QCowAIOCB {
1164 BlockDriverAIOCB common;
1165 int64_t sector_num;
1166 uint8_t *buf;
1167 int nb_sectors;
1168 int n;
1169 uint64_t cluster_offset;
1170 uint8_t *cluster_data;
1171 BlockDriverAIOCB *hd_aiocb;
1172 } QCowAIOCB;
1174 static void qcow_aio_read_cb(void *opaque, int ret)
1176 QCowAIOCB *acb = opaque;
1177 BlockDriverState *bs = acb->common.bs;
1178 BDRVQcowState *s = bs->opaque;
1179 int index_in_cluster, n1;
1181 acb->hd_aiocb = NULL;
1182 if (ret < 0) {
1183 fail:
1184 acb->common.cb(acb->common.opaque, ret);
1185 qemu_aio_release(acb);
1186 return;
1189 redo:
1190 /* post process the read buffer */
1191 if (!acb->cluster_offset) {
1192 /* nothing to do */
1193 } else if (acb->cluster_offset & QCOW_OFLAG_COMPRESSED) {
1194 /* nothing to do */
1195 } else {
1196 if (s->crypt_method) {
1197 encrypt_sectors(s, acb->sector_num, acb->buf, acb->buf,
1198 acb->n, 0,
1199 &s->aes_decrypt_key);
1203 acb->nb_sectors -= acb->n;
1204 acb->sector_num += acb->n;
1205 acb->buf += acb->n * 512;
1207 if (acb->nb_sectors == 0) {
1208 /* request completed */
1209 acb->common.cb(acb->common.opaque, 0);
1210 qemu_aio_release(acb);
1211 return;
1214 /* prepare next AIO request */
1215 acb->n = acb->nb_sectors;
1216 acb->cluster_offset = get_cluster_offset(bs, acb->sector_num << 9, &acb->n);
1217 index_in_cluster = acb->sector_num & (s->cluster_sectors - 1);
1219 if (!acb->cluster_offset) {
1220 if (bs->backing_hd) {
1221 /* read from the base image */
1222 n1 = backing_read1(bs->backing_hd, acb->sector_num,
1223 acb->buf, acb->n);
1224 if (n1 > 0) {
1225 acb->hd_aiocb = bdrv_aio_read(bs->backing_hd, acb->sector_num,
1226 acb->buf, acb->n, qcow_aio_read_cb, acb);
1227 if (acb->hd_aiocb == NULL)
1228 goto fail;
1229 } else {
1230 goto redo;
1232 } else {
1233 /* Note: in this case, no need to wait */
1234 memset(acb->buf, 0, 512 * acb->n);
1235 goto redo;
1237 } else if (acb->cluster_offset & QCOW_OFLAG_COMPRESSED) {
1238 /* add AIO support for compressed blocks ? */
1239 if (decompress_cluster(s, acb->cluster_offset) < 0)
1240 goto fail;
1241 memcpy(acb->buf,
1242 s->cluster_cache + index_in_cluster * 512, 512 * acb->n);
1243 goto redo;
1244 } else {
1245 if ((acb->cluster_offset & 511) != 0) {
1246 ret = -EIO;
1247 goto fail;
1249 acb->hd_aiocb = bdrv_aio_read(s->hd,
1250 (acb->cluster_offset >> 9) + index_in_cluster,
1251 acb->buf, acb->n, qcow_aio_read_cb, acb);
1252 if (acb->hd_aiocb == NULL)
1253 goto fail;
1257 static QCowAIOCB *qcow_aio_setup(BlockDriverState *bs,
1258 int64_t sector_num, uint8_t *buf, int nb_sectors,
1259 BlockDriverCompletionFunc *cb, void *opaque)
1261 QCowAIOCB *acb;
1263 acb = qemu_aio_get(bs, cb, opaque);
1264 if (!acb)
1265 return NULL;
1266 acb->hd_aiocb = NULL;
1267 acb->sector_num = sector_num;
1268 acb->buf = buf;
1269 acb->nb_sectors = nb_sectors;
1270 acb->n = 0;
1271 acb->cluster_offset = 0;
1272 return acb;
1275 static BlockDriverAIOCB *qcow_aio_read(BlockDriverState *bs,
1276 int64_t sector_num, uint8_t *buf, int nb_sectors,
1277 BlockDriverCompletionFunc *cb, void *opaque)
1279 QCowAIOCB *acb;
1281 acb = qcow_aio_setup(bs, sector_num, buf, nb_sectors, cb, opaque);
1282 if (!acb)
1283 return NULL;
1285 qcow_aio_read_cb(acb, 0);
1286 return &acb->common;
1289 static void qcow_aio_write_cb(void *opaque, int ret)
1291 QCowAIOCB *acb = opaque;
1292 BlockDriverState *bs = acb->common.bs;
1293 BDRVQcowState *s = bs->opaque;
1294 int index_in_cluster;
1295 uint64_t cluster_offset;
1296 const uint8_t *src_buf;
1297 int n_end;
1299 acb->hd_aiocb = NULL;
1301 if (ret < 0) {
1302 fail:
1303 acb->common.cb(acb->common.opaque, ret);
1304 qemu_aio_release(acb);
1305 return;
1308 acb->nb_sectors -= acb->n;
1309 acb->sector_num += acb->n;
1310 acb->buf += acb->n * 512;
1312 if (acb->nb_sectors == 0) {
1313 /* request completed */
1314 acb->common.cb(acb->common.opaque, 0);
1315 qemu_aio_release(acb);
1316 return;
1319 index_in_cluster = acb->sector_num & (s->cluster_sectors - 1);
1320 n_end = index_in_cluster + acb->nb_sectors;
1321 if (s->crypt_method &&
1322 n_end > QCOW_MAX_CRYPT_CLUSTERS * s->cluster_sectors)
1323 n_end = QCOW_MAX_CRYPT_CLUSTERS * s->cluster_sectors;
1325 cluster_offset = alloc_cluster_offset(bs, acb->sector_num << 9,
1326 index_in_cluster,
1327 n_end, &acb->n);
1328 if (!cluster_offset || (cluster_offset & 511) != 0) {
1329 ret = -EIO;
1330 goto fail;
1332 if (s->crypt_method) {
1333 if (!acb->cluster_data) {
1334 acb->cluster_data = qemu_mallocz(QCOW_MAX_CRYPT_CLUSTERS *
1335 s->cluster_size);
1336 if (!acb->cluster_data) {
1337 ret = -ENOMEM;
1338 goto fail;
1341 encrypt_sectors(s, acb->sector_num, acb->cluster_data, acb->buf,
1342 acb->n, 1, &s->aes_encrypt_key);
1343 src_buf = acb->cluster_data;
1344 } else {
1345 src_buf = acb->buf;
1347 acb->hd_aiocb = bdrv_aio_write(s->hd,
1348 (cluster_offset >> 9) + index_in_cluster,
1349 src_buf, acb->n,
1350 qcow_aio_write_cb, acb);
1351 if (acb->hd_aiocb == NULL)
1352 goto fail;
1355 static BlockDriverAIOCB *qcow_aio_write(BlockDriverState *bs,
1356 int64_t sector_num, const uint8_t *buf, int nb_sectors,
1357 BlockDriverCompletionFunc *cb, void *opaque)
1359 BDRVQcowState *s = bs->opaque;
1360 QCowAIOCB *acb;
1362 s->cluster_cache_offset = -1; /* disable compressed cache */
1364 acb = qcow_aio_setup(bs, sector_num, (uint8_t*)buf, nb_sectors, cb, opaque);
1365 if (!acb)
1366 return NULL;
1368 qcow_aio_write_cb(acb, 0);
1369 return &acb->common;
1372 static void qcow_aio_cancel(BlockDriverAIOCB *blockacb)
1374 QCowAIOCB *acb = (QCowAIOCB *)blockacb;
1375 if (acb->hd_aiocb)
1376 bdrv_aio_cancel(acb->hd_aiocb);
1377 qemu_aio_release(acb);
1380 static void qcow_close(BlockDriverState *bs)
1382 BDRVQcowState *s = bs->opaque;
1383 qemu_free(s->l1_table);
1384 qemu_free(s->l2_cache);
1385 qemu_free(s->cluster_cache);
1386 qemu_free(s->cluster_data);
1387 refcount_close(bs);
1388 bdrv_delete(s->hd);
1391 /* XXX: use std qcow open function ? */
1392 typedef struct QCowCreateState {
1393 int cluster_size;
1394 int cluster_bits;
1395 uint16_t *refcount_block;
1396 uint64_t *refcount_table;
1397 int64_t l1_table_offset;
1398 int64_t refcount_table_offset;
1399 int64_t refcount_block_offset;
1400 } QCowCreateState;
1402 static void create_refcount_update(QCowCreateState *s,
1403 int64_t offset, int64_t size)
1405 int refcount;
1406 int64_t start, last, cluster_offset;
1407 uint16_t *p;
1409 start = offset & ~(s->cluster_size - 1);
1410 last = (offset + size - 1) & ~(s->cluster_size - 1);
1411 for(cluster_offset = start; cluster_offset <= last;
1412 cluster_offset += s->cluster_size) {
1413 p = &s->refcount_block[cluster_offset >> s->cluster_bits];
1414 refcount = be16_to_cpu(*p);
1415 refcount++;
1416 *p = cpu_to_be16(refcount);
1420 static int qcow_create(const char *filename, int64_t total_size,
1421 const char *backing_file, int flags)
1423 int fd, header_size, backing_filename_len, l1_size, i, shift, l2_bits;
1424 QCowHeader header;
1425 uint64_t tmp, offset;
1426 QCowCreateState s1, *s = &s1;
1428 memset(s, 0, sizeof(*s));
1430 fd = open(filename, O_WRONLY | O_CREAT | O_TRUNC | O_BINARY, 0644);
1431 if (fd < 0)
1432 return -1;
1433 memset(&header, 0, sizeof(header));
1434 header.magic = cpu_to_be32(QCOW_MAGIC);
1435 header.version = cpu_to_be32(QCOW_VERSION);
1436 header.size = cpu_to_be64(total_size * 512);
1437 header_size = sizeof(header);
1438 backing_filename_len = 0;
1439 if (backing_file) {
1440 header.backing_file_offset = cpu_to_be64(header_size);
1441 backing_filename_len = strlen(backing_file);
1442 header.backing_file_size = cpu_to_be32(backing_filename_len);
1443 header_size += backing_filename_len;
1445 s->cluster_bits = 12; /* 4 KB clusters */
1446 s->cluster_size = 1 << s->cluster_bits;
1447 header.cluster_bits = cpu_to_be32(s->cluster_bits);
1448 header_size = (header_size + 7) & ~7;
1449 if (flags & BLOCK_FLAG_ENCRYPT) {
1450 header.crypt_method = cpu_to_be32(QCOW_CRYPT_AES);
1451 } else {
1452 header.crypt_method = cpu_to_be32(QCOW_CRYPT_NONE);
1454 l2_bits = s->cluster_bits - 3;
1455 shift = s->cluster_bits + l2_bits;
1456 l1_size = (((total_size * 512) + (1LL << shift) - 1) >> shift);
1457 offset = align_offset(header_size, s->cluster_size);
1458 s->l1_table_offset = offset;
1459 header.l1_table_offset = cpu_to_be64(s->l1_table_offset);
1460 header.l1_size = cpu_to_be32(l1_size);
1461 offset += align_offset(l1_size * sizeof(uint64_t), s->cluster_size);
1463 s->refcount_table = qemu_mallocz(s->cluster_size);
1464 if (!s->refcount_table)
1465 goto fail;
1466 s->refcount_block = qemu_mallocz(s->cluster_size);
1467 if (!s->refcount_block)
1468 goto fail;
1470 s->refcount_table_offset = offset;
1471 header.refcount_table_offset = cpu_to_be64(offset);
1472 header.refcount_table_clusters = cpu_to_be32(1);
1473 offset += s->cluster_size;
1475 s->refcount_table[0] = cpu_to_be64(offset);
1476 s->refcount_block_offset = offset;
1477 offset += s->cluster_size;
1479 /* update refcounts */
1480 create_refcount_update(s, 0, header_size);
1481 create_refcount_update(s, s->l1_table_offset, l1_size * sizeof(uint64_t));
1482 create_refcount_update(s, s->refcount_table_offset, s->cluster_size);
1483 create_refcount_update(s, s->refcount_block_offset, s->cluster_size);
1485 /* write all the data */
1486 write(fd, &header, sizeof(header));
1487 if (backing_file) {
1488 write(fd, backing_file, backing_filename_len);
1490 lseek(fd, s->l1_table_offset, SEEK_SET);
1491 tmp = 0;
1492 for(i = 0;i < l1_size; i++) {
1493 write(fd, &tmp, sizeof(tmp));
1495 lseek(fd, s->refcount_table_offset, SEEK_SET);
1496 write(fd, s->refcount_table, s->cluster_size);
1498 lseek(fd, s->refcount_block_offset, SEEK_SET);
1499 write(fd, s->refcount_block, s->cluster_size);
1501 qemu_free(s->refcount_table);
1502 qemu_free(s->refcount_block);
1503 close(fd);
1504 return 0;
1505 fail:
1506 qemu_free(s->refcount_table);
1507 qemu_free(s->refcount_block);
1508 close(fd);
1509 return -ENOMEM;
1512 static int qcow_make_empty(BlockDriverState *bs)
1514 #if 0
1515 /* XXX: not correct */
1516 BDRVQcowState *s = bs->opaque;
1517 uint32_t l1_length = s->l1_size * sizeof(uint64_t);
1518 int ret;
1520 memset(s->l1_table, 0, l1_length);
1521 if (bdrv_pwrite(s->hd, s->l1_table_offset, s->l1_table, l1_length) < 0)
1522 return -1;
1523 ret = bdrv_truncate(s->hd, s->l1_table_offset + l1_length);
1524 if (ret < 0)
1525 return ret;
1527 l2_cache_reset(bs);
1528 #endif
1529 return 0;
1532 /* XXX: put compressed sectors first, then all the cluster aligned
1533 tables to avoid losing bytes in alignment */
1534 static int qcow_write_compressed(BlockDriverState *bs, int64_t sector_num,
1535 const uint8_t *buf, int nb_sectors)
1537 BDRVQcowState *s = bs->opaque;
1538 z_stream strm;
1539 int ret, out_len;
1540 uint8_t *out_buf;
1541 uint64_t cluster_offset;
1543 if (nb_sectors == 0) {
1544 /* align end of file to a sector boundary to ease reading with
1545 sector based I/Os */
1546 cluster_offset = bdrv_getlength(s->hd);
1547 cluster_offset = (cluster_offset + 511) & ~511;
1548 bdrv_truncate(s->hd, cluster_offset);
1549 return 0;
1552 if (nb_sectors != s->cluster_sectors)
1553 return -EINVAL;
1555 out_buf = qemu_malloc(s->cluster_size + (s->cluster_size / 1000) + 128);
1556 if (!out_buf)
1557 return -ENOMEM;
1559 /* best compression, small window, no zlib header */
1560 memset(&strm, 0, sizeof(strm));
1561 ret = deflateInit2(&strm, Z_DEFAULT_COMPRESSION,
1562 Z_DEFLATED, -12,
1563 9, Z_DEFAULT_STRATEGY);
1564 if (ret != 0) {
1565 qemu_free(out_buf);
1566 return -1;
1569 strm.avail_in = s->cluster_size;
1570 strm.next_in = (uint8_t *)buf;
1571 strm.avail_out = s->cluster_size;
1572 strm.next_out = out_buf;
1574 ret = deflate(&strm, Z_FINISH);
1575 if (ret != Z_STREAM_END && ret != Z_OK) {
1576 qemu_free(out_buf);
1577 deflateEnd(&strm);
1578 return -1;
1580 out_len = strm.next_out - out_buf;
1582 deflateEnd(&strm);
1584 if (ret != Z_STREAM_END || out_len >= s->cluster_size) {
1585 /* could not compress: write normal cluster */
1586 qcow_write(bs, sector_num, buf, s->cluster_sectors);
1587 } else {
1588 cluster_offset = alloc_compressed_cluster_offset(bs, sector_num << 9,
1589 out_len);
1590 if (!cluster_offset)
1591 return -1;
1592 cluster_offset &= s->cluster_offset_mask;
1593 if (bdrv_pwrite(s->hd, cluster_offset, out_buf, out_len) != out_len) {
1594 qemu_free(out_buf);
1595 return -1;
1599 qemu_free(out_buf);
1600 return 0;
1603 static void qcow_flush(BlockDriverState *bs)
1605 BDRVQcowState *s = bs->opaque;
1606 bdrv_flush(s->hd);
1609 static int qcow_get_info(BlockDriverState *bs, BlockDriverInfo *bdi)
1611 BDRVQcowState *s = bs->opaque;
1612 bdi->cluster_size = s->cluster_size;
1613 bdi->vm_state_offset = (int64_t)s->l1_vm_state_index <<
1614 (s->cluster_bits + s->l2_bits);
1615 return 0;
1618 /*********************************************************/
1619 /* snapshot support */
1621 /* update the refcounts of snapshots and the copied flag */
1622 static int update_snapshot_refcount(BlockDriverState *bs,
1623 int64_t l1_table_offset,
1624 int l1_size,
1625 int addend)
1627 BDRVQcowState *s = bs->opaque;
1628 uint64_t *l1_table, *l2_table, l2_offset, offset, l1_size2, l1_allocated;
1629 int64_t old_offset, old_l2_offset;
1630 int l2_size, i, j, l1_modified, l2_modified, nb_csectors, refcount;
1632 l2_cache_reset(bs);
1634 l2_table = NULL;
1635 l1_table = NULL;
1636 l1_size2 = l1_size * sizeof(uint64_t);
1637 l1_allocated = 0;
1638 if (l1_table_offset != s->l1_table_offset) {
1639 l1_table = qemu_malloc(l1_size2);
1640 if (!l1_table)
1641 goto fail;
1642 l1_allocated = 1;
1643 if (bdrv_pread(s->hd, l1_table_offset,
1644 l1_table, l1_size2) != l1_size2)
1645 goto fail;
1646 for(i = 0;i < l1_size; i++)
1647 be64_to_cpus(&l1_table[i]);
1648 } else {
1649 assert(l1_size == s->l1_size);
1650 l1_table = s->l1_table;
1651 l1_allocated = 0;
1654 l2_size = s->l2_size * sizeof(uint64_t);
1655 l2_table = qemu_malloc(l2_size);
1656 if (!l2_table)
1657 goto fail;
1658 l1_modified = 0;
1659 for(i = 0; i < l1_size; i++) {
1660 l2_offset = l1_table[i];
1661 if (l2_offset) {
1662 old_l2_offset = l2_offset;
1663 l2_offset &= ~QCOW_OFLAG_COPIED;
1664 l2_modified = 0;
1665 if (bdrv_pread(s->hd, l2_offset, l2_table, l2_size) != l2_size)
1666 goto fail;
1667 for(j = 0; j < s->l2_size; j++) {
1668 offset = be64_to_cpu(l2_table[j]);
1669 if (offset != 0) {
1670 old_offset = offset;
1671 offset &= ~QCOW_OFLAG_COPIED;
1672 if (offset & QCOW_OFLAG_COMPRESSED) {
1673 nb_csectors = ((offset >> s->csize_shift) &
1674 s->csize_mask) + 1;
1675 if (addend != 0)
1676 update_refcount(bs, (offset & s->cluster_offset_mask) & ~511,
1677 nb_csectors * 512, addend);
1678 /* compressed clusters are never modified */
1679 refcount = 2;
1680 } else {
1681 if (addend != 0) {
1682 refcount = update_cluster_refcount(bs, offset >> s->cluster_bits, addend);
1683 } else {
1684 refcount = get_refcount(bs, offset >> s->cluster_bits);
1688 if (refcount == 1) {
1689 offset |= QCOW_OFLAG_COPIED;
1691 if (offset != old_offset) {
1692 l2_table[j] = cpu_to_be64(offset);
1693 l2_modified = 1;
1697 if (l2_modified) {
1698 if (bdrv_pwrite(s->hd,
1699 l2_offset, l2_table, l2_size) != l2_size)
1700 goto fail;
1703 if (addend != 0) {
1704 refcount = update_cluster_refcount(bs, l2_offset >> s->cluster_bits, addend);
1705 } else {
1706 refcount = get_refcount(bs, l2_offset >> s->cluster_bits);
1708 if (refcount == 1) {
1709 l2_offset |= QCOW_OFLAG_COPIED;
1711 if (l2_offset != old_l2_offset) {
1712 l1_table[i] = l2_offset;
1713 l1_modified = 1;
1717 if (l1_modified) {
1718 for(i = 0; i < l1_size; i++)
1719 cpu_to_be64s(&l1_table[i]);
1720 if (bdrv_pwrite(s->hd, l1_table_offset, l1_table,
1721 l1_size2) != l1_size2)
1722 goto fail;
1723 for(i = 0; i < l1_size; i++)
1724 be64_to_cpus(&l1_table[i]);
1726 if (l1_allocated)
1727 qemu_free(l1_table);
1728 qemu_free(l2_table);
1729 return 0;
1730 fail:
1731 if (l1_allocated)
1732 qemu_free(l1_table);
1733 qemu_free(l2_table);
1734 return -EIO;
1737 static void qcow_free_snapshots(BlockDriverState *bs)
1739 BDRVQcowState *s = bs->opaque;
1740 int i;
1742 for(i = 0; i < s->nb_snapshots; i++) {
1743 qemu_free(s->snapshots[i].name);
1744 qemu_free(s->snapshots[i].id_str);
1746 qemu_free(s->snapshots);
1747 s->snapshots = NULL;
1748 s->nb_snapshots = 0;
1751 static int qcow_read_snapshots(BlockDriverState *bs)
1753 BDRVQcowState *s = bs->opaque;
1754 QCowSnapshotHeader h;
1755 QCowSnapshot *sn;
1756 int i, id_str_size, name_size;
1757 int64_t offset;
1758 uint32_t extra_data_size;
1760 offset = s->snapshots_offset;
1761 s->snapshots = qemu_mallocz(s->nb_snapshots * sizeof(QCowSnapshot));
1762 if (!s->snapshots)
1763 goto fail;
1764 for(i = 0; i < s->nb_snapshots; i++) {
1765 offset = align_offset(offset, 8);
1766 if (bdrv_pread(s->hd, offset, &h, sizeof(h)) != sizeof(h))
1767 goto fail;
1768 offset += sizeof(h);
1769 sn = s->snapshots + i;
1770 sn->l1_table_offset = be64_to_cpu(h.l1_table_offset);
1771 sn->l1_size = be32_to_cpu(h.l1_size);
1772 sn->vm_state_size = be32_to_cpu(h.vm_state_size);
1773 sn->date_sec = be32_to_cpu(h.date_sec);
1774 sn->date_nsec = be32_to_cpu(h.date_nsec);
1775 sn->vm_clock_nsec = be64_to_cpu(h.vm_clock_nsec);
1776 extra_data_size = be32_to_cpu(h.extra_data_size);
1778 id_str_size = be16_to_cpu(h.id_str_size);
1779 name_size = be16_to_cpu(h.name_size);
1781 offset += extra_data_size;
1783 sn->id_str = qemu_malloc(id_str_size + 1);
1784 if (!sn->id_str)
1785 goto fail;
1786 if (bdrv_pread(s->hd, offset, sn->id_str, id_str_size) != id_str_size)
1787 goto fail;
1788 offset += id_str_size;
1789 sn->id_str[id_str_size] = '\0';
1791 sn->name = qemu_malloc(name_size + 1);
1792 if (!sn->name)
1793 goto fail;
1794 if (bdrv_pread(s->hd, offset, sn->name, name_size) != name_size)
1795 goto fail;
1796 offset += name_size;
1797 sn->name[name_size] = '\0';
1799 s->snapshots_size = offset - s->snapshots_offset;
1800 return 0;
1801 fail:
1802 qcow_free_snapshots(bs);
1803 return -1;
1806 /* add at the end of the file a new list of snapshots */
1807 static int qcow_write_snapshots(BlockDriverState *bs)
1809 BDRVQcowState *s = bs->opaque;
1810 QCowSnapshot *sn;
1811 QCowSnapshotHeader h;
1812 int i, name_size, id_str_size, snapshots_size;
1813 uint64_t data64;
1814 uint32_t data32;
1815 int64_t offset, snapshots_offset;
1817 /* compute the size of the snapshots */
1818 offset = 0;
1819 for(i = 0; i < s->nb_snapshots; i++) {
1820 sn = s->snapshots + i;
1821 offset = align_offset(offset, 8);
1822 offset += sizeof(h);
1823 offset += strlen(sn->id_str);
1824 offset += strlen(sn->name);
1826 snapshots_size = offset;
1828 snapshots_offset = alloc_clusters(bs, snapshots_size);
1829 offset = snapshots_offset;
1831 for(i = 0; i < s->nb_snapshots; i++) {
1832 sn = s->snapshots + i;
1833 memset(&h, 0, sizeof(h));
1834 h.l1_table_offset = cpu_to_be64(sn->l1_table_offset);
1835 h.l1_size = cpu_to_be32(sn->l1_size);
1836 h.vm_state_size = cpu_to_be32(sn->vm_state_size);
1837 h.date_sec = cpu_to_be32(sn->date_sec);
1838 h.date_nsec = cpu_to_be32(sn->date_nsec);
1839 h.vm_clock_nsec = cpu_to_be64(sn->vm_clock_nsec);
1841 id_str_size = strlen(sn->id_str);
1842 name_size = strlen(sn->name);
1843 h.id_str_size = cpu_to_be16(id_str_size);
1844 h.name_size = cpu_to_be16(name_size);
1845 offset = align_offset(offset, 8);
1846 if (bdrv_pwrite(s->hd, offset, &h, sizeof(h)) != sizeof(h))
1847 goto fail;
1848 offset += sizeof(h);
1849 if (bdrv_pwrite(s->hd, offset, sn->id_str, id_str_size) != id_str_size)
1850 goto fail;
1851 offset += id_str_size;
1852 if (bdrv_pwrite(s->hd, offset, sn->name, name_size) != name_size)
1853 goto fail;
1854 offset += name_size;
1857 /* update the various header fields */
1858 data64 = cpu_to_be64(snapshots_offset);
1859 if (bdrv_pwrite(s->hd, offsetof(QCowHeader, snapshots_offset),
1860 &data64, sizeof(data64)) != sizeof(data64))
1861 goto fail;
1862 data32 = cpu_to_be32(s->nb_snapshots);
1863 if (bdrv_pwrite(s->hd, offsetof(QCowHeader, nb_snapshots),
1864 &data32, sizeof(data32)) != sizeof(data32))
1865 goto fail;
1867 /* free the old snapshot table */
1868 free_clusters(bs, s->snapshots_offset, s->snapshots_size);
1869 s->snapshots_offset = snapshots_offset;
1870 s->snapshots_size = snapshots_size;
1871 return 0;
1872 fail:
1873 return -1;
1876 static void find_new_snapshot_id(BlockDriverState *bs,
1877 char *id_str, int id_str_size)
1879 BDRVQcowState *s = bs->opaque;
1880 QCowSnapshot *sn;
1881 int i, id, id_max = 0;
1883 for(i = 0; i < s->nb_snapshots; i++) {
1884 sn = s->snapshots + i;
1885 id = strtoul(sn->id_str, NULL, 10);
1886 if (id > id_max)
1887 id_max = id;
1889 snprintf(id_str, id_str_size, "%d", id_max + 1);
1892 static int find_snapshot_by_id(BlockDriverState *bs, const char *id_str)
1894 BDRVQcowState *s = bs->opaque;
1895 int i;
1897 for(i = 0; i < s->nb_snapshots; i++) {
1898 if (!strcmp(s->snapshots[i].id_str, id_str))
1899 return i;
1901 return -1;
1904 static int find_snapshot_by_id_or_name(BlockDriverState *bs, const char *name)
1906 BDRVQcowState *s = bs->opaque;
1907 int i, ret;
1909 ret = find_snapshot_by_id(bs, name);
1910 if (ret >= 0)
1911 return ret;
1912 for(i = 0; i < s->nb_snapshots; i++) {
1913 if (!strcmp(s->snapshots[i].name, name))
1914 return i;
1916 return -1;
1919 /* if no id is provided, a new one is constructed */
1920 static int qcow_snapshot_create(BlockDriverState *bs,
1921 QEMUSnapshotInfo *sn_info)
1923 BDRVQcowState *s = bs->opaque;
1924 QCowSnapshot *snapshots1, sn1, *sn = &sn1;
1925 int i, ret;
1926 uint64_t *l1_table = NULL;
1928 memset(sn, 0, sizeof(*sn));
1930 if (sn_info->id_str[0] == '\0') {
1931 /* compute a new id */
1932 find_new_snapshot_id(bs, sn_info->id_str, sizeof(sn_info->id_str));
1935 /* check that the ID is unique */
1936 if (find_snapshot_by_id(bs, sn_info->id_str) >= 0)
1937 return -ENOENT;
1939 sn->id_str = qemu_strdup(sn_info->id_str);
1940 if (!sn->id_str)
1941 goto fail;
1942 sn->name = qemu_strdup(sn_info->name);
1943 if (!sn->name)
1944 goto fail;
1945 sn->vm_state_size = sn_info->vm_state_size;
1946 sn->date_sec = sn_info->date_sec;
1947 sn->date_nsec = sn_info->date_nsec;
1948 sn->vm_clock_nsec = sn_info->vm_clock_nsec;
1950 ret = update_snapshot_refcount(bs, s->l1_table_offset, s->l1_size, 1);
1951 if (ret < 0)
1952 goto fail;
1954 /* create the L1 table of the snapshot */
1955 sn->l1_table_offset = alloc_clusters(bs, s->l1_size * sizeof(uint64_t));
1956 sn->l1_size = s->l1_size;
1958 l1_table = qemu_malloc(s->l1_size * sizeof(uint64_t));
1959 if (!l1_table)
1960 goto fail;
1961 for(i = 0; i < s->l1_size; i++) {
1962 l1_table[i] = cpu_to_be64(s->l1_table[i]);
1964 if (bdrv_pwrite(s->hd, sn->l1_table_offset,
1965 l1_table, s->l1_size * sizeof(uint64_t)) !=
1966 (s->l1_size * sizeof(uint64_t)))
1967 goto fail;
1968 qemu_free(l1_table);
1969 l1_table = NULL;
1971 snapshots1 = qemu_malloc((s->nb_snapshots + 1) * sizeof(QCowSnapshot));
1972 if (!snapshots1)
1973 goto fail;
1974 memcpy(snapshots1, s->snapshots, s->nb_snapshots * sizeof(QCowSnapshot));
1975 s->snapshots = snapshots1;
1976 s->snapshots[s->nb_snapshots++] = *sn;
1978 if (qcow_write_snapshots(bs) < 0)
1979 goto fail;
1980 #ifdef DEBUG_ALLOC
1981 check_refcounts(bs);
1982 #endif
1983 return 0;
1984 fail:
1985 qemu_free(sn->name);
1986 qemu_free(l1_table);
1987 return -1;
1990 /* copy the snapshot 'snapshot_name' into the current disk image */
1991 static int qcow_snapshot_goto(BlockDriverState *bs,
1992 const char *snapshot_id)
1994 BDRVQcowState *s = bs->opaque;
1995 QCowSnapshot *sn;
1996 int i, snapshot_index, l1_size2;
1998 snapshot_index = find_snapshot_by_id_or_name(bs, snapshot_id);
1999 if (snapshot_index < 0)
2000 return -ENOENT;
2001 sn = &s->snapshots[snapshot_index];
2003 if (update_snapshot_refcount(bs, s->l1_table_offset, s->l1_size, -1) < 0)
2004 goto fail;
2006 if (grow_l1_table(bs, sn->l1_size) < 0)
2007 goto fail;
2009 s->l1_size = sn->l1_size;
2010 l1_size2 = s->l1_size * sizeof(uint64_t);
2011 /* copy the snapshot l1 table to the current l1 table */
2012 if (bdrv_pread(s->hd, sn->l1_table_offset,
2013 s->l1_table, l1_size2) != l1_size2)
2014 goto fail;
2015 if (bdrv_pwrite(s->hd, s->l1_table_offset,
2016 s->l1_table, l1_size2) != l1_size2)
2017 goto fail;
2018 for(i = 0;i < s->l1_size; i++) {
2019 be64_to_cpus(&s->l1_table[i]);
2022 if (update_snapshot_refcount(bs, s->l1_table_offset, s->l1_size, 1) < 0)
2023 goto fail;
2025 #ifdef DEBUG_ALLOC
2026 check_refcounts(bs);
2027 #endif
2028 return 0;
2029 fail:
2030 return -EIO;
2033 static int qcow_snapshot_delete(BlockDriverState *bs, const char *snapshot_id)
2035 BDRVQcowState *s = bs->opaque;
2036 QCowSnapshot *sn;
2037 int snapshot_index, ret;
2039 snapshot_index = find_snapshot_by_id_or_name(bs, snapshot_id);
2040 if (snapshot_index < 0)
2041 return -ENOENT;
2042 sn = &s->snapshots[snapshot_index];
2044 ret = update_snapshot_refcount(bs, sn->l1_table_offset, sn->l1_size, -1);
2045 if (ret < 0)
2046 return ret;
2047 /* must update the copied flag on the current cluster offsets */
2048 ret = update_snapshot_refcount(bs, s->l1_table_offset, s->l1_size, 0);
2049 if (ret < 0)
2050 return ret;
2051 free_clusters(bs, sn->l1_table_offset, sn->l1_size * sizeof(uint64_t));
2053 qemu_free(sn->id_str);
2054 qemu_free(sn->name);
2055 memmove(sn, sn + 1, (s->nb_snapshots - snapshot_index - 1) * sizeof(*sn));
2056 s->nb_snapshots--;
2057 ret = qcow_write_snapshots(bs);
2058 if (ret < 0) {
2059 /* XXX: restore snapshot if error ? */
2060 return ret;
2062 #ifdef DEBUG_ALLOC
2063 check_refcounts(bs);
2064 #endif
2065 return 0;
2068 static int qcow_snapshot_list(BlockDriverState *bs,
2069 QEMUSnapshotInfo **psn_tab)
2071 BDRVQcowState *s = bs->opaque;
2072 QEMUSnapshotInfo *sn_tab, *sn_info;
2073 QCowSnapshot *sn;
2074 int i;
2076 sn_tab = qemu_mallocz(s->nb_snapshots * sizeof(QEMUSnapshotInfo));
2077 if (!sn_tab)
2078 goto fail;
2079 for(i = 0; i < s->nb_snapshots; i++) {
2080 sn_info = sn_tab + i;
2081 sn = s->snapshots + i;
2082 pstrcpy(sn_info->id_str, sizeof(sn_info->id_str),
2083 sn->id_str);
2084 pstrcpy(sn_info->name, sizeof(sn_info->name),
2085 sn->name);
2086 sn_info->vm_state_size = sn->vm_state_size;
2087 sn_info->date_sec = sn->date_sec;
2088 sn_info->date_nsec = sn->date_nsec;
2089 sn_info->vm_clock_nsec = sn->vm_clock_nsec;
2091 *psn_tab = sn_tab;
2092 return s->nb_snapshots;
2093 fail:
2094 qemu_free(sn_tab);
2095 *psn_tab = NULL;
2096 return -ENOMEM;
2099 /*********************************************************/
2100 /* refcount handling */
2102 static int refcount_init(BlockDriverState *bs)
2104 BDRVQcowState *s = bs->opaque;
2105 int ret, refcount_table_size2, i;
2107 s->refcount_block_cache = qemu_malloc(s->cluster_size);
2108 if (!s->refcount_block_cache)
2109 goto fail;
2110 refcount_table_size2 = s->refcount_table_size * sizeof(uint64_t);
2111 s->refcount_table = qemu_malloc(refcount_table_size2);
2112 if (!s->refcount_table)
2113 goto fail;
2114 if (s->refcount_table_size > 0) {
2115 ret = bdrv_pread(s->hd, s->refcount_table_offset,
2116 s->refcount_table, refcount_table_size2);
2117 if (ret != refcount_table_size2)
2118 goto fail;
2119 for(i = 0; i < s->refcount_table_size; i++)
2120 be64_to_cpus(&s->refcount_table[i]);
2122 return 0;
2123 fail:
2124 return -ENOMEM;
2127 static void refcount_close(BlockDriverState *bs)
2129 BDRVQcowState *s = bs->opaque;
2130 qemu_free(s->refcount_block_cache);
2131 qemu_free(s->refcount_table);
2135 static int load_refcount_block(BlockDriverState *bs,
2136 int64_t refcount_block_offset)
2138 BDRVQcowState *s = bs->opaque;
2139 int ret;
2140 ret = bdrv_pread(s->hd, refcount_block_offset, s->refcount_block_cache,
2141 s->cluster_size);
2142 if (ret != s->cluster_size)
2143 return -EIO;
2144 s->refcount_block_cache_offset = refcount_block_offset;
2145 return 0;
2148 static int get_refcount(BlockDriverState *bs, int64_t cluster_index)
2150 BDRVQcowState *s = bs->opaque;
2151 int refcount_table_index, block_index;
2152 int64_t refcount_block_offset;
2154 refcount_table_index = cluster_index >> (s->cluster_bits - REFCOUNT_SHIFT);
2155 if (refcount_table_index >= s->refcount_table_size)
2156 return 0;
2157 refcount_block_offset = s->refcount_table[refcount_table_index];
2158 if (!refcount_block_offset)
2159 return 0;
2160 if (refcount_block_offset != s->refcount_block_cache_offset) {
2161 /* better than nothing: return allocated if read error */
2162 if (load_refcount_block(bs, refcount_block_offset) < 0)
2163 return 1;
2165 block_index = cluster_index &
2166 ((1 << (s->cluster_bits - REFCOUNT_SHIFT)) - 1);
2167 return be16_to_cpu(s->refcount_block_cache[block_index]);
2170 /* return < 0 if error */
2171 static int64_t alloc_clusters_noref(BlockDriverState *bs, int64_t size)
2173 BDRVQcowState *s = bs->opaque;
2174 int i, nb_clusters;
2176 nb_clusters = (size + s->cluster_size - 1) >> s->cluster_bits;
2177 for(;;) {
2178 if (get_refcount(bs, s->free_cluster_index) == 0) {
2179 s->free_cluster_index++;
2180 for(i = 1; i < nb_clusters; i++) {
2181 if (get_refcount(bs, s->free_cluster_index) != 0)
2182 goto not_found;
2183 s->free_cluster_index++;
2185 #ifdef DEBUG_ALLOC2
2186 printf("alloc_clusters: size=%lld -> %lld\n",
2187 size,
2188 (s->free_cluster_index - nb_clusters) << s->cluster_bits);
2189 #endif
2190 return (s->free_cluster_index - nb_clusters) << s->cluster_bits;
2191 } else {
2192 not_found:
2193 s->free_cluster_index++;
2198 static int64_t alloc_clusters(BlockDriverState *bs, int64_t size)
2200 int64_t offset;
2202 offset = alloc_clusters_noref(bs, size);
2203 update_refcount(bs, offset, size, 1);
2204 return offset;
2207 /* only used to allocate compressed sectors. We try to allocate
2208 contiguous sectors. size must be <= cluster_size */
2209 static int64_t alloc_bytes(BlockDriverState *bs, int size)
2211 BDRVQcowState *s = bs->opaque;
2212 int64_t offset, cluster_offset;
2213 int free_in_cluster;
2215 assert(size > 0 && size <= s->cluster_size);
2216 if (s->free_byte_offset == 0) {
2217 s->free_byte_offset = alloc_clusters(bs, s->cluster_size);
2219 redo:
2220 free_in_cluster = s->cluster_size -
2221 (s->free_byte_offset & (s->cluster_size - 1));
2222 if (size <= free_in_cluster) {
2223 /* enough space in current cluster */
2224 offset = s->free_byte_offset;
2225 s->free_byte_offset += size;
2226 free_in_cluster -= size;
2227 if (free_in_cluster == 0)
2228 s->free_byte_offset = 0;
2229 if ((offset & (s->cluster_size - 1)) != 0)
2230 update_cluster_refcount(bs, offset >> s->cluster_bits, 1);
2231 } else {
2232 offset = alloc_clusters(bs, s->cluster_size);
2233 cluster_offset = s->free_byte_offset & ~(s->cluster_size - 1);
2234 if ((cluster_offset + s->cluster_size) == offset) {
2235 /* we are lucky: contiguous data */
2236 offset = s->free_byte_offset;
2237 update_cluster_refcount(bs, offset >> s->cluster_bits, 1);
2238 s->free_byte_offset += size;
2239 } else {
2240 s->free_byte_offset = offset;
2241 goto redo;
2244 return offset;
2247 static void free_clusters(BlockDriverState *bs,
2248 int64_t offset, int64_t size)
2250 update_refcount(bs, offset, size, -1);
2253 static int grow_refcount_table(BlockDriverState *bs, int min_size)
2255 BDRVQcowState *s = bs->opaque;
2256 int new_table_size, new_table_size2, refcount_table_clusters, i, ret;
2257 uint64_t *new_table;
2258 int64_t table_offset;
2259 uint64_t data64;
2260 uint32_t data32;
2261 int old_table_size;
2262 int64_t old_table_offset;
2264 if (min_size <= s->refcount_table_size)
2265 return 0;
2266 /* compute new table size */
2267 refcount_table_clusters = s->refcount_table_size >> (s->cluster_bits - 3);
2268 for(;;) {
2269 if (refcount_table_clusters == 0) {
2270 refcount_table_clusters = 1;
2271 } else {
2272 refcount_table_clusters = (refcount_table_clusters * 3 + 1) / 2;
2274 new_table_size = refcount_table_clusters << (s->cluster_bits - 3);
2275 if (min_size <= new_table_size)
2276 break;
2278 #ifdef DEBUG_ALLOC2
2279 printf("grow_refcount_table from %d to %d\n",
2280 s->refcount_table_size,
2281 new_table_size);
2282 #endif
2283 new_table_size2 = new_table_size * sizeof(uint64_t);
2284 new_table = qemu_mallocz(new_table_size2);
2285 if (!new_table)
2286 return -ENOMEM;
2287 memcpy(new_table, s->refcount_table,
2288 s->refcount_table_size * sizeof(uint64_t));
2289 for(i = 0; i < s->refcount_table_size; i++)
2290 cpu_to_be64s(&new_table[i]);
2291 /* Note: we cannot update the refcount now to avoid recursion */
2292 table_offset = alloc_clusters_noref(bs, new_table_size2);
2293 ret = bdrv_pwrite(s->hd, table_offset, new_table, new_table_size2);
2294 if (ret != new_table_size2)
2295 goto fail;
2296 for(i = 0; i < s->refcount_table_size; i++)
2297 be64_to_cpus(&new_table[i]);
2299 data64 = cpu_to_be64(table_offset);
2300 if (bdrv_pwrite(s->hd, offsetof(QCowHeader, refcount_table_offset),
2301 &data64, sizeof(data64)) != sizeof(data64))
2302 goto fail;
2303 data32 = cpu_to_be32(refcount_table_clusters);
2304 if (bdrv_pwrite(s->hd, offsetof(QCowHeader, refcount_table_clusters),
2305 &data32, sizeof(data32)) != sizeof(data32))
2306 goto fail;
2307 qemu_free(s->refcount_table);
2308 old_table_offset = s->refcount_table_offset;
2309 old_table_size = s->refcount_table_size;
2310 s->refcount_table = new_table;
2311 s->refcount_table_size = new_table_size;
2312 s->refcount_table_offset = table_offset;
2314 update_refcount(bs, table_offset, new_table_size2, 1);
2315 free_clusters(bs, old_table_offset, old_table_size * sizeof(uint64_t));
2316 return 0;
2317 fail:
2318 free_clusters(bs, table_offset, new_table_size2);
2319 qemu_free(new_table);
2320 return -EIO;
2323 /* addend must be 1 or -1 */
2324 /* XXX: cache several refcount block clusters ? */
2325 static int update_cluster_refcount(BlockDriverState *bs,
2326 int64_t cluster_index,
2327 int addend)
2329 BDRVQcowState *s = bs->opaque;
2330 int64_t offset, refcount_block_offset;
2331 int ret, refcount_table_index, block_index, refcount;
2332 uint64_t data64;
2334 refcount_table_index = cluster_index >> (s->cluster_bits - REFCOUNT_SHIFT);
2335 if (refcount_table_index >= s->refcount_table_size) {
2336 if (addend < 0)
2337 return -EINVAL;
2338 ret = grow_refcount_table(bs, refcount_table_index + 1);
2339 if (ret < 0)
2340 return ret;
2342 refcount_block_offset = s->refcount_table[refcount_table_index];
2343 if (!refcount_block_offset) {
2344 if (addend < 0)
2345 return -EINVAL;
2346 /* create a new refcount block */
2347 /* Note: we cannot update the refcount now to avoid recursion */
2348 offset = alloc_clusters_noref(bs, s->cluster_size);
2349 memset(s->refcount_block_cache, 0, s->cluster_size);
2350 ret = bdrv_pwrite(s->hd, offset, s->refcount_block_cache, s->cluster_size);
2351 if (ret != s->cluster_size)
2352 return -EINVAL;
2353 s->refcount_table[refcount_table_index] = offset;
2354 data64 = cpu_to_be64(offset);
2355 ret = bdrv_pwrite(s->hd, s->refcount_table_offset +
2356 refcount_table_index * sizeof(uint64_t),
2357 &data64, sizeof(data64));
2358 if (ret != sizeof(data64))
2359 return -EINVAL;
2361 refcount_block_offset = offset;
2362 s->refcount_block_cache_offset = offset;
2363 update_refcount(bs, offset, s->cluster_size, 1);
2364 } else {
2365 if (refcount_block_offset != s->refcount_block_cache_offset) {
2366 if (load_refcount_block(bs, refcount_block_offset) < 0)
2367 return -EIO;
2370 /* we can update the count and save it */
2371 block_index = cluster_index &
2372 ((1 << (s->cluster_bits - REFCOUNT_SHIFT)) - 1);
2373 refcount = be16_to_cpu(s->refcount_block_cache[block_index]);
2374 refcount += addend;
2375 if (refcount < 0 || refcount > 0xffff)
2376 return -EINVAL;
2377 if (refcount == 0 && cluster_index < s->free_cluster_index) {
2378 s->free_cluster_index = cluster_index;
2380 s->refcount_block_cache[block_index] = cpu_to_be16(refcount);
2381 if (bdrv_pwrite(s->hd,
2382 refcount_block_offset + (block_index << REFCOUNT_SHIFT),
2383 &s->refcount_block_cache[block_index], 2) != 2)
2384 return -EIO;
2385 return refcount;
2388 static void update_refcount(BlockDriverState *bs,
2389 int64_t offset, int64_t length,
2390 int addend)
2392 BDRVQcowState *s = bs->opaque;
2393 int64_t start, last, cluster_offset;
2395 #ifdef DEBUG_ALLOC2
2396 printf("update_refcount: offset=%lld size=%lld addend=%d\n",
2397 offset, length, addend);
2398 #endif
2399 if (length <= 0)
2400 return;
2401 start = offset & ~(s->cluster_size - 1);
2402 last = (offset + length - 1) & ~(s->cluster_size - 1);
2403 for(cluster_offset = start; cluster_offset <= last;
2404 cluster_offset += s->cluster_size) {
2405 update_cluster_refcount(bs, cluster_offset >> s->cluster_bits, addend);
2409 #ifdef DEBUG_ALLOC
2410 static void inc_refcounts(BlockDriverState *bs,
2411 uint16_t *refcount_table,
2412 int refcount_table_size,
2413 int64_t offset, int64_t size)
2415 BDRVQcowState *s = bs->opaque;
2416 int64_t start, last, cluster_offset;
2417 int k;
2419 if (size <= 0)
2420 return;
2422 start = offset & ~(s->cluster_size - 1);
2423 last = (offset + size - 1) & ~(s->cluster_size - 1);
2424 for(cluster_offset = start; cluster_offset <= last;
2425 cluster_offset += s->cluster_size) {
2426 k = cluster_offset >> s->cluster_bits;
2427 if (k < 0 || k >= refcount_table_size) {
2428 printf("ERROR: invalid cluster offset=0x%llx\n", cluster_offset);
2429 } else {
2430 if (++refcount_table[k] == 0) {
2431 printf("ERROR: overflow cluster offset=0x%llx\n", cluster_offset);
2437 static int check_refcounts_l1(BlockDriverState *bs,
2438 uint16_t *refcount_table,
2439 int refcount_table_size,
2440 int64_t l1_table_offset, int l1_size,
2441 int check_copied)
2443 BDRVQcowState *s = bs->opaque;
2444 uint64_t *l1_table, *l2_table, l2_offset, offset, l1_size2;
2445 int l2_size, i, j, nb_csectors, refcount;
2447 l2_table = NULL;
2448 l1_size2 = l1_size * sizeof(uint64_t);
2450 inc_refcounts(bs, refcount_table, refcount_table_size,
2451 l1_table_offset, l1_size2);
2453 l1_table = qemu_malloc(l1_size2);
2454 if (!l1_table)
2455 goto fail;
2456 if (bdrv_pread(s->hd, l1_table_offset,
2457 l1_table, l1_size2) != l1_size2)
2458 goto fail;
2459 for(i = 0;i < l1_size; i++)
2460 be64_to_cpus(&l1_table[i]);
2462 l2_size = s->l2_size * sizeof(uint64_t);
2463 l2_table = qemu_malloc(l2_size);
2464 if (!l2_table)
2465 goto fail;
2466 for(i = 0; i < l1_size; i++) {
2467 l2_offset = l1_table[i];
2468 if (l2_offset) {
2469 if (check_copied) {
2470 refcount = get_refcount(bs, (l2_offset & ~QCOW_OFLAG_COPIED) >> s->cluster_bits);
2471 if ((refcount == 1) != ((l2_offset & QCOW_OFLAG_COPIED) != 0)) {
2472 printf("ERROR OFLAG_COPIED: l2_offset=%llx refcount=%d\n",
2473 l2_offset, refcount);
2476 l2_offset &= ~QCOW_OFLAG_COPIED;
2477 if (bdrv_pread(s->hd, l2_offset, l2_table, l2_size) != l2_size)
2478 goto fail;
2479 for(j = 0; j < s->l2_size; j++) {
2480 offset = be64_to_cpu(l2_table[j]);
2481 if (offset != 0) {
2482 if (offset & QCOW_OFLAG_COMPRESSED) {
2483 if (offset & QCOW_OFLAG_COPIED) {
2484 printf("ERROR: cluster %lld: copied flag must never be set for compressed clusters\n",
2485 offset >> s->cluster_bits);
2486 offset &= ~QCOW_OFLAG_COPIED;
2488 nb_csectors = ((offset >> s->csize_shift) &
2489 s->csize_mask) + 1;
2490 offset &= s->cluster_offset_mask;
2491 inc_refcounts(bs, refcount_table,
2492 refcount_table_size,
2493 offset & ~511, nb_csectors * 512);
2494 } else {
2495 if (check_copied) {
2496 refcount = get_refcount(bs, (offset & ~QCOW_OFLAG_COPIED) >> s->cluster_bits);
2497 if ((refcount == 1) != ((offset & QCOW_OFLAG_COPIED) != 0)) {
2498 printf("ERROR OFLAG_COPIED: offset=%llx refcount=%d\n",
2499 offset, refcount);
2502 offset &= ~QCOW_OFLAG_COPIED;
2503 inc_refcounts(bs, refcount_table,
2504 refcount_table_size,
2505 offset, s->cluster_size);
2509 inc_refcounts(bs, refcount_table,
2510 refcount_table_size,
2511 l2_offset,
2512 s->cluster_size);
2515 qemu_free(l1_table);
2516 qemu_free(l2_table);
2517 return 0;
2518 fail:
2519 printf("ERROR: I/O error in check_refcounts_l1\n");
2520 qemu_free(l1_table);
2521 qemu_free(l2_table);
2522 return -EIO;
2525 static void check_refcounts(BlockDriverState *bs)
2527 BDRVQcowState *s = bs->opaque;
2528 int64_t size;
2529 int nb_clusters, refcount1, refcount2, i;
2530 QCowSnapshot *sn;
2531 uint16_t *refcount_table;
2533 size = bdrv_getlength(s->hd);
2534 nb_clusters = (size + s->cluster_size - 1) >> s->cluster_bits;
2535 refcount_table = qemu_mallocz(nb_clusters * sizeof(uint16_t));
2537 /* header */
2538 inc_refcounts(bs, refcount_table, nb_clusters,
2539 0, s->cluster_size);
2541 check_refcounts_l1(bs, refcount_table, nb_clusters,
2542 s->l1_table_offset, s->l1_size, 1);
2544 /* snapshots */
2545 for(i = 0; i < s->nb_snapshots; i++) {
2546 sn = s->snapshots + i;
2547 check_refcounts_l1(bs, refcount_table, nb_clusters,
2548 sn->l1_table_offset, sn->l1_size, 0);
2550 inc_refcounts(bs, refcount_table, nb_clusters,
2551 s->snapshots_offset, s->snapshots_size);
2553 /* refcount data */
2554 inc_refcounts(bs, refcount_table, nb_clusters,
2555 s->refcount_table_offset,
2556 s->refcount_table_size * sizeof(uint64_t));
2557 for(i = 0; i < s->refcount_table_size; i++) {
2558 int64_t offset;
2559 offset = s->refcount_table[i];
2560 if (offset != 0) {
2561 inc_refcounts(bs, refcount_table, nb_clusters,
2562 offset, s->cluster_size);
2566 /* compare ref counts */
2567 for(i = 0; i < nb_clusters; i++) {
2568 refcount1 = get_refcount(bs, i);
2569 refcount2 = refcount_table[i];
2570 if (refcount1 != refcount2)
2571 printf("ERROR cluster %d refcount=%d reference=%d\n",
2572 i, refcount1, refcount2);
2575 qemu_free(refcount_table);
2578 #if 0
2579 static void dump_refcounts(BlockDriverState *bs)
2581 BDRVQcowState *s = bs->opaque;
2582 int64_t nb_clusters, k, k1, size;
2583 int refcount;
2585 size = bdrv_getlength(s->hd);
2586 nb_clusters = (size + s->cluster_size - 1) >> s->cluster_bits;
2587 for(k = 0; k < nb_clusters;) {
2588 k1 = k;
2589 refcount = get_refcount(bs, k);
2590 k++;
2591 while (k < nb_clusters && get_refcount(bs, k) == refcount)
2592 k++;
2593 printf("%lld: refcount=%d nb=%lld\n", k, refcount, k - k1);
2596 #endif
2597 #endif
2599 BlockDriver bdrv_qcow2 = {
2600 "qcow2",
2601 sizeof(BDRVQcowState),
2602 qcow_probe,
2603 qcow_open,
2604 NULL,
2605 NULL,
2606 qcow_close,
2607 qcow_create,
2608 qcow_flush,
2609 qcow_is_allocated,
2610 qcow_set_key,
2611 qcow_make_empty,
2613 .bdrv_aio_read = qcow_aio_read,
2614 .bdrv_aio_write = qcow_aio_write,
2615 .bdrv_aio_cancel = qcow_aio_cancel,
2616 .aiocb_size = sizeof(QCowAIOCB),
2617 .bdrv_write_compressed = qcow_write_compressed,
2619 .bdrv_snapshot_create = qcow_snapshot_create,
2620 .bdrv_snapshot_goto = qcow_snapshot_goto,
2621 .bdrv_snapshot_delete = qcow_snapshot_delete,
2622 .bdrv_snapshot_list = qcow_snapshot_list,
2623 .bdrv_get_info = qcow_get_info,