#include "qemu/error-report.h"
#include "qapi/qmp/qerror.h"
#include "qapi/qmp/qbool.h"
-#include "qapi/util.h"
#include "qapi/qmp/types.h"
#include "qapi-event.h"
#include "trace.h"
}
if (!(s->autoclear_features & QCOW2_AUTOCLEAR_BITMAPS)) {
- error_report("WARNING: a program lacking bitmap support "
- "modified this file, so all bitmaps are now "
- "considered inconsistent. Some clusters may be "
- "leaked, run 'qemu-img check -r' on the image "
+ warn_report("a program lacking bitmap support "
+ "modified this file, so all bitmaps are now "
+ "considered inconsistent");
+ error_printf("Some clusters may be leaked, "
+ "run 'qemu-img check -r' on the image "
"file to fix.");
if (need_update_header != NULL) {
/* Updating is needed to drop invalid bitmap extension. */
goto fail;
}
- s->cluster_cache = g_malloc(s->cluster_size);
- /* one more sector for decompressed data alignment */
- s->cluster_data = qemu_try_blockalign(bs->file->bs, QCOW_MAX_CRYPT_CLUSTERS
- * s->cluster_size + 512);
- if (s->cluster_data == NULL) {
- error_setg(errp, "Could not allocate temporary cluster buffer");
- ret = -ENOMEM;
- goto fail;
- }
-
s->cluster_cache_offset = -1;
s->flags = flags;
if (s->refcount_block_cache) {
qcow2_cache_destroy(bs, s->refcount_block_cache);
}
- g_free(s->cluster_cache);
- qemu_vfree(s->cluster_data);
qcrypto_block_free(s->crypto);
qapi_free_QCryptoBlockOpenOptions(s->crypto_opts);
return ret;
assert(s->crypto);
assert((offset & (BDRV_SECTOR_SIZE - 1)) == 0);
assert((cur_bytes & (BDRV_SECTOR_SIZE - 1)) == 0);
- Error *err = NULL;
if (qcrypto_block_decrypt(s->crypto,
(s->crypt_physical_offset ?
cluster_offset + offset_in_cluster :
- offset) >> BDRV_SECTOR_BITS,
+ offset),
cluster_data,
cur_bytes,
- &err) < 0) {
- error_free(err);
+ NULL) < 0) {
ret = -EIO;
goto fail;
}
qemu_iovec_concat(&hd_qiov, qiov, bytes_done, cur_bytes);
if (bs->encrypted) {
- Error *err = NULL;
assert(s->crypto);
if (!cluster_data) {
cluster_data = qemu_try_blockalign(bs->file->bs,
if (qcrypto_block_encrypt(s->crypto,
(s->crypt_physical_offset ?
cluster_offset + offset_in_cluster :
- offset) >> BDRV_SECTOR_BITS,
+ offset),
cluster_data,
- cur_bytes, &err) < 0) {
- error_free(err);
+ cur_bytes, NULL) < 0) {
ret = -EIO;
goto fail;
}
ret = 0;
fail:
- qemu_co_mutex_unlock(&s->lock);
-
while (l2meta != NULL) {
QCowL2Meta *next;
l2meta = next;
}
+ qemu_co_mutex_unlock(&s->lock);
+
qemu_iovec_destroy(&hd_qiov);
qemu_vfree(cluster_data);
trace_qcow2_writev_done_req(qemu_coroutine_self(), ret);
int ret, result = 0;
Error *local_err = NULL;
+ qcow2_store_persistent_dirty_bitmaps(bs, &local_err);
+ if (local_err != NULL) {
+ result = -EINVAL;
+ error_report_err(local_err);
+ error_report("Persistent bitmaps are lost for node '%s'",
+ bdrv_get_device_or_node_name(bs));
+ }
+
ret = qcow2_cache_flush(bs, s->l2_table_cache);
if (ret) {
result = ret;
strerror(-ret));
}
- qcow2_store_persistent_dirty_bitmaps(bs, &local_err);
- if (local_err != NULL) {
- result = -EINVAL;
- error_report_err(local_err);
- error_report("Persistent bitmaps are lost for node '%s'",
- bdrv_get_device_or_node_name(bs));
- }
-
if (result == 0) {
qcow2_mark_clean(bs);
}
}
-static int preallocate(BlockDriverState *bs)
+/**
+ * Preallocates metadata structures for data clusters between @offset (in the
+ * guest disk) and @new_length (which is thus generally the new guest disk
+ * size).
+ *
+ * Returns: 0 on success, -errno on failure.
+ */
+static int preallocate(BlockDriverState *bs,
+ uint64_t offset, uint64_t new_length)
{
+ BDRVQcow2State *s = bs->opaque;
uint64_t bytes;
- uint64_t offset;
uint64_t host_offset = 0;
unsigned int cur_bytes;
int ret;
QCowL2Meta *meta;
- bytes = bdrv_getlength(bs);
- offset = 0;
+ if (qemu_in_coroutine()) {
+ qemu_co_mutex_lock(&s->lock);
+ }
+
+ assert(offset <= new_length);
+ bytes = new_length - offset;
while (bytes) {
cur_bytes = MIN(bytes, INT_MAX);
ret = qcow2_alloc_cluster_offset(bs, offset, &cur_bytes,
&host_offset, &meta);
if (ret < 0) {
- return ret;
+ goto done;
}
while (meta) {
if (ret < 0) {
qcow2_free_any_clusters(bs, meta->alloc_offset,
meta->nb_clusters, QCOW2_DISCARD_NEVER);
- return ret;
+ goto done;
}
/* There are no dependent requests, but we need to remove our
ret = bdrv_pwrite(bs->file, (host_offset + cur_bytes) - 1,
&data, 1);
if (ret < 0) {
- return ret;
+ goto done;
}
}
- return 0;
+ ret = 0;
+
+done:
+ if (qemu_in_coroutine()) {
+ qemu_co_mutex_unlock(&s->lock);
+ }
+ return ret;
+}
+
+/* qcow2_refcount_metadata_size:
+ * @clusters: number of clusters to refcount (including data and L1/L2 tables)
+ * @cluster_size: size of a cluster, in bytes
+ * @refcount_order: refcount bits power-of-2 exponent
+ * @generous_increase: allow for the refcount table to be 1.5x as large as it
+ * needs to be
+ *
+ * Returns: Number of bytes required for refcount blocks and table metadata.
+ */
+int64_t qcow2_refcount_metadata_size(int64_t clusters, size_t cluster_size,
+ int refcount_order, bool generous_increase,
+ uint64_t *refblock_count)
+{
+ /*
+ * Every host cluster is reference-counted, including metadata (even
+ * refcount metadata is recursively included).
+ *
+ * An accurate formula for the size of refcount metadata size is difficult
+ * to derive. An easier method of calculation is finding the fixed point
+ * where no further refcount blocks or table clusters are required to
+ * reference count every cluster.
+ */
+ int64_t blocks_per_table_cluster = cluster_size / sizeof(uint64_t);
+ int64_t refcounts_per_block = cluster_size * 8 / (1 << refcount_order);
+ int64_t table = 0; /* number of refcount table clusters */
+ int64_t blocks = 0; /* number of refcount block clusters */
+ int64_t last;
+ int64_t n = 0;
+
+ do {
+ last = n;
+ blocks = DIV_ROUND_UP(clusters + table + blocks, refcounts_per_block);
+ table = DIV_ROUND_UP(blocks, blocks_per_table_cluster);
+ n = clusters + blocks + table;
+
+ if (n == last && generous_increase) {
+ clusters += DIV_ROUND_UP(table, 2);
+ n = 0; /* force another loop */
+ generous_increase = false;
+ }
+ } while (n != last);
+
+ if (refblock_count) {
+ *refblock_count = blocks;
+ }
+
+ return (blocks + table) * cluster_size;
}
/**
size_t cluster_size,
int refcount_order)
{
- /* Note: The following calculation does not need to be exact; if it is a
- * bit off, either some bytes will be "leaked" (which is fine) or we
- * will need to increase the file size by some bytes (which is fine,
- * too, as long as the bulk is allocated here). Therefore, using
- * floating point arithmetic is fine. */
int64_t meta_size = 0;
- uint64_t nreftablee, nrefblocke, nl1e, nl2e, refblock_count;
+ uint64_t nl1e, nl2e;
int64_t aligned_total_size = align_offset(total_size, cluster_size);
- int cluster_bits = ctz32(cluster_size);
- int refblock_bits, refblock_size;
- /* refcount entry size in bytes */
- double rces = (1 << refcount_order) / 8.;
-
- /* see qcow2_open() */
- refblock_bits = cluster_bits - (refcount_order - 3);
- refblock_size = 1 << refblock_bits;
/* header: 1 cluster */
meta_size += cluster_size;
nl1e = align_offset(nl1e, cluster_size / sizeof(uint64_t));
meta_size += nl1e * sizeof(uint64_t);
- /* total size of refcount blocks
- *
- * note: every host cluster is reference-counted, including metadata
- * (even refcount blocks are recursively included).
- * Let:
- * a = total_size (this is the guest disk size)
- * m = meta size not including refcount blocks and refcount tables
- * c = cluster size
- * y1 = number of refcount blocks entries
- * y2 = meta size including everything
- * rces = refcount entry size in bytes
- * then,
- * y1 = (y2 + a)/c
- * y2 = y1 * rces + y1 * rces * sizeof(u64) / c + m
- * we can get y1:
- * y1 = (a + m) / (c - rces - rces * sizeof(u64) / c)
- */
- nrefblocke = (aligned_total_size + meta_size + cluster_size)
- / (cluster_size - rces - rces * sizeof(uint64_t)
- / cluster_size);
- refblock_count = DIV_ROUND_UP(nrefblocke, refblock_size);
- meta_size += refblock_count * cluster_size;
-
- /* total size of refcount tables */
- nreftablee = align_offset(refblock_count,
- cluster_size / sizeof(uint64_t));
- meta_size += nreftablee * sizeof(uint64_t);
+ /* total size of refcount table and blocks */
+ meta_size += qcow2_refcount_metadata_size(
+ (meta_size + aligned_total_size) / cluster_size,
+ cluster_size, refcount_order, false, NULL);
return meta_size + aligned_total_size;
}
-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, PreallocMode prealloc,
- QemuOpts *opts, int version, int refcount_order,
- const char *encryptfmt, Error **errp)
+static size_t qcow2_opt_get_cluster_size_del(QemuOpts *opts, Error **errp)
{
+ size_t cluster_size;
int cluster_bits;
- QDict *options;
- /* Calculate cluster_bits */
+ cluster_size = qemu_opt_get_size_del(opts, BLOCK_OPT_CLUSTER_SIZE,
+ DEFAULT_CLUSTER_SIZE);
cluster_bits = ctz32(cluster_size);
if (cluster_bits < MIN_CLUSTER_BITS || cluster_bits > MAX_CLUSTER_BITS ||
(1 << cluster_bits) != cluster_size)
{
error_setg(errp, "Cluster size must be a power of two between %d and "
"%dk", 1 << MIN_CLUSTER_BITS, 1 << (MAX_CLUSTER_BITS - 10));
- return -EINVAL;
+ return 0;
+ }
+ return cluster_size;
+}
+
+static int qcow2_opt_get_version_del(QemuOpts *opts, Error **errp)
+{
+ char *buf;
+ int ret;
+
+ buf = qemu_opt_get_del(opts, BLOCK_OPT_COMPAT_LEVEL);
+ if (!buf) {
+ ret = 3; /* default */
+ } else if (!strcmp(buf, "0.10")) {
+ ret = 2;
+ } else if (!strcmp(buf, "1.1")) {
+ ret = 3;
+ } else {
+ error_setg(errp, "Invalid compatibility level: '%s'", buf);
+ ret = -EINVAL;
+ }
+ g_free(buf);
+ return ret;
+}
+
+static uint64_t qcow2_opt_get_refcount_bits_del(QemuOpts *opts, int version,
+ Error **errp)
+{
+ uint64_t refcount_bits;
+
+ refcount_bits = qemu_opt_get_number_del(opts, BLOCK_OPT_REFCOUNT_BITS, 16);
+ if (refcount_bits > 64 || !is_power_of_2(refcount_bits)) {
+ error_setg(errp, "Refcount width must be a power of two and may not "
+ "exceed 64 bits");
+ return 0;
+ }
+
+ if (version < 3 && refcount_bits != 16) {
+ error_setg(errp, "Different refcount widths than 16 bits require "
+ "compatibility level 1.1 or above (use compat=1.1 or "
+ "greater)");
+ return 0;
}
+ return refcount_bits;
+}
+
+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, PreallocMode prealloc,
+ QemuOpts *opts, int version, int refcount_order,
+ const char *encryptfmt, Error **errp)
+{
+ QDict *options;
+
/*
* Open the image file and write a minimal qcow2 header.
*
int64_t prealloc_size =
qcow2_calc_prealloc_size(total_size, cluster_size, refcount_order);
qemu_opt_set_number(opts, BLOCK_OPT_SIZE, prealloc_size, &error_abort);
- qemu_opt_set(opts, BLOCK_OPT_PREALLOC, PreallocMode_lookup[prealloc],
+ qemu_opt_set(opts, BLOCK_OPT_PREALLOC, PreallocMode_str(prealloc),
&error_abort);
}
*header = (QCowHeader) {
.magic = cpu_to_be32(QCOW_MAGIC),
.version = cpu_to_be32(version),
- .cluster_bits = cpu_to_be32(cluster_bits),
+ .cluster_bits = cpu_to_be32(ctz32(cluster_size)),
.size = cpu_to_be64(0),
.l1_table_offset = cpu_to_be64(0),
.l1_size = cpu_to_be32(0),
}
/* Okay, now that we have a valid image, let's give it the right size */
- ret = blk_truncate(blk, total_size, errp);
+ ret = blk_truncate(blk, total_size, PREALLOC_MODE_OFF, errp);
if (ret < 0) {
error_prepend(errp, "Could not resize image: ");
goto out;
/* And if we're supposed to preallocate metadata, do that now */
if (prealloc != PREALLOC_MODE_OFF) {
- BDRVQcow2State *s = blk_bs(blk)->opaque;
- qemu_co_mutex_lock(&s->lock);
- ret = preallocate(blk_bs(blk));
- qemu_co_mutex_unlock(&s->lock);
+ ret = preallocate(blk_bs(blk), 0, total_size);
if (ret < 0) {
error_setg_errno(errp, -ret, "Could not preallocate metadata");
goto out;
int flags = 0;
size_t cluster_size = DEFAULT_CLUSTER_SIZE;
PreallocMode prealloc;
- int version = 3;
- uint64_t refcount_bits = 16;
+ int version;
+ uint64_t refcount_bits;
int refcount_order;
- const char *encryptfmt = NULL;
+ char *encryptfmt = NULL;
Error *local_err = NULL;
int ret;
backing_fmt = qemu_opt_get_del(opts, BLOCK_OPT_BACKING_FMT);
encryptfmt = qemu_opt_get_del(opts, BLOCK_OPT_ENCRYPT_FORMAT);
if (encryptfmt) {
- if (qemu_opt_get_del(opts, BLOCK_OPT_ENCRYPT)) {
+ if (qemu_opt_get(opts, BLOCK_OPT_ENCRYPT)) {
error_setg(errp, "Options " BLOCK_OPT_ENCRYPT " and "
BLOCK_OPT_ENCRYPT_FORMAT " are mutually exclusive");
ret = -EINVAL;
goto finish;
}
} else if (qemu_opt_get_bool_del(opts, BLOCK_OPT_ENCRYPT, false)) {
- encryptfmt = "aes";
+ encryptfmt = g_strdup("aes");
+ }
+ cluster_size = qcow2_opt_get_cluster_size_del(opts, &local_err);
+ if (local_err) {
+ error_propagate(errp, local_err);
+ ret = -EINVAL;
+ goto finish;
}
- cluster_size = qemu_opt_get_size_del(opts, BLOCK_OPT_CLUSTER_SIZE,
- DEFAULT_CLUSTER_SIZE);
buf = qemu_opt_get_del(opts, BLOCK_OPT_PREALLOC);
- prealloc = qapi_enum_parse(PreallocMode_lookup, buf,
- PREALLOC_MODE__MAX, PREALLOC_MODE_OFF,
- &local_err);
+ prealloc = qapi_enum_parse(&PreallocMode_lookup, buf,
+ PREALLOC_MODE_OFF, &local_err);
if (local_err) {
error_propagate(errp, local_err);
ret = -EINVAL;
goto finish;
}
- g_free(buf);
- buf = qemu_opt_get_del(opts, BLOCK_OPT_COMPAT_LEVEL);
- if (!buf) {
- /* keep the default */
- } else if (!strcmp(buf, "0.10")) {
- version = 2;
- } else if (!strcmp(buf, "1.1")) {
- version = 3;
- } else {
- error_setg(errp, "Invalid compatibility level: '%s'", buf);
+
+ version = qcow2_opt_get_version_del(opts, &local_err);
+ if (local_err) {
+ error_propagate(errp, local_err);
ret = -EINVAL;
goto finish;
}
goto finish;
}
- refcount_bits = qemu_opt_get_number_del(opts, BLOCK_OPT_REFCOUNT_BITS,
- refcount_bits);
- if (refcount_bits > 64 || !is_power_of_2(refcount_bits)) {
- error_setg(errp, "Refcount width must be a power of two and may not "
- "exceed 64 bits");
- ret = -EINVAL;
- goto finish;
- }
-
- if (version < 3 && refcount_bits != 16) {
- error_setg(errp, "Different refcount widths than 16 bits require "
- "compatibility level 1.1 or above (use compat=1.1 or "
- "greater)");
+ refcount_bits = qcow2_opt_get_refcount_bits_del(opts, version, &local_err);
+ if (local_err) {
+ error_propagate(errp, local_err);
ret = -EINVAL;
goto finish;
}
finish:
g_free(backing_file);
g_free(backing_fmt);
+ g_free(encryptfmt);
g_free(buf);
return ret;
}
-static bool is_zero_sectors(BlockDriverState *bs, int64_t start,
- uint32_t count)
+static bool is_zero(BlockDriverState *bs, int64_t offset, int64_t bytes)
{
- int nr;
- BlockDriverState *file;
- int64_t res;
+ int64_t nr;
+ int res;
+ int64_t start;
+
+ /* TODO: Widening to sector boundaries should only be needed as
+ * long as we can't query finer granularity. */
+ start = QEMU_ALIGN_DOWN(offset, BDRV_SECTOR_SIZE);
+ bytes = QEMU_ALIGN_UP(offset + bytes, BDRV_SECTOR_SIZE) - start;
- if (start + count > bs->total_sectors) {
- count = bs->total_sectors - start;
+ /* Clamp to image length, before checking status of underlying sectors */
+ if (start + bytes > bs->total_sectors * BDRV_SECTOR_SIZE) {
+ bytes = bs->total_sectors * BDRV_SECTOR_SIZE - start;
}
- if (!count) {
+ if (!bytes) {
return true;
}
- res = bdrv_get_block_status_above(bs, NULL, start, count,
- &nr, &file);
- return res >= 0 && (res & BDRV_BLOCK_ZERO) && nr == count;
+ res = bdrv_block_status_above(bs, NULL, start, bytes, &nr, NULL, NULL);
+ return res >= 0 && (res & BDRV_BLOCK_ZERO) && nr == bytes;
}
static coroutine_fn int qcow2_co_pwrite_zeroes(BlockDriverState *bs,
}
if (head || tail) {
- int64_t cl_start = (offset - head) >> BDRV_SECTOR_BITS;
uint64_t off;
unsigned int nr;
assert(head + bytes <= s->cluster_size);
/* check whether remainder of cluster already reads as zero */
- if (!(is_zero_sectors(bs, cl_start,
- DIV_ROUND_UP(head, BDRV_SECTOR_SIZE)) &&
- is_zero_sectors(bs, (offset + bytes) >> BDRV_SECTOR_BITS,
- DIV_ROUND_UP(-tail & (s->cluster_size - 1),
- BDRV_SECTOR_SIZE)))) {
+ if (!(is_zero(bs, offset - head, head) &&
+ is_zero(bs, offset + bytes,
+ tail ? s->cluster_size - tail : 0))) {
return -ENOTSUP;
}
qemu_co_mutex_lock(&s->lock);
/* We can have new write after previous check */
- offset = cl_start << BDRV_SECTOR_BITS;
+ offset = QEMU_ALIGN_DOWN(offset, s->cluster_size);
bytes = s->cluster_size;
nr = s->cluster_size;
ret = qcow2_get_cluster_offset(bs, offset, &nr, &off);
return ret;
}
-static int qcow2_truncate(BlockDriverState *bs, int64_t offset, Error **errp)
+static int qcow2_truncate(BlockDriverState *bs, int64_t offset,
+ PreallocMode prealloc, Error **errp)
{
BDRVQcow2State *s = bs->opaque;
+ uint64_t old_length;
int64_t new_l1_size;
int ret;
+ if (prealloc != PREALLOC_MODE_OFF && prealloc != PREALLOC_MODE_METADATA &&
+ prealloc != PREALLOC_MODE_FALLOC && prealloc != PREALLOC_MODE_FULL)
+ {
+ error_setg(errp, "Unsupported preallocation mode '%s'",
+ PreallocMode_str(prealloc));
+ return -ENOTSUP;
+ }
+
if (offset & 511) {
error_setg(errp, "The new size must be a multiple of 512");
return -EINVAL;
return -ENOTSUP;
}
- /* shrinking is currently not supported */
- if (offset < bs->total_sectors * 512) {
- error_setg(errp, "qcow2 doesn't support shrinking images yet");
- return -ENOTSUP;
+ old_length = bs->total_sectors * 512;
+ new_l1_size = size_to_l1(s, offset);
+
+ if (offset < old_length) {
+ int64_t last_cluster, old_file_size;
+ if (prealloc != PREALLOC_MODE_OFF) {
+ error_setg(errp,
+ "Preallocation can't be used for shrinking an image");
+ return -EINVAL;
+ }
+
+ ret = qcow2_cluster_discard(bs, ROUND_UP(offset, s->cluster_size),
+ old_length - ROUND_UP(offset,
+ s->cluster_size),
+ QCOW2_DISCARD_ALWAYS, true);
+ if (ret < 0) {
+ error_setg_errno(errp, -ret, "Failed to discard cropped clusters");
+ return ret;
+ }
+
+ ret = qcow2_shrink_l1_table(bs, new_l1_size);
+ if (ret < 0) {
+ error_setg_errno(errp, -ret,
+ "Failed to reduce the number of L2 tables");
+ return ret;
+ }
+
+ ret = qcow2_shrink_reftable(bs);
+ if (ret < 0) {
+ error_setg_errno(errp, -ret,
+ "Failed to discard unused refblocks");
+ return ret;
+ }
+
+ old_file_size = bdrv_getlength(bs->file->bs);
+ if (old_file_size < 0) {
+ error_setg_errno(errp, -old_file_size,
+ "Failed to inquire current file length");
+ return old_file_size;
+ }
+ last_cluster = qcow2_get_last_cluster(bs, old_file_size);
+ if (last_cluster < 0) {
+ error_setg_errno(errp, -last_cluster,
+ "Failed to find the last cluster");
+ return last_cluster;
+ }
+ if ((last_cluster + 1) * s->cluster_size < old_file_size) {
+ ret = bdrv_truncate(bs->file, (last_cluster + 1) * s->cluster_size,
+ PREALLOC_MODE_OFF, NULL);
+ if (ret < 0) {
+ warn_report("Failed to truncate the tail of the image: %s",
+ strerror(-ret));
+ ret = 0;
+ }
+ }
+ } else {
+ ret = qcow2_grow_l1_table(bs, new_l1_size, true);
+ if (ret < 0) {
+ error_setg_errno(errp, -ret, "Failed to grow the L1 table");
+ return ret;
+ }
}
- new_l1_size = size_to_l1(s, offset);
- ret = qcow2_grow_l1_table(bs, new_l1_size, true);
- if (ret < 0) {
- error_setg_errno(errp, -ret, "Failed to grow the L1 table");
- return ret;
+ switch (prealloc) {
+ case PREALLOC_MODE_OFF:
+ break;
+
+ case PREALLOC_MODE_METADATA:
+ ret = preallocate(bs, old_length, offset);
+ if (ret < 0) {
+ error_setg_errno(errp, -ret, "Preallocation failed");
+ return ret;
+ }
+ break;
+
+ case PREALLOC_MODE_FALLOC:
+ case PREALLOC_MODE_FULL:
+ {
+ int64_t allocation_start, host_offset, guest_offset;
+ int64_t clusters_allocated;
+ int64_t old_file_size, new_file_size;
+ uint64_t nb_new_data_clusters, nb_new_l2_tables;
+
+ old_file_size = bdrv_getlength(bs->file->bs);
+ if (old_file_size < 0) {
+ error_setg_errno(errp, -old_file_size,
+ "Failed to inquire current file length");
+ return old_file_size;
+ }
+
+ nb_new_data_clusters = DIV_ROUND_UP(offset - old_length,
+ s->cluster_size);
+
+ /* This is an overestimation; we will not actually allocate space for
+ * these in the file but just make sure the new refcount structures are
+ * able to cover them so we will not have to allocate new refblocks
+ * while entering the data blocks in the potentially new L2 tables.
+ * (We do not actually care where the L2 tables are placed. Maybe they
+ * are already allocated or they can be placed somewhere before
+ * @old_file_size. It does not matter because they will be fully
+ * allocated automatically, so they do not need to be covered by the
+ * preallocation. All that matters is that we will not have to allocate
+ * new refcount structures for them.) */
+ nb_new_l2_tables = DIV_ROUND_UP(nb_new_data_clusters,
+ s->cluster_size / sizeof(uint64_t));
+ /* The cluster range may not be aligned to L2 boundaries, so add one L2
+ * table for a potential head/tail */
+ nb_new_l2_tables++;
+
+ allocation_start = qcow2_refcount_area(bs, old_file_size,
+ nb_new_data_clusters +
+ nb_new_l2_tables,
+ true, 0, 0);
+ if (allocation_start < 0) {
+ error_setg_errno(errp, -allocation_start,
+ "Failed to resize refcount structures");
+ return allocation_start;
+ }
+
+ clusters_allocated = qcow2_alloc_clusters_at(bs, allocation_start,
+ nb_new_data_clusters);
+ if (clusters_allocated < 0) {
+ error_setg_errno(errp, -clusters_allocated,
+ "Failed to allocate data clusters");
+ return -clusters_allocated;
+ }
+
+ assert(clusters_allocated == nb_new_data_clusters);
+
+ /* Allocate the data area */
+ new_file_size = allocation_start +
+ nb_new_data_clusters * s->cluster_size;
+ ret = bdrv_truncate(bs->file, new_file_size, prealloc, errp);
+ if (ret < 0) {
+ error_prepend(errp, "Failed to resize underlying file: ");
+ qcow2_free_clusters(bs, allocation_start,
+ nb_new_data_clusters * s->cluster_size,
+ QCOW2_DISCARD_OTHER);
+ return ret;
+ }
+
+ /* Create the necessary L2 entries */
+ host_offset = allocation_start;
+ guest_offset = old_length;
+ while (nb_new_data_clusters) {
+ int64_t guest_cluster = guest_offset >> s->cluster_bits;
+ int64_t nb_clusters = MIN(nb_new_data_clusters,
+ s->l2_size - guest_cluster % s->l2_size);
+ QCowL2Meta allocation = {
+ .offset = guest_offset,
+ .alloc_offset = host_offset,
+ .nb_clusters = nb_clusters,
+ };
+ qemu_co_queue_init(&allocation.dependent_requests);
+
+ ret = qcow2_alloc_cluster_link_l2(bs, &allocation);
+ if (ret < 0) {
+ error_setg_errno(errp, -ret, "Failed to update L2 tables");
+ qcow2_free_clusters(bs, host_offset,
+ nb_new_data_clusters * s->cluster_size,
+ QCOW2_DISCARD_OTHER);
+ return ret;
+ }
+
+ guest_offset += nb_clusters * s->cluster_size;
+ host_offset += nb_clusters * s->cluster_size;
+ nb_new_data_clusters -= nb_clusters;
+ }
+ break;
+ }
+
+ default:
+ g_assert_not_reached();
+ }
+
+ if (prealloc != PREALLOC_MODE_OFF) {
+ /* Flush metadata before actually changing the image size */
+ ret = bdrv_flush(bs);
+ if (ret < 0) {
+ error_setg_errno(errp, -ret,
+ "Failed to flush the preallocated area to disk");
+ return ret;
+ }
}
/* write updated header.size */
z_stream strm;
int ret, out_len;
uint8_t *buf, *out_buf;
- uint64_t cluster_offset;
+ int64_t cluster_offset;
if (bytes == 0) {
/* align end of file to a sector boundary to ease reading with
sector based I/Os */
cluster_offset = bdrv_getlength(bs->file->bs);
- return bdrv_truncate(bs->file, cluster_offset, NULL);
+ if (cluster_offset < 0) {
+ return cluster_offset;
+ }
+ return bdrv_truncate(bs->file, cluster_offset, PREALLOC_MODE_OFF, NULL);
}
buf = qemu_blockalign(bs, s->cluster_size);
}
ret = bdrv_truncate(bs->file, (3 + l1_clusters) * s->cluster_size,
- &local_err);
+ PREALLOC_MODE_OFF, &local_err);
if (ret < 0) {
error_report_err(local_err);
goto fail;
return 0;
}
+static BlockMeasureInfo *qcow2_measure(QemuOpts *opts, BlockDriverState *in_bs,
+ Error **errp)
+{
+ Error *local_err = NULL;
+ BlockMeasureInfo *info;
+ uint64_t required = 0; /* bytes that contribute to required size */
+ uint64_t virtual_size; /* disk size as seen by guest */
+ uint64_t refcount_bits;
+ uint64_t l2_tables;
+ size_t cluster_size;
+ int version;
+ char *optstr;
+ PreallocMode prealloc;
+ bool has_backing_file;
+
+ /* Parse image creation options */
+ cluster_size = qcow2_opt_get_cluster_size_del(opts, &local_err);
+ if (local_err) {
+ goto err;
+ }
+
+ version = qcow2_opt_get_version_del(opts, &local_err);
+ if (local_err) {
+ goto err;
+ }
+
+ refcount_bits = qcow2_opt_get_refcount_bits_del(opts, version, &local_err);
+ if (local_err) {
+ goto err;
+ }
+
+ optstr = qemu_opt_get_del(opts, BLOCK_OPT_PREALLOC);
+ prealloc = qapi_enum_parse(&PreallocMode_lookup, optstr,
+ PREALLOC_MODE_OFF, &local_err);
+ g_free(optstr);
+ if (local_err) {
+ goto err;
+ }
+
+ optstr = qemu_opt_get_del(opts, BLOCK_OPT_BACKING_FILE);
+ has_backing_file = !!optstr;
+ g_free(optstr);
+
+ virtual_size = align_offset(qemu_opt_get_size_del(opts, BLOCK_OPT_SIZE, 0),
+ cluster_size);
+
+ /* Check that virtual disk size is valid */
+ l2_tables = DIV_ROUND_UP(virtual_size / cluster_size,
+ cluster_size / sizeof(uint64_t));
+ if (l2_tables * sizeof(uint64_t) > QCOW_MAX_L1_SIZE) {
+ error_setg(&local_err, "The image size is too large "
+ "(try using a larger cluster size)");
+ goto err;
+ }
+
+ /* Account for input image */
+ if (in_bs) {
+ int64_t ssize = bdrv_getlength(in_bs);
+ if (ssize < 0) {
+ error_setg_errno(&local_err, -ssize,
+ "Unable to get image virtual_size");
+ goto err;
+ }
+
+ virtual_size = align_offset(ssize, cluster_size);
+
+ if (has_backing_file) {
+ /* We don't how much of the backing chain is shared by the input
+ * image and the new image file. In the worst case the new image's
+ * backing file has nothing in common with the input image. Be
+ * conservative and assume all clusters need to be written.
+ */
+ required = virtual_size;
+ } else {
+ int64_t offset;
+ int64_t pnum = 0;
+
+ for (offset = 0; offset < ssize; offset += pnum) {
+ int ret;
+
+ ret = bdrv_block_status_above(in_bs, NULL, offset,
+ ssize - offset, &pnum, NULL,
+ NULL);
+ if (ret < 0) {
+ error_setg_errno(&local_err, -ret,
+ "Unable to get block status");
+ goto err;
+ }
+
+ if (ret & BDRV_BLOCK_ZERO) {
+ /* Skip zero regions (safe with no backing file) */
+ } else if ((ret & (BDRV_BLOCK_DATA | BDRV_BLOCK_ALLOCATED)) ==
+ (BDRV_BLOCK_DATA | BDRV_BLOCK_ALLOCATED)) {
+ /* Extend pnum to end of cluster for next iteration */
+ pnum = ROUND_UP(offset + pnum, cluster_size) - offset;
+
+ /* Count clusters we've seen */
+ required += offset % cluster_size + pnum;
+ }
+ }
+ }
+ }
+
+ /* Take into account preallocation. Nothing special is needed for
+ * PREALLOC_MODE_METADATA since metadata is always counted.
+ */
+ if (prealloc == PREALLOC_MODE_FULL || prealloc == PREALLOC_MODE_FALLOC) {
+ required = virtual_size;
+ }
+
+ info = g_new(BlockMeasureInfo, 1);
+ info->fully_allocated =
+ qcow2_calc_prealloc_size(virtual_size, cluster_size,
+ ctz32(refcount_bits));
+
+ /* Remove data clusters that are not required. This overestimates the
+ * required size because metadata needed for the fully allocated file is
+ * still counted.
+ */
+ info->required = info->fully_allocated - virtual_size + required;
+ return info;
+
+err:
+ error_propagate(errp, local_err);
+ return NULL;
+}
+
static int qcow2_get_info(BlockDriverState *bs, BlockDriverInfo *bdi)
{
BDRVQcow2State *s = bs->opaque;
return spec_info;
}
-#if 0
-static void dump_refcounts(BlockDriverState *bs)
-{
- BDRVQcow2State *s = bs->opaque;
- int64_t nb_clusters, k, k1, size;
- int refcount;
-
- size = bdrv_getlength(bs->file->bs);
- nb_clusters = size_to_clusters(s, size);
- for(k = 0; k < nb_clusters;) {
- k1 = k;
- refcount = get_refcount(bs, k);
- k++;
- while (k < nb_clusters && get_refcount(bs, k) == refcount)
- k++;
- printf("%" PRId64 ": refcount=%d nb=%" PRId64 "\n", k, refcount,
- k - k1);
- }
-}
-#endif
-
static int qcow2_save_vmstate(BlockDriverState *bs, QEMUIOVector *qiov,
int64_t pos)
{
return ret;
}
- ret = blk_truncate(blk, new_size, &local_err);
+ ret = blk_truncate(blk, new_size, PREALLOC_MODE_OFF, &local_err);
blk_unref(blk);
if (ret < 0) {
error_report_err(local_err);
.bdrv_snapshot_delete = qcow2_snapshot_delete,
.bdrv_snapshot_list = qcow2_snapshot_list,
.bdrv_snapshot_load_tmp = qcow2_snapshot_load_tmp,
+ .bdrv_measure = qcow2_measure,
.bdrv_get_info = qcow2_get_info,
.bdrv_get_specific_info = qcow2_get_specific_info,
projects / qemu.git / blobdiff