uint32_t magic;
uint32_t len;
} QCowExtension;
+
#define QCOW2_EXT_MAGIC_END 0
#define QCOW2_EXT_MAGIC_BACKING_FORMAT 0xE2792ACA
+#define QCOW2_EXT_MAGIC_FEATURE_TABLE 0x6803f857
static int qcow2_probe(const uint8_t *buf, int buf_size, const char *filename)
{
* return 0 upon success, non-0 otherwise
*/
static int qcow2_read_extensions(BlockDriverState *bs, uint64_t start_offset,
- uint64_t end_offset)
+ uint64_t end_offset, void **p_feature_table)
{
BDRVQcowState *s = bs->opaque;
QCowExtension ext;
#endif
break;
+ case QCOW2_EXT_MAGIC_FEATURE_TABLE:
+ if (p_feature_table != NULL) {
+ void* feature_table = g_malloc0(ext.len + 2 * sizeof(Qcow2Feature));
+ ret = bdrv_pread(bs->file, offset , feature_table, ext.len);
+ if (ret < 0) {
+ return ret;
+ }
+
+ *p_feature_table = feature_table;
+ }
+ break;
+
default:
/* unknown magic - save it in case we need to rewrite the header */
{
}
}
-static void report_unsupported(BlockDriverState *bs, const char *fmt, ...)
+static void GCC_FMT_ATTR(2, 3) report_unsupported(BlockDriverState *bs,
+ const char *fmt, ...)
{
char msg[64];
va_list ap;
bs->device_name, "qcow2", msg);
}
+static void report_unsupported_feature(BlockDriverState *bs,
+ Qcow2Feature *table, uint64_t mask)
+{
+ while (table && table->name[0] != '\0') {
+ if (table->type == QCOW2_FEAT_TYPE_INCOMPATIBLE) {
+ if (mask & (1 << table->bit)) {
+ report_unsupported(bs, "%.46s",table->name);
+ mask &= ~(1 << table->bit);
+ }
+ }
+ table++;
+ }
+
+ if (mask) {
+ report_unsupported(bs, "Unknown incompatible feature: %" PRIx64, mask);
+ }
+}
+
+/*
+ * Sets the dirty bit and flushes afterwards if necessary.
+ *
+ * The incompatible_features bit is only set if the image file header was
+ * updated successfully. Therefore it is not required to check the return
+ * value of this function.
+ */
+static int qcow2_mark_dirty(BlockDriverState *bs)
+{
+ BDRVQcowState *s = bs->opaque;
+ uint64_t val;
+ int ret;
+
+ assert(s->qcow_version >= 3);
+
+ if (s->incompatible_features & QCOW2_INCOMPAT_DIRTY) {
+ return 0; /* already dirty */
+ }
+
+ val = cpu_to_be64(s->incompatible_features | QCOW2_INCOMPAT_DIRTY);
+ ret = bdrv_pwrite(bs->file, offsetof(QCowHeader, incompatible_features),
+ &val, sizeof(val));
+ if (ret < 0) {
+ return ret;
+ }
+ ret = bdrv_flush(bs->file);
+ if (ret < 0) {
+ return ret;
+ }
+
+ /* Only treat image as dirty if the header was updated successfully */
+ s->incompatible_features |= QCOW2_INCOMPAT_DIRTY;
+ return 0;
+}
+
+/*
+ * Clears the dirty bit and flushes before if necessary. Only call this
+ * function when there are no pending requests, it does not guard against
+ * concurrent requests dirtying the image.
+ */
+static int qcow2_mark_clean(BlockDriverState *bs)
+{
+ BDRVQcowState *s = bs->opaque;
+
+ if (s->incompatible_features & QCOW2_INCOMPAT_DIRTY) {
+ int ret = bdrv_flush(bs);
+ if (ret < 0) {
+ return ret;
+ }
+
+ s->incompatible_features &= ~QCOW2_INCOMPAT_DIRTY;
+ return qcow2_update_header(bs);
+ }
+ return 0;
+}
+
+static int qcow2_check(BlockDriverState *bs, BdrvCheckResult *result,
+ BdrvCheckMode fix)
+{
+ int ret = qcow2_check_refcounts(bs, result, fix);
+ if (ret < 0) {
+ return ret;
+ }
+
+ if (fix && result->check_errors == 0 && result->corruptions == 0) {
+ return qcow2_mark_clean(bs);
+ }
+ return ret;
+}
+
static int qcow2_open(BlockDriverState *bs, int flags)
{
BDRVQcowState *s = bs->opaque;
int len, i, ret = 0;
QCowHeader header;
uint64_t ext_end;
- bool writethrough;
ret = bdrv_pread(bs->file, 0, &header, sizeof(header));
if (ret < 0) {
}
}
+ if (header.backing_file_offset) {
+ ext_end = header.backing_file_offset;
+ } else {
+ ext_end = 1 << header.cluster_bits;
+ }
+
/* Handle feature bits */
s->incompatible_features = header.incompatible_features;
s->compatible_features = header.compatible_features;
s->autoclear_features = header.autoclear_features;
- if (s->incompatible_features != 0) {
- report_unsupported(bs, "incompatible features mask %" PRIx64,
- header.incompatible_features);
+ if (s->incompatible_features & ~QCOW2_INCOMPAT_MASK) {
+ void *feature_table = NULL;
+ qcow2_read_extensions(bs, header.header_length, ext_end,
+ &feature_table);
+ report_unsupported_feature(bs, feature_table,
+ s->incompatible_features &
+ ~QCOW2_INCOMPAT_MASK);
ret = -ENOTSUP;
goto fail;
}
- if (!bs->read_only && s->autoclear_features != 0) {
- s->autoclear_features = 0;
- qcow2_update_header(bs);
- }
-
/* Check support for various header values */
if (header.refcount_order != 4) {
report_unsupported(bs, "%d bit reference counts",
}
/* alloc L2 table/refcount block cache */
- writethrough = ((flags & BDRV_O_CACHE_WB) == 0);
- s->l2_table_cache = qcow2_cache_create(bs, L2_CACHE_SIZE, writethrough);
- s->refcount_block_cache = qcow2_cache_create(bs, REFCOUNT_CACHE_SIZE,
- writethrough);
+ s->l2_table_cache = qcow2_cache_create(bs, L2_CACHE_SIZE);
+ s->refcount_block_cache = qcow2_cache_create(bs, REFCOUNT_CACHE_SIZE);
s->cluster_cache = g_malloc(s->cluster_size);
/* one more sector for decompressed data alignment */
QLIST_INIT(&s->cluster_allocs);
/* read qcow2 extensions */
- if (header.backing_file_offset) {
- ext_end = header.backing_file_offset;
- } else {
- ext_end = s->cluster_size;
- }
- if (qcow2_read_extensions(bs, header.header_length, ext_end)) {
+ if (qcow2_read_extensions(bs, header.header_length, ext_end, NULL)) {
ret = -EINVAL;
goto fail;
}
goto fail;
}
+ /* Clear unknown autoclear feature bits */
+ if (!bs->read_only && s->autoclear_features != 0) {
+ s->autoclear_features = 0;
+ ret = qcow2_update_header(bs);
+ if (ret < 0) {
+ goto fail;
+ }
+ }
+
/* Initialise locks */
qemu_co_mutex_init(&s->lock);
+ /* Repair image if dirty */
+ if (!(flags & BDRV_O_CHECK) && !bs->read_only &&
+ (s->incompatible_features & QCOW2_INCOMPAT_DIRTY)) {
+ BdrvCheckResult result = {0};
+
+ ret = qcow2_check(bs, &result, BDRV_FIX_ERRORS);
+ if (ret < 0) {
+ goto fail;
+ }
+ }
+
#ifdef DEBUG_ALLOC
{
BdrvCheckResult result = {0};
- qcow2_check_refcounts(bs, &result);
+ qcow2_check_refcounts(bs, &result, 0);
}
#endif
return ret;
return 0;
}
+/* We have nothing to do for QCOW2 reopen, stubs just return
+ * success */
+static int qcow2_reopen_prepare(BDRVReopenState *state,
+ BlockReopenQueue *queue, Error **errp)
+{
+ return 0;
+}
+
static int coroutine_fn qcow2_co_is_allocated(BlockDriverState *bs,
int64_t sector_num, int nb_sectors, int *pnum)
{
else
n1 = bs->total_sectors - sector_num;
- qemu_iovec_memset_skip(qiov, 0, 512 * (nb_sectors - n1), 512 * n1);
+ qemu_iovec_memset(qiov, 512 * n1, 0, 512 * (nb_sectors - n1));
return n1;
}
index_in_cluster = sector_num & (s->cluster_sectors - 1);
qemu_iovec_reset(&hd_qiov);
- qemu_iovec_copy(&hd_qiov, qiov, bytes_done,
+ qemu_iovec_concat(&hd_qiov, qiov, bytes_done,
cur_nr_sectors * 512);
switch (ret) {
}
} else {
/* Note: in this case, no need to wait */
- qemu_iovec_memset(&hd_qiov, 0, 512 * cur_nr_sectors);
+ qemu_iovec_memset(&hd_qiov, 0, 0, 512 * cur_nr_sectors);
+ }
+ break;
+
+ case QCOW2_CLUSTER_ZERO:
+ if (s->qcow_version < 3) {
+ ret = -EIO;
+ goto fail;
}
+ qemu_iovec_memset(&hd_qiov, 0, 0, 512 * cur_nr_sectors);
break;
case QCOW2_CLUSTER_COMPRESSED:
goto fail;
}
- qemu_iovec_from_buffer(&hd_qiov,
+ qemu_iovec_from_buf(&hd_qiov, 0,
s->cluster_cache + index_in_cluster * 512,
512 * cur_nr_sectors);
break;
if (s->crypt_method) {
qcow2_encrypt_sectors(s, sector_num, cluster_data,
cluster_data, cur_nr_sectors, 0, &s->aes_decrypt_key);
- qemu_iovec_reset(&hd_qiov);
- qemu_iovec_copy(&hd_qiov, qiov, bytes_done,
- cur_nr_sectors * 512);
- qemu_iovec_from_buffer(&hd_qiov, cluster_data,
- 512 * cur_nr_sectors);
+ qemu_iovec_from_buf(qiov, bytes_done,
+ cluster_data, 512 * cur_nr_sectors);
}
break;
goto fail;
}
+ if (l2meta.nb_clusters > 0 &&
+ (s->compatible_features & QCOW2_COMPAT_LAZY_REFCOUNTS)) {
+ qcow2_mark_dirty(bs);
+ }
+
cluster_offset = l2meta.cluster_offset;
assert((cluster_offset & 511) == 0);
qemu_iovec_reset(&hd_qiov);
- qemu_iovec_copy(&hd_qiov, qiov, bytes_done,
+ qemu_iovec_concat(&hd_qiov, qiov, bytes_done,
cur_nr_sectors * 512);
if (s->crypt_method) {
assert(hd_qiov.size <=
QCOW_MAX_CRYPT_CLUSTERS * s->cluster_size);
- qemu_iovec_to_buffer(&hd_qiov, cluster_data);
+ qemu_iovec_to_buf(&hd_qiov, 0, cluster_data, hd_qiov.size);
qcow2_encrypt_sectors(s, sector_num, cluster_data,
cluster_data, cur_nr_sectors, 1, &s->aes_encrypt_key);
qcow2_cache_flush(bs, s->l2_table_cache);
qcow2_cache_flush(bs, s->refcount_block_cache);
+ qcow2_mark_clean(bs);
+
qcow2_cache_destroy(bs, s->l2_table_cache);
qcow2_cache_destroy(bs, s->refcount_block_cache);
ret = sizeof(*header);
break;
default:
- return -EINVAL;
+ ret = -EINVAL;
+ goto fail;
}
buf += ret;
buflen -= ret;
}
+ /* Feature table */
+ Qcow2Feature features[] = {
+ {
+ .type = QCOW2_FEAT_TYPE_INCOMPATIBLE,
+ .bit = QCOW2_INCOMPAT_DIRTY_BITNR,
+ .name = "dirty bit",
+ },
+ {
+ .type = QCOW2_FEAT_TYPE_COMPATIBLE,
+ .bit = QCOW2_COMPAT_LAZY_REFCOUNTS_BITNR,
+ .name = "lazy refcounts",
+ },
+ };
+
+ ret = header_ext_add(buf, QCOW2_EXT_MAGIC_FEATURE_TABLE,
+ features, sizeof(features), buflen);
+ if (ret < 0) {
+ goto fail;
+ }
+ buf += ret;
+ buflen -= ret;
+
/* Keep unknown header extensions */
QLIST_FOREACH(uext, &s->unknown_header_ext, next) {
ret = header_ext_add(buf, uext->magic, uext->data, uext->len, buflen);
goto fail;
}
+ /* Using strncpy is ok here, since buf is not NUL-terminated. */
strncpy(buf, bs->backing_file, buflen);
header->backing_file_offset = cpu_to_be64(buf - ((char*) header));
static int qcow2_change_backing_file(BlockDriverState *bs,
const char *backing_file, const char *backing_fmt)
{
- /* Backing file format doesn't make sense without a backing file */
- if (backing_fmt && !backing_file) {
- return -EINVAL;
- }
-
pstrcpy(bs->backing_file, sizeof(bs->backing_file), backing_file ?: "");
pstrcpy(bs->backing_format, sizeof(bs->backing_format), backing_fmt ?: "");
header.crypt_method = cpu_to_be32(QCOW_CRYPT_NONE);
}
+ if (flags & BLOCK_FLAG_LAZY_REFCOUNTS) {
+ header.compatible_features |=
+ cpu_to_be64(QCOW2_COMPAT_LAZY_REFCOUNTS);
+ }
+
ret = bdrv_pwrite(bs, 0, &header, sizeof(header));
if (ret < 0) {
goto out;
/* And if we're supposed to preallocate metadata, do that now */
if (prealloc) {
+ BDRVQcowState *s = bs->opaque;
+ qemu_co_mutex_lock(&s->lock);
ret = preallocate(bs);
+ qemu_co_mutex_unlock(&s->lock);
if (ret < 0) {
goto out;
}
options->value.s);
return -EINVAL;
}
+ } else if (!strcmp(options->name, BLOCK_OPT_LAZY_REFCOUNTS)) {
+ flags |= options->value.n ? BLOCK_FLAG_LAZY_REFCOUNTS : 0;
}
options++;
}
return -EINVAL;
}
+ if (version < 3 && (flags & BLOCK_FLAG_LAZY_REFCOUNTS)) {
+ fprintf(stderr, "Lazy refcounts only supported with compatibility "
+ "level 1.1 and above (use compat=1.1 or greater)\n");
+ return -EINVAL;
+ }
+
return qcow2_create2(filename, sectors, backing_file, backing_fmt, flags,
cluster_size, prealloc, options, version);
}
return 0;
}
+static coroutine_fn int qcow2_co_write_zeroes(BlockDriverState *bs,
+ int64_t sector_num, int nb_sectors)
+{
+ int ret;
+ BDRVQcowState *s = bs->opaque;
+
+ /* Emulate misaligned zero writes */
+ if (sector_num % s->cluster_sectors || nb_sectors % s->cluster_sectors) {
+ return -ENOTSUP;
+ }
+
+ /* Whatever is left can use real zero clusters */
+ qemu_co_mutex_lock(&s->lock);
+ ret = qcow2_zero_clusters(bs, sector_num << BDRV_SECTOR_BITS,
+ nb_sectors);
+ qemu_co_mutex_unlock(&s->lock);
+
+ return ret;
+}
+
static coroutine_fn int qcow2_co_discard(BlockDriverState *bs,
int64_t sector_num, int nb_sectors)
{
return ret;
}
- ret = qcow2_cache_flush(bs, s->refcount_block_cache);
- if (ret < 0) {
- qemu_co_mutex_unlock(&s->lock);
- return ret;
+ if (qcow2_need_accurate_refcounts(s)) {
+ ret = qcow2_cache_flush(bs, s->refcount_block_cache);
+ if (ret < 0) {
+ qemu_co_mutex_unlock(&s->lock);
+ return ret;
+ }
}
qemu_co_mutex_unlock(&s->lock);
return 0;
}
-
-static int qcow2_check(BlockDriverState *bs, BdrvCheckResult *result)
-{
- return qcow2_check_refcounts(bs, result);
-}
-
#if 0
static void dump_refcounts(BlockDriverState *bs)
{
.type = OPT_STRING,
.help = "Preallocation mode (allowed values: off, metadata)"
},
+ {
+ .name = BLOCK_OPT_LAZY_REFCOUNTS,
+ .type = OPT_FLAG,
+ .help = "Postpone refcount updates",
+ },
{ NULL }
};
.bdrv_probe = qcow2_probe,
.bdrv_open = qcow2_open,
.bdrv_close = qcow2_close,
+ .bdrv_reopen_prepare = qcow2_reopen_prepare,
.bdrv_create = qcow2_create,
.bdrv_co_is_allocated = qcow2_co_is_allocated,
.bdrv_set_key = qcow2_set_key,
.bdrv_co_writev = qcow2_co_writev,
.bdrv_co_flush_to_os = qcow2_co_flush_to_os,
+ .bdrv_co_write_zeroes = qcow2_co_write_zeroes,
.bdrv_co_discard = qcow2_co_discard,
.bdrv_truncate = qcow2_truncate,
.bdrv_write_compressed = qcow2_write_compressed,
projects / qemu.git / blobdiff