* THE SOFTWARE.
*/
#include "qemu-common.h"
-#include "block_int.h"
+#include "block/block_int.h"
#include "module.h"
#include <zlib.h>
-#include "aes.h"
+#include "block/aes.h"
#include "block/qcow2.h"
#include "qemu-error.h"
-#include "qerror.h"
+#include "qapi/qmp/qerror.h"
+#include "trace.h"
/*
Differences with QCOW:
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)
{
if (buf_size >= sizeof(QCowHeader) &&
be32_to_cpu(cow_header->magic) == QCOW_MAGIC &&
- be32_to_cpu(cow_header->version) >= QCOW_VERSION)
+ be32_to_cpu(cow_header->version) >= 2)
return 100;
else
return 0;
* 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 GCC_FMT_ATTR(2, 3) report_unsupported(BlockDriverState *bs,
+ const char *fmt, ...)
+{
+ char msg[64];
+ va_list ap;
+
+ va_start(ap, fmt);
+ vsnprintf(msg, sizeof(msg), fmt, ap);
+ va_end(ap);
+
+ qerror_report(QERR_UNKNOWN_BLOCK_FORMAT_FEATURE,
+ 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.
+ */
+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) {
ret = -EINVAL;
goto fail;
}
- if (header.version != QCOW_VERSION) {
- char version[64];
- snprintf(version, sizeof(version), "QCOW version %d", header.version);
- qerror_report(QERR_UNKNOWN_BLOCK_FORMAT_FEATURE,
- bs->device_name, "qcow2", version);
+ if (header.version < 2 || header.version > 3) {
+ report_unsupported(bs, "QCOW version %d", header.version);
+ ret = -ENOTSUP;
+ goto fail;
+ }
+
+ s->qcow_version = header.version;
+
+ /* Initialise version 3 header fields */
+ if (header.version == 2) {
+ header.incompatible_features = 0;
+ header.compatible_features = 0;
+ header.autoclear_features = 0;
+ header.refcount_order = 4;
+ header.header_length = 72;
+ } else {
+ be64_to_cpus(&header.incompatible_features);
+ be64_to_cpus(&header.compatible_features);
+ be64_to_cpus(&header.autoclear_features);
+ be32_to_cpus(&header.refcount_order);
+ be32_to_cpus(&header.header_length);
+ }
+
+ if (header.header_length > sizeof(header)) {
+ s->unknown_header_fields_size = header.header_length - sizeof(header);
+ s->unknown_header_fields = g_malloc(s->unknown_header_fields_size);
+ ret = bdrv_pread(bs->file, sizeof(header), s->unknown_header_fields,
+ s->unknown_header_fields_size);
+ if (ret < 0) {
+ goto fail;
+ }
+ }
+
+ 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 & ~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;
+ }
+
+ /* Check support for various header values */
+ if (header.refcount_order != 4) {
+ report_unsupported(bs, "%d bit reference counts",
+ 1 << header.refcount_order);
ret = -ENOTSUP;
goto fail;
}
+
if (header.cluster_bits < MIN_CLUSTER_BITS ||
header.cluster_bits > MAX_CLUSTER_BITS) {
ret = -EINVAL;
}
/* 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, sizeof(header), 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;
fail:
+ g_free(s->unknown_header_fields);
cleanup_unknown_header_ext(bs);
qcow2_free_snapshots(bs);
qcow2_refcount_close(bs);
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);
- if (!cluster_offset) {
+ switch (ret) {
+ case QCOW2_CLUSTER_UNALLOCATED:
if (bs->backing_hd) {
/* read from the base image */
}
} 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;
}
- } else if (cluster_offset & QCOW_OFLAG_COMPRESSED) {
+ qemu_iovec_memset(&hd_qiov, 0, 0, 512 * cur_nr_sectors);
+ break;
+
+ case QCOW2_CLUSTER_COMPRESSED:
/* add AIO support for compressed blocks ? */
ret = qcow2_decompress_cluster(bs, cluster_offset);
if (ret < 0) {
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);
- } else {
+ break;
+
+ case QCOW2_CLUSTER_NORMAL:
if ((cluster_offset & 511) != 0) {
ret = -EIO;
goto fail;
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;
+
+ default:
+ g_assert_not_reached();
+ ret = -EIO;
+ goto fail;
}
remaining_sectors -= cur_nr_sectors;
return ret;
}
-static void run_dependent_requests(BDRVQcowState *s, QCowL2Meta *m)
-{
- /* Take the request off the list of running requests */
- if (m->nb_clusters != 0) {
- QLIST_REMOVE(m, next_in_flight);
- }
-
- /* Restart all dependent requests */
- if (!qemu_co_queue_empty(&m->dependent_requests)) {
- qemu_co_mutex_unlock(&s->lock);
- qemu_co_queue_restart_all(&m->dependent_requests);
- qemu_co_mutex_lock(&s->lock);
- }
-}
-
static coroutine_fn int qcow2_co_writev(BlockDriverState *bs,
int64_t sector_num,
int remaining_sectors,
QEMUIOVector hd_qiov;
uint64_t bytes_done = 0;
uint8_t *cluster_data = NULL;
- QCowL2Meta l2meta = {
- .nb_clusters = 0,
- };
+ QCowL2Meta *l2meta;
- qemu_co_queue_init(&l2meta.dependent_requests);
+ trace_qcow2_writev_start_req(qemu_coroutine_self(), sector_num,
+ remaining_sectors);
qemu_iovec_init(&hd_qiov, qiov->niov);
while (remaining_sectors != 0) {
+ l2meta = NULL;
+
+ trace_qcow2_writev_start_part(qemu_coroutine_self());
index_in_cluster = sector_num & (s->cluster_sectors - 1);
n_end = index_in_cluster + remaining_sectors;
if (s->crypt_method &&
}
ret = qcow2_alloc_cluster_offset(bs, sector_num << 9,
- index_in_cluster, n_end, &cur_nr_sectors, &l2meta);
+ index_in_cluster, n_end, &cur_nr_sectors, &cluster_offset, &l2meta);
if (ret < 0) {
goto fail;
}
- 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);
cur_nr_sectors * 512);
}
- BLKDBG_EVENT(bs->file, BLKDBG_WRITE_AIO);
qemu_co_mutex_unlock(&s->lock);
+ BLKDBG_EVENT(bs->file, BLKDBG_WRITE_AIO);
+ trace_qcow2_writev_data(qemu_coroutine_self(),
+ (cluster_offset >> 9) + index_in_cluster);
ret = bdrv_co_writev(bs->file,
(cluster_offset >> 9) + index_in_cluster,
cur_nr_sectors, &hd_qiov);
goto fail;
}
- ret = qcow2_alloc_cluster_link_l2(bs, &l2meta);
- if (ret < 0) {
- goto fail;
- }
+ if (l2meta != NULL) {
+ ret = qcow2_alloc_cluster_link_l2(bs, l2meta);
+ if (ret < 0) {
+ goto fail;
+ }
+
+ /* Take the request off the list of running requests */
+ if (l2meta->nb_clusters != 0) {
+ QLIST_REMOVE(l2meta, next_in_flight);
+ }
- run_dependent_requests(s, &l2meta);
+ qemu_co_mutex_unlock(&s->lock);
+ qemu_co_queue_restart_all(&l2meta->dependent_requests);
+ qemu_co_mutex_lock(&s->lock);
+
+ g_free(l2meta);
+ l2meta = NULL;
+ }
remaining_sectors -= cur_nr_sectors;
sector_num += cur_nr_sectors;
bytes_done += cur_nr_sectors * 512;
+ trace_qcow2_writev_done_part(qemu_coroutine_self(), cur_nr_sectors);
}
ret = 0;
fail:
- run_dependent_requests(s, &l2meta);
-
qemu_co_mutex_unlock(&s->lock);
+ if (l2meta != NULL) {
+ if (l2meta->nb_clusters != 0) {
+ QLIST_REMOVE(l2meta, next_in_flight);
+ }
+ qemu_co_queue_restart_all(&l2meta->dependent_requests);
+ g_free(l2meta);
+ }
+
qemu_iovec_destroy(&hd_qiov);
qemu_vfree(cluster_data);
+ trace_qcow2_writev_done_req(qemu_coroutine_self(), ret);
return ret;
}
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);
+ g_free(s->unknown_header_fields);
cleanup_unknown_header_ext(bs);
+
g_free(s->cluster_cache);
qemu_vfree(s->cluster_data);
qcow2_refcount_close(bs);
int ret;
uint64_t total_size;
uint32_t refcount_table_clusters;
+ size_t header_length;
Qcow2UnknownHeaderExtension *uext;
buf = qemu_blockalign(bs, buflen);
- memset(buf, 0, s->cluster_size);
/* Header structure */
header = (QCowHeader*) buf;
goto fail;
}
+ header_length = sizeof(*header) + s->unknown_header_fields_size;
total_size = bs->total_sectors * BDRV_SECTOR_SIZE;
refcount_table_clusters = s->refcount_table_size >> (s->cluster_bits - 3);
*header = (QCowHeader) {
+ /* Version 2 fields */
.magic = cpu_to_be32(QCOW_MAGIC),
- .version = cpu_to_be32(QCOW_VERSION),
+ .version = cpu_to_be32(s->qcow_version),
.backing_file_offset = 0,
.backing_file_size = 0,
.cluster_bits = cpu_to_be32(s->cluster_bits),
.refcount_table_clusters = cpu_to_be32(refcount_table_clusters),
.nb_snapshots = cpu_to_be32(s->nb_snapshots),
.snapshots_offset = cpu_to_be64(s->snapshots_offset),
+
+ /* Version 3 fields */
+ .incompatible_features = cpu_to_be64(s->incompatible_features),
+ .compatible_features = cpu_to_be64(s->compatible_features),
+ .autoclear_features = cpu_to_be64(s->autoclear_features),
+ .refcount_order = cpu_to_be32(3 + REFCOUNT_SHIFT),
+ .header_length = cpu_to_be32(header_length),
};
- buf += sizeof(*header);
- buflen -= sizeof(*header);
+ /* For older versions, write a shorter header */
+ switch (s->qcow_version) {
+ case 2:
+ ret = offsetof(QCowHeader, incompatible_features);
+ break;
+ case 3:
+ ret = sizeof(*header);
+ break;
+ default:
+ ret = -EINVAL;
+ goto fail;
+ }
+
+ buf += ret;
+ buflen -= ret;
+ memset(buf, 0, buflen);
+
+ /* Preserve any unknown field in the header */
+ if (s->unknown_header_fields_size) {
+ if (buflen < s->unknown_header_fields_size) {
+ ret = -ENOSPC;
+ goto fail;
+ }
+
+ memcpy(buf, s->unknown_header_fields, s->unknown_header_fields_size);
+ buf += s->unknown_header_fields_size;
+ buflen -= s->unknown_header_fields_size;
+ }
/* Backing file format header extension */
if (*bs->backing_format) {
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 ?: "");
{
uint64_t nb_sectors;
uint64_t offset;
+ uint64_t host_offset = 0;
int num;
int ret;
- QCowL2Meta meta;
+ QCowL2Meta *meta;
nb_sectors = bdrv_getlength(bs) >> 9;
offset = 0;
- qemu_co_queue_init(&meta.dependent_requests);
- meta.cluster_offset = 0;
while (nb_sectors) {
num = MIN(nb_sectors, INT_MAX >> 9);
- ret = qcow2_alloc_cluster_offset(bs, offset, 0, num, &num, &meta);
+ ret = qcow2_alloc_cluster_offset(bs, offset, 0, num, &num,
+ &host_offset, &meta);
if (ret < 0) {
return ret;
}
- ret = qcow2_alloc_cluster_link_l2(bs, &meta);
+ ret = qcow2_alloc_cluster_link_l2(bs, meta);
if (ret < 0) {
- qcow2_free_any_clusters(bs, meta.cluster_offset, meta.nb_clusters);
+ qcow2_free_any_clusters(bs, meta->alloc_offset, meta->nb_clusters);
return ret;
}
/* There are no dependent requests, but we need to remove our request
* from the list of in-flight requests */
- run_dependent_requests(bs->opaque, &meta);
+ if (meta != NULL) {
+ QLIST_REMOVE(meta, next_in_flight);
+ }
/* TODO Preallocate data if requested */
* all of the allocated clusters (otherwise we get failing reads after
* EOF). Extend the image to the last allocated sector.
*/
- if (meta.cluster_offset != 0) {
+ if (host_offset != 0) {
uint8_t buf[512];
memset(buf, 0, 512);
- ret = bdrv_write(bs->file, (meta.cluster_offset >> 9) + num - 1, buf, 1);
+ ret = bdrv_write(bs->file, (host_offset >> 9) + num - 1, buf, 1);
if (ret < 0) {
return ret;
}
static int qcow2_create2(const char *filename, int64_t total_size,
const char *backing_file, const char *backing_format,
int flags, size_t cluster_size, int prealloc,
- QEMUOptionParameter *options)
+ QEMUOptionParameter *options, int version)
{
/* Calculate cluster_bits */
int cluster_bits;
/* Write the header */
memset(&header, 0, sizeof(header));
header.magic = cpu_to_be32(QCOW_MAGIC);
- header.version = cpu_to_be32(QCOW_VERSION);
+ header.version = cpu_to_be32(version);
header.cluster_bits = cpu_to_be32(cluster_bits);
header.size = cpu_to_be64(0);
header.l1_table_offset = cpu_to_be64(0);
header.l1_size = cpu_to_be32(0);
header.refcount_table_offset = cpu_to_be64(cluster_size);
header.refcount_table_clusters = cpu_to_be32(1);
+ header.refcount_order = cpu_to_be32(3 + REFCOUNT_SHIFT);
+ header.header_length = cpu_to_be32(sizeof(header));
if (flags & BLOCK_FLAG_ENCRYPT) {
header.crypt_method = cpu_to_be32(QCOW_CRYPT_AES);
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;
}
int flags = 0;
size_t cluster_size = DEFAULT_CLUSTER_SIZE;
int prealloc = 0;
+ int version = 2;
/* Read out options */
while (options && options->name) {
options->value.s);
return -EINVAL;
}
+ } else if (!strcmp(options->name, BLOCK_OPT_COMPAT_LEVEL)) {
+ if (!options->value.s || !strcmp(options->value.s, "0.10")) {
+ version = 2;
+ } else if (!strcmp(options->value.s, "1.1")) {
+ version = 3;
+ } else {
+ fprintf(stderr, "Invalid compatibility level: '%s'\n",
+ 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);
+ cluster_size, prealloc, options, version);
}
static int qcow2_make_empty(BlockDriverState *bs)
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)
{
int ret, new_l1_size;
if (offset & 511) {
+ error_report("The new size must be a multiple of 512");
return -EINVAL;
}
/* cannot proceed if image has snapshots */
if (s->nb_snapshots) {
+ error_report("Can't resize an image which has snapshots");
return -ENOTSUP;
}
/* shrinking is currently not supported */
if (offset < bs->total_sectors * 512) {
+ error_report("qcow2 doesn't support shrinking images yet");
return -ENOTSUP;
}
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 coroutine_fn int qcow2_co_flush_to_disk(BlockDriverState *bs)
-{
- return bdrv_co_flush(bs->file);
-}
-
static int64_t qcow2_vm_state_offset(BDRVQcowState *s)
{
return (int64_t)s->l1_vm_state_index << (s->cluster_bits + s->l2_bits);
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_SIZE,
.help = "Virtual disk size"
},
+ {
+ .name = BLOCK_OPT_COMPAT_LEVEL,
+ .type = OPT_STRING,
+ .help = "Compatibility level (0.10 or 1.1)"
+ },
{
.name = BLOCK_OPT_BACKING_FILE,
.type = OPT_STRING,
.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_readv = qcow2_co_readv,
.bdrv_co_writev = qcow2_co_writev,
.bdrv_co_flush_to_os = qcow2_co_flush_to_os,
- .bdrv_co_flush_to_disk = qcow2_co_flush_to_disk,
+ .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