aboutsummaryrefslogtreecommitdiff
path: root/block
diff options
context:
space:
mode:
Diffstat (limited to 'block')
-rw-r--r--block/blkverify.c24
-rw-r--r--block/cow.c92
-rw-r--r--block/curl.c4
-rw-r--r--block/nbd.c4
-rw-r--r--block/qcow.c24
-rw-r--r--block/qcow2-cluster.c115
-rw-r--r--block/qcow2-refcount.c7
-rw-r--r--block/qcow2-snapshot.c362
-rw-r--r--block/qcow2.c33
-rw-r--r--block/qcow2.h2
-rw-r--r--block/qed-table.c24
-rw-r--r--block/qed.c75
-rw-r--r--block/raw-posix.c4
-rw-r--r--block/rbd.c6
-rw-r--r--block/sheepdog.c6
-rw-r--r--block/vdi.c89
-rw-r--r--block/vmdk.c24
-rw-r--r--block/vpc.c18
-rw-r--r--block/vvfat.c21
19 files changed, 559 insertions, 375 deletions
diff --git a/block/blkverify.c b/block/blkverify.c
index 483f3b3cfe..4ca8584b88 100644
--- a/block/blkverify.c
+++ b/block/blkverify.c
@@ -310,14 +310,10 @@ static BlockDriverAIOCB *blkverify_aio_readv(BlockDriverState *bs,
qemu_iovec_init(&acb->raw_qiov, acb->qiov->niov);
blkverify_iovec_clone(&acb->raw_qiov, qiov, acb->buf);
- if (!bdrv_aio_readv(s->test_file, sector_num, qiov, nb_sectors,
- blkverify_aio_cb, acb)) {
- blkverify_aio_cb(acb, -EIO);
- }
- if (!bdrv_aio_readv(bs->file, sector_num, &acb->raw_qiov, nb_sectors,
- blkverify_aio_cb, acb)) {
- blkverify_aio_cb(acb, -EIO);
- }
+ bdrv_aio_readv(s->test_file, sector_num, qiov, nb_sectors,
+ blkverify_aio_cb, acb);
+ bdrv_aio_readv(bs->file, sector_num, &acb->raw_qiov, nb_sectors,
+ blkverify_aio_cb, acb);
return &acb->common;
}
@@ -329,14 +325,10 @@ static BlockDriverAIOCB *blkverify_aio_writev(BlockDriverState *bs,
BlkverifyAIOCB *acb = blkverify_aio_get(bs, true, sector_num, qiov,
nb_sectors, cb, opaque);
- if (!bdrv_aio_writev(s->test_file, sector_num, qiov, nb_sectors,
- blkverify_aio_cb, acb)) {
- blkverify_aio_cb(acb, -EIO);
- }
- if (!bdrv_aio_writev(bs->file, sector_num, qiov, nb_sectors,
- blkverify_aio_cb, acb)) {
- blkverify_aio_cb(acb, -EIO);
- }
+ bdrv_aio_writev(s->test_file, sector_num, qiov, nb_sectors,
+ blkverify_aio_cb, acb);
+ bdrv_aio_writev(bs->file, sector_num, qiov, nb_sectors,
+ blkverify_aio_cb, acb);
return &acb->common;
}
diff --git a/block/cow.c b/block/cow.c
index 089d395c40..bb5927c6aa 100644
--- a/block/cow.c
+++ b/block/cow.c
@@ -64,15 +64,26 @@ static int cow_open(BlockDriverState *bs, int flags)
struct cow_header_v2 cow_header;
int bitmap_size;
int64_t size;
+ int ret;
/* see if it is a cow image */
- if (bdrv_pread(bs->file, 0, &cow_header, sizeof(cow_header)) !=
- sizeof(cow_header)) {
+ ret = bdrv_pread(bs->file, 0, &cow_header, sizeof(cow_header));
+ if (ret < 0) {
+ goto fail;
+ }
+
+ if (be32_to_cpu(cow_header.magic) != COW_MAGIC) {
+ ret = -EINVAL;
goto fail;
}
- if (be32_to_cpu(cow_header.magic) != COW_MAGIC ||
- be32_to_cpu(cow_header.version) != COW_VERSION) {
+ if (be32_to_cpu(cow_header.version) != COW_VERSION) {
+ char version[64];
+ snprintf(version, sizeof(version),
+ "COW version %d", cow_header.version);
+ qerror_report(QERR_UNKNOWN_BLOCK_FORMAT_FEATURE,
+ bs->device_name, "cow", version);
+ ret = -ENOTSUP;
goto fail;
}
@@ -88,11 +99,11 @@ static int cow_open(BlockDriverState *bs, int flags)
qemu_co_mutex_init(&s->lock);
return 0;
fail:
- return -1;
+ return ret;
}
/*
- * XXX(hch): right now these functions are extremly ineffcient.
+ * XXX(hch): right now these functions are extremely ineffcient.
* We should just read the whole bitmap we'll need in one go instead.
*/
static inline int cow_set_bit(BlockDriverState *bs, int64_t bitnum)
@@ -132,8 +143,8 @@ static inline int is_bit_set(BlockDriverState *bs, int64_t bitnum)
/* Return true if first block has been changed (ie. current version is
* in COW file). Set the number of continuous blocks for which that
* is true. */
-static int cow_is_allocated(BlockDriverState *bs, int64_t sector_num,
- int nb_sectors, int *num_same)
+static int coroutine_fn cow_co_is_allocated(BlockDriverState *bs,
+ int64_t sector_num, int nb_sectors, int *num_same)
{
int changed;
@@ -171,28 +182,30 @@ static int cow_update_bitmap(BlockDriverState *bs, int64_t sector_num,
return error;
}
-static int cow_read(BlockDriverState *bs, int64_t sector_num,
- uint8_t *buf, int nb_sectors)
+static int coroutine_fn cow_read(BlockDriverState *bs, int64_t sector_num,
+ uint8_t *buf, int nb_sectors)
{
BDRVCowState *s = bs->opaque;
int ret, n;
while (nb_sectors > 0) {
- if (cow_is_allocated(bs, sector_num, nb_sectors, &n)) {
+ if (bdrv_co_is_allocated(bs, sector_num, nb_sectors, &n)) {
ret = bdrv_pread(bs->file,
s->cow_sectors_offset + sector_num * 512,
buf, n * 512);
- if (ret != n * 512)
- return -1;
+ if (ret < 0) {
+ return ret;
+ }
} else {
if (bs->backing_hd) {
/* read from the base image */
ret = bdrv_read(bs->backing_hd, sector_num, buf, n);
- if (ret < 0)
- return -1;
+ if (ret < 0) {
+ return ret;
+ }
} else {
- memset(buf, 0, n * 512);
- }
+ memset(buf, 0, n * 512);
+ }
}
nb_sectors -= n;
sector_num += n;
@@ -220,8 +233,9 @@ static int cow_write(BlockDriverState *bs, int64_t sector_num,
ret = bdrv_pwrite(bs->file, s->cow_sectors_offset + sector_num * 512,
buf, nb_sectors * 512);
- if (ret != nb_sectors * 512)
- return -1;
+ if (ret < 0) {
+ return ret;
+ }
return cow_update_bitmap(bs, sector_num, nb_sectors);
}
@@ -243,12 +257,12 @@ static void cow_close(BlockDriverState *bs)
static int cow_create(const char *filename, QEMUOptionParameter *options)
{
- int fd, cow_fd;
struct cow_header_v2 cow_header;
struct stat st;
int64_t image_sectors = 0;
const char *image_filename = NULL;
int ret;
+ BlockDriverState *cow_bs;
/* Read out options */
while (options && options->name) {
@@ -260,10 +274,16 @@ static int cow_create(const char *filename, QEMUOptionParameter *options)
options++;
}
- cow_fd = open(filename, O_WRONLY | O_CREAT | O_TRUNC | O_BINARY,
- 0644);
- if (cow_fd < 0)
- return -errno;
+ ret = bdrv_create_file(filename, options);
+ if (ret < 0) {
+ return ret;
+ }
+
+ ret = bdrv_file_open(&cow_bs, filename, BDRV_O_RDWR);
+ if (ret < 0) {
+ return ret;
+ }
+
memset(&cow_header, 0, sizeof(cow_header));
cow_header.magic = cpu_to_be32(COW_MAGIC);
cow_header.version = cpu_to_be32(COW_VERSION);
@@ -271,16 +291,9 @@ static int cow_create(const char *filename, QEMUOptionParameter *options)
/* Note: if no file, we put a dummy mtime */
cow_header.mtime = cpu_to_be32(0);
- fd = open(image_filename, O_RDONLY | O_BINARY);
- if (fd < 0) {
- close(cow_fd);
+ if (stat(image_filename, &st) != 0) {
goto mtime_fail;
}
- if (fstat(fd, &st) != 0) {
- close(fd);
- goto mtime_fail;
- }
- close(fd);
cow_header.mtime = cpu_to_be32(st.st_mtime);
mtime_fail:
pstrcpy(cow_header.backing_file, sizeof(cow_header.backing_file),
@@ -288,21 +301,20 @@ static int cow_create(const char *filename, QEMUOptionParameter *options)
}
cow_header.sectorsize = cpu_to_be32(512);
cow_header.size = cpu_to_be64(image_sectors * 512);
- ret = qemu_write_full(cow_fd, &cow_header, sizeof(cow_header));
- if (ret != sizeof(cow_header)) {
- ret = -errno;
+ ret = bdrv_pwrite(cow_bs, 0, &cow_header, sizeof(cow_header));
+ if (ret < 0) {
goto exit;
}
/* resize to include at least all the bitmap */
- ret = ftruncate(cow_fd, sizeof(cow_header) + ((image_sectors + 7) >> 3));
- if (ret) {
- ret = -errno;
+ ret = bdrv_truncate(cow_bs,
+ sizeof(cow_header) + ((image_sectors + 7) >> 3));
+ if (ret < 0) {
goto exit;
}
exit:
- close(cow_fd);
+ bdrv_delete(cow_bs);
return ret;
}
@@ -337,7 +349,7 @@ static BlockDriver bdrv_cow = {
.bdrv_read = cow_co_read,
.bdrv_write = cow_co_write,
.bdrv_co_flush_to_disk = cow_co_flush,
- .bdrv_is_allocated = cow_is_allocated,
+ .bdrv_co_is_allocated = cow_co_is_allocated,
.create_options = cow_create_options,
};
diff --git a/block/curl.c b/block/curl.c
index 4209ac88ce..e9102e3e20 100644
--- a/block/curl.c
+++ b/block/curl.c
@@ -509,10 +509,6 @@ static BlockDriverAIOCB *curl_aio_readv(BlockDriverState *bs,
acb = qemu_aio_get(&curl_aio_pool, bs, cb, opaque);
- if (!acb) {
- return NULL;
- }
-
acb->qiov = qiov;
acb->sector_num = sector_num;
acb->nb_sectors = nb_sectors;
diff --git a/block/nbd.c b/block/nbd.c
index 882b2dc84a..95212dac64 100644
--- a/block/nbd.c
+++ b/block/nbd.c
@@ -189,7 +189,7 @@ static int nbd_read(BlockDriverState *bs, int64_t sector_num,
request.type = NBD_CMD_READ;
request.handle = (uint64_t)(intptr_t)bs;
- request.from = sector_num * 512;;
+ request.from = sector_num * 512;
request.len = nb_sectors * 512;
if (nbd_send_request(s->sock, &request) == -1)
@@ -219,7 +219,7 @@ static int nbd_write(BlockDriverState *bs, int64_t sector_num,
request.type = NBD_CMD_WRITE;
request.handle = (uint64_t)(intptr_t)bs;
- request.from = sector_num * 512;;
+ request.from = sector_num * 512;
request.len = nb_sectors * 512;
if (nbd_send_request(s->sock, &request) == -1)
diff --git a/block/qcow.c b/block/qcow.c
index adecee06c9..b16955d764 100644
--- a/block/qcow.c
+++ b/block/qcow.c
@@ -26,6 +26,7 @@
#include "module.h"
#include <zlib.h>
#include "aes.h"
+#include "migration.h"
/**************************************************************/
/* QEMU COW block driver with compression and encryption support */
@@ -74,6 +75,7 @@ typedef struct BDRVQcowState {
AES_KEY aes_encrypt_key;
AES_KEY aes_decrypt_key;
CoMutex lock;
+ Error *migration_blocker;
} BDRVQcowState;
static int decompress_cluster(BlockDriverState *bs, uint64_t cluster_offset);
@@ -160,6 +162,12 @@ static int qcow_open(BlockDriverState *bs, int flags)
bs->backing_file[len] = '\0';
}
+ /* Disable migration when qcow images are used */
+ error_set(&s->migration_blocker,
+ QERR_BLOCK_FORMAT_FEATURE_NOT_SUPPORTED,
+ "qcow", bs->device_name, "live migration");
+ migrate_add_blocker(s->migration_blocker);
+
qemu_co_mutex_init(&s->lock);
return 0;
@@ -360,14 +368,16 @@ static uint64_t get_cluster_offset(BlockDriverState *bs,
return cluster_offset;
}
-static int qcow_is_allocated(BlockDriverState *bs, int64_t sector_num,
- int nb_sectors, int *pnum)
+static int coroutine_fn qcow_co_is_allocated(BlockDriverState *bs,
+ int64_t sector_num, int nb_sectors, int *pnum)
{
BDRVQcowState *s = bs->opaque;
int index_in_cluster, n;
uint64_t cluster_offset;
+ qemu_co_mutex_lock(&s->lock);
cluster_offset = get_cluster_offset(bs, sector_num << 9, 0, 0, 0, 0);
+ qemu_co_mutex_unlock(&s->lock);
index_in_cluster = sector_num & (s->cluster_sectors - 1);
n = s->cluster_sectors - index_in_cluster;
if (n > nb_sectors)
@@ -425,7 +435,7 @@ static int decompress_cluster(BlockDriverState *bs, uint64_t cluster_offset)
return 0;
}
-static int qcow_co_readv(BlockDriverState *bs, int64_t sector_num,
+static coroutine_fn int qcow_co_readv(BlockDriverState *bs, int64_t sector_num,
int nb_sectors, QEMUIOVector *qiov)
{
BDRVQcowState *s = bs->opaque;
@@ -523,7 +533,7 @@ fail:
goto done;
}
-static int qcow_co_writev(BlockDriverState *bs, int64_t sector_num,
+static coroutine_fn int qcow_co_writev(BlockDriverState *bs, int64_t sector_num,
int nb_sectors, QEMUIOVector *qiov)
{
BDRVQcowState *s = bs->opaque;
@@ -604,10 +614,14 @@ static int qcow_co_writev(BlockDriverState *bs, int64_t sector_num,
static void qcow_close(BlockDriverState *bs)
{
BDRVQcowState *s = bs->opaque;
+
g_free(s->l1_table);
g_free(s->l2_cache);
g_free(s->cluster_cache);
g_free(s->cluster_data);
+
+ migrate_del_blocker(s->migration_blocker);
+ error_free(s->migration_blocker);
}
static int qcow_create(const char *filename, QEMUOptionParameter *options)
@@ -832,7 +846,7 @@ static BlockDriver bdrv_qcow = {
.bdrv_co_readv = qcow_co_readv,
.bdrv_co_writev = qcow_co_writev,
.bdrv_co_flush_to_disk = qcow_co_flush,
- .bdrv_is_allocated = qcow_is_allocated,
+ .bdrv_co_is_allocated = qcow_co_is_allocated,
.bdrv_set_key = qcow_set_key,
.bdrv_make_empty = qcow_make_empty,
diff --git a/block/qcow2-cluster.c b/block/qcow2-cluster.c
index f4e049fa90..07a2e936fd 100644
--- a/block/qcow2-cluster.c
+++ b/block/qcow2-cluster.c
@@ -289,89 +289,62 @@ void qcow2_encrypt_sectors(BDRVQcowState *s, int64_t sector_num,
}
}
-
-static int qcow2_read(BlockDriverState *bs, int64_t sector_num,
- uint8_t *buf, int nb_sectors)
+static int coroutine_fn copy_sectors(BlockDriverState *bs,
+ uint64_t start_sect,
+ uint64_t cluster_offset,
+ int n_start, int n_end)
{
BDRVQcowState *s = bs->opaque;
- int ret, index_in_cluster, n, n1;
- uint64_t cluster_offset;
- struct iovec iov;
QEMUIOVector qiov;
+ struct iovec iov;
+ int n, ret;
- while (nb_sectors > 0) {
- n = nb_sectors;
-
- ret = qcow2_get_cluster_offset(bs, sector_num << 9, &n,
- &cluster_offset);
- if (ret < 0) {
- return ret;
- }
-
- index_in_cluster = sector_num & (s->cluster_sectors - 1);
- if (!cluster_offset) {
- if (bs->backing_hd) {
- /* read from the base image */
- iov.iov_base = buf;
- iov.iov_len = n * 512;
- qemu_iovec_init_external(&qiov, &iov, 1);
-
- n1 = qcow2_backing_read1(bs->backing_hd, &qiov, sector_num, n);
- if (n1 > 0) {
- BLKDBG_EVENT(bs->file, BLKDBG_READ_BACKING);
- ret = bdrv_read(bs->backing_hd, sector_num, buf, n1);
- if (ret < 0)
- return -1;
- }
- } else {
- memset(buf, 0, 512 * n);
- }
- } else if (cluster_offset & QCOW_OFLAG_COMPRESSED) {
- if (qcow2_decompress_cluster(bs, cluster_offset) < 0)
- return -1;
- memcpy(buf, s->cluster_cache + index_in_cluster * 512, 512 * n);
- } else {
- BLKDBG_EVENT(bs->file, BLKDBG_READ);
- ret = bdrv_pread(bs->file, cluster_offset + index_in_cluster * 512, buf, n * 512);
- if (ret != n * 512)
- return -1;
- if (s->crypt_method) {
- qcow2_encrypt_sectors(s, sector_num, buf, buf, n, 0,
- &s->aes_decrypt_key);
- }
- }
- nb_sectors -= n;
- sector_num += n;
- buf += n * 512;
+ /*
+ * If this is the last cluster and it is only partially used, we must only
+ * copy until the end of the image, or bdrv_check_request will fail for the
+ * bdrv_read/write calls below.
+ */
+ if (start_sect + n_end > bs->total_sectors) {
+ n_end = bs->total_sectors - start_sect;
}
- return 0;
-}
-
-static int copy_sectors(BlockDriverState *bs, uint64_t start_sect,
- uint64_t cluster_offset, int n_start, int n_end)
-{
- BDRVQcowState *s = bs->opaque;
- int n, ret;
n = n_end - n_start;
- if (n <= 0)
+ if (n <= 0) {
return 0;
+ }
+
+ iov.iov_len = n * BDRV_SECTOR_SIZE;
+ iov.iov_base = qemu_blockalign(bs, iov.iov_len);
+
+ qemu_iovec_init_external(&qiov, &iov, 1);
+
BLKDBG_EVENT(bs->file, BLKDBG_COW_READ);
- ret = qcow2_read(bs, start_sect + n_start, s->cluster_data, n);
- if (ret < 0)
- return ret;
+
+ /* Call .bdrv_co_readv() directly instead of using the public block-layer
+ * interface. This avoids double I/O throttling and request tracking,
+ * which can lead to deadlock when block layer copy-on-read is enabled.
+ */
+ ret = bs->drv->bdrv_co_readv(bs, start_sect + n_start, n, &qiov);
+ if (ret < 0) {
+ goto out;
+ }
+
if (s->crypt_method) {
qcow2_encrypt_sectors(s, start_sect + n_start,
- s->cluster_data,
- s->cluster_data, n, 1,
+ iov.iov_base, iov.iov_base, n, 1,
&s->aes_encrypt_key);
}
+
BLKDBG_EVENT(bs->file, BLKDBG_COW_WRITE);
- ret = bdrv_write(bs->file, (cluster_offset >> 9) + n_start,
- s->cluster_data, n);
- if (ret < 0)
- return ret;
- return 0;
+ ret = bdrv_co_writev(bs->file, (cluster_offset >> 9) + n_start, n, &qiov);
+ if (ret < 0) {
+ goto out;
+ }
+
+ ret = 0;
+out:
+ qemu_vfree(iov.iov_base);
+ return ret;
}
@@ -620,7 +593,9 @@ int qcow2_alloc_cluster_link_l2(BlockDriverState *bs, QCowL2Meta *m)
start_sect = (m->offset & ~(s->cluster_size - 1)) >> 9;
if (m->n_start) {
cow = true;
+ qemu_co_mutex_unlock(&s->lock);
ret = copy_sectors(bs, start_sect, cluster_offset, 0, m->n_start);
+ qemu_co_mutex_lock(&s->lock);
if (ret < 0)
goto err;
}
@@ -628,8 +603,10 @@ int qcow2_alloc_cluster_link_l2(BlockDriverState *bs, QCowL2Meta *m)
if (m->nb_available & (s->cluster_sectors - 1)) {
uint64_t end = m->nb_available & ~(uint64_t)(s->cluster_sectors - 1);
cow = true;
+ qemu_co_mutex_unlock(&s->lock);
ret = copy_sectors(bs, start_sect + end, cluster_offset + (end << 9),
m->nb_available - end, s->cluster_sectors);
+ qemu_co_mutex_lock(&s->lock);
if (ret < 0)
goto err;
}
diff --git a/block/qcow2-refcount.c b/block/qcow2-refcount.c
index 9605367777..2db2ede3d1 100644
--- a/block/qcow2-refcount.c
+++ b/block/qcow2-refcount.c
@@ -700,6 +700,10 @@ int qcow2_update_snapshot_refcount(BlockDriverState *bs,
l2_table = NULL;
l1_table = NULL;
l1_size2 = l1_size * sizeof(uint64_t);
+
+ /* WARNING: qcow2_snapshot_goto relies on this function not using the
+ * l1_table_offset when it is the current s->l1_table_offset! Be careful
+ * when changing this! */
if (l1_table_offset != s->l1_table_offset) {
if (l1_size2 != 0) {
l1_table = g_malloc0(align_offset(l1_size2, 512));
@@ -819,7 +823,8 @@ fail:
qcow2_cache_set_writethrough(bs, s->refcount_block_cache,
old_refcount_writethrough);
- if (l1_modified) {
+ /* Update L1 only if it isn't deleted anyway (addend = -1) */
+ if (addend >= 0 && l1_modified) {
for(i = 0; i < l1_size; i++)
cpu_to_be64s(&l1_table[i]);
if (bdrv_pwrite_sync(bs->file, l1_table_offset, l1_table,
diff --git a/block/qcow2-snapshot.c b/block/qcow2-snapshot.c
index bdc33ba94c..7d3fde5a8a 100644
--- a/block/qcow2-snapshot.c
+++ b/block/qcow2-snapshot.c
@@ -46,6 +46,10 @@ typedef struct QEMU_PACKED QCowSnapshotHeader {
/* name follows */
} QCowSnapshotHeader;
+typedef struct QEMU_PACKED QCowSnapshotExtraData {
+ uint64_t vm_state_size_large;
+} QCowSnapshotExtraData;
+
void qcow2_free_snapshots(BlockDriverState *bs)
{
BDRVQcowState *s = bs->opaque;
@@ -64,10 +68,12 @@ int qcow2_read_snapshots(BlockDriverState *bs)
{
BDRVQcowState *s = bs->opaque;
QCowSnapshotHeader h;
+ QCowSnapshotExtraData extra;
QCowSnapshot *sn;
int i, id_str_size, name_size;
int64_t offset;
uint32_t extra_data_size;
+ int ret;
if (!s->nb_snapshots) {
s->snapshots = NULL;
@@ -77,10 +83,15 @@ int qcow2_read_snapshots(BlockDriverState *bs)
offset = s->snapshots_offset;
s->snapshots = g_malloc0(s->nb_snapshots * sizeof(QCowSnapshot));
+
for(i = 0; i < s->nb_snapshots; i++) {
+ /* Read statically sized part of the snapshot header */
offset = align_offset(offset, 8);
- if (bdrv_pread(bs->file, offset, &h, sizeof(h)) != sizeof(h))
+ ret = bdrv_pread(bs->file, offset, &h, sizeof(h));
+ if (ret < 0) {
goto fail;
+ }
+
offset += sizeof(h);
sn = s->snapshots + i;
sn->l1_table_offset = be64_to_cpu(h.l1_table_offset);
@@ -94,25 +105,43 @@ int qcow2_read_snapshots(BlockDriverState *bs)
id_str_size = be16_to_cpu(h.id_str_size);
name_size = be16_to_cpu(h.name_size);
+ /* Read extra data */
+ ret = bdrv_pread(bs->file, offset, &extra,
+ MIN(sizeof(extra), extra_data_size));
+ if (ret < 0) {
+ goto fail;
+ }
offset += extra_data_size;
+ if (extra_data_size >= 8) {
+ sn->vm_state_size = be64_to_cpu(extra.vm_state_size_large);
+ }
+
+ /* Read snapshot ID */
sn->id_str = g_malloc(id_str_size + 1);
- if (bdrv_pread(bs->file, offset, sn->id_str, id_str_size) != id_str_size)
+ ret = bdrv_pread(bs->file, offset, sn->id_str, id_str_size);
+ if (ret < 0) {
goto fail;
+ }
offset += id_str_size;
sn->id_str[id_str_size] = '\0';
+ /* Read snapshot name */
sn->name = g_malloc(name_size + 1);
- if (bdrv_pread(bs->file, offset, sn->name, name_size) != name_size)
+ ret = bdrv_pread(bs->file, offset, sn->name, name_size);
+ if (ret < 0) {
goto fail;
+ }
offset += name_size;
sn->name[name_size] = '\0';
}
+
s->snapshots_size = offset - s->snapshots_offset;
return 0;
- fail:
+
+fail:
qcow2_free_snapshots(bs);
- return -1;
+ return ret;
}
/* add at the end of the file a new list of snapshots */
@@ -121,10 +150,14 @@ static int qcow2_write_snapshots(BlockDriverState *bs)
BDRVQcowState *s = bs->opaque;
QCowSnapshot *sn;
QCowSnapshotHeader h;
+ QCowSnapshotExtraData extra;
int i, name_size, id_str_size, snapshots_size;
- uint64_t data64;
- uint32_t data32;
+ struct {
+ uint32_t nb_snapshots;
+ uint64_t snapshots_offset;
+ } QEMU_PACKED header_data;
int64_t offset, snapshots_offset;
+ int ret;
/* compute the size of the snapshots */
offset = 0;
@@ -132,11 +165,13 @@ static int qcow2_write_snapshots(BlockDriverState *bs)
sn = s->snapshots + i;
offset = align_offset(offset, 8);
offset += sizeof(h);
+ offset += sizeof(extra);
offset += strlen(sn->id_str);
offset += strlen(sn->name);
}
snapshots_size = offset;
+ /* Allocate space for the new snapshot list */
snapshots_offset = qcow2_alloc_clusters(bs, snapshots_size);
bdrv_flush(bs->file);
offset = snapshots_offset;
@@ -144,49 +179,85 @@ static int qcow2_write_snapshots(BlockDriverState *bs)
return offset;
}
+ /* Write all snapshots to the new list */
for(i = 0; i < s->nb_snapshots; i++) {
sn = s->snapshots + i;
memset(&h, 0, sizeof(h));
h.l1_table_offset = cpu_to_be64(sn->l1_table_offset);
h.l1_size = cpu_to_be32(sn->l1_size);
- h.vm_state_size = cpu_to_be32(sn->vm_state_size);
+ /* If it doesn't fit in 32 bit, older implementations should treat it
+ * as a disk-only snapshot rather than truncate the VM state */
+ if (sn->vm_state_size <= 0xffffffff) {
+ h.vm_state_size = cpu_to_be32(sn->vm_state_size);
+ }
h.date_sec = cpu_to_be32(sn->date_sec);
h.date_nsec = cpu_to_be32(sn->date_nsec);
h.vm_clock_nsec = cpu_to_be64(sn->vm_clock_nsec);
+ h.extra_data_size = cpu_to_be32(sizeof(extra));
+
+ memset(&extra, 0, sizeof(extra));
+ extra.vm_state_size_large = cpu_to_be64(sn->vm_state_size);
id_str_size = strlen(sn->id_str);
name_size = strlen(sn->name);
h.id_str_size = cpu_to_be16(id_str_size);
h.name_size = cpu_to_be16(name_size);
offset = align_offset(offset, 8);
- if (bdrv_pwrite_sync(bs->file, offset, &h, sizeof(h)) < 0)
+
+ ret = bdrv_pwrite(bs->file, offset, &h, sizeof(h));
+ if (ret < 0) {
goto fail;
+ }
offset += sizeof(h);
- if (bdrv_pwrite_sync(bs->file, offset, sn->id_str, id_str_size) < 0)
+
+ ret = bdrv_pwrite(bs->file, offset, &extra, sizeof(extra));
+ if (ret < 0) {
goto fail;
+ }
+ offset += sizeof(extra);
+
+ ret = bdrv_pwrite(bs->file, offset, sn->id_str, id_str_size);
+ if (ret < 0) {
+ goto fail;
+ }
offset += id_str_size;
- if (bdrv_pwrite_sync(bs->file, offset, sn->name, name_size) < 0)
+
+ ret = bdrv_pwrite(bs->file, offset, sn->name, name_size);
+ if (ret < 0) {
goto fail;
+ }
offset += name_size;
}
- /* update the various header fields */
- data64 = cpu_to_be64(snapshots_offset);
- if (bdrv_pwrite_sync(bs->file, offsetof(QCowHeader, snapshots_offset),
- &data64, sizeof(data64)) < 0)
+ /*
+ * Update the header to point to the new snapshot table. This requires the
+ * new table and its refcounts to be stable on disk.
+ */
+ ret = bdrv_flush(bs);
+ if (ret < 0) {
goto fail;
- data32 = cpu_to_be32(s->nb_snapshots);
- if (bdrv_pwrite_sync(bs->file, offsetof(QCowHeader, nb_snapshots),
- &data32, sizeof(data32)) < 0)
+ }
+
+ QEMU_BUILD_BUG_ON(offsetof(QCowHeader, snapshots_offset) !=
+ offsetof(QCowHeader, nb_snapshots) + sizeof(header_data.nb_snapshots));
+
+ header_data.nb_snapshots = cpu_to_be32(s->nb_snapshots);
+ header_data.snapshots_offset = cpu_to_be64(snapshots_offset);
+
+ ret = bdrv_pwrite_sync(bs->file, offsetof(QCowHeader, nb_snapshots),
+ &header_data, sizeof(header_data));
+ if (ret < 0) {
goto fail;
+ }
/* free the old snapshot table */
qcow2_free_clusters(bs, s->snapshots_offset, s->snapshots_size);
s->snapshots_offset = snapshots_offset;
s->snapshots_size = snapshots_size;
return 0;
- fail:
- return -1;
+
+fail:
+ return ret;
}
static void find_new_snapshot_id(BlockDriverState *bs,
@@ -236,72 +307,92 @@ static int find_snapshot_by_id_or_name(BlockDriverState *bs, const char *name)
int qcow2_snapshot_create(BlockDriverState *bs, QEMUSnapshotInfo *sn_info)
{
BDRVQcowState *s = bs->opaque;
- QCowSnapshot *snapshots1, sn1, *sn = &sn1;
+ QCowSnapshot *new_snapshot_list = NULL;
+ QCowSnapshot *old_snapshot_list = NULL;
+ QCowSnapshot sn1, *sn = &sn1;
int i, ret;
uint64_t *l1_table = NULL;
int64_t l1_table_offset;
memset(sn, 0, sizeof(*sn));
+ /* Generate an ID if it wasn't passed */
if (sn_info->id_str[0] == '\0') {
- /* compute a new id */
find_new_snapshot_id(bs, sn_info->id_str, sizeof(sn_info->id_str));
}
- /* check that the ID is unique */
- if (find_snapshot_by_id(bs, sn_info->id_str) >= 0)
+ /* Check that the ID is unique */
+ if (find_snapshot_by_id(bs, sn_info->id_str) >= 0) {
return -ENOENT;
+ }
+ /* Populate sn with passed data */
sn->id_str = g_strdup(sn_info->id_str);
- if (!sn->id_str)
- goto fail;
sn->name = g_strdup(sn_info->name);
- if (!sn->name)
- goto fail;
+
sn->vm_state_size = sn_info->vm_state_size;
sn->date_sec = sn_info->date_sec;
sn->date_nsec = sn_info->date_nsec;
sn->vm_clock_nsec = sn_info->vm_clock_nsec;
- ret = qcow2_update_snapshot_refcount(bs, s->l1_table_offset, s->l1_size, 1);
- if (ret < 0)
- goto fail;
-
- /* create the L1 table of the snapshot */
+ /* Allocate the L1 table of the snapshot and copy the current one there. */
l1_table_offset = qcow2_alloc_clusters(bs, s->l1_size * sizeof(uint64_t));
if (l1_table_offset < 0) {
+ ret = l1_table_offset;
goto fail;
}
- bdrv_flush(bs->file);
sn->l1_table_offset = l1_table_offset;
sn->l1_size = s->l1_size;
- if (s->l1_size != 0) {
- l1_table = g_malloc(s->l1_size * sizeof(uint64_t));
- } else {
- l1_table = NULL;
- }
-
+ l1_table = g_malloc(s->l1_size * sizeof(uint64_t));
for(i = 0; i < s->l1_size; i++) {
l1_table[i] = cpu_to_be64(s->l1_table[i]);
}
- if (bdrv_pwrite_sync(bs->file, sn->l1_table_offset,
- l1_table, s->l1_size * sizeof(uint64_t)) < 0)
+
+ ret = bdrv_pwrite(bs->file, sn->l1_table_offset, l1_table,
+ s->l1_size * sizeof(uint64_t));
+ if (ret < 0) {
goto fail;
+ }
+
g_free(l1_table);
l1_table = NULL;
- snapshots1 = g_malloc((s->nb_snapshots + 1) * sizeof(QCowSnapshot));
+ /*
+ * Increase the refcounts of all clusters and make sure everything is
+ * stable on disk before updating the snapshot table to contain a pointer
+ * to the new L1 table.
+ */
+ ret = qcow2_update_snapshot_refcount(bs, s->l1_table_offset, s->l1_size, 1);
+ if (ret < 0) {
+ goto fail;
+ }
+
+ ret = bdrv_flush(bs);
+ if (ret < 0) {
+ goto fail;
+ }
+
+ /* Append the new snapshot to the snapshot list */
+ new_snapshot_list = g_malloc((s->nb_snapshots + 1) * sizeof(QCowSnapshot));
if (s->snapshots) {
- memcpy(snapshots1, s->snapshots, s->nb_snapshots * sizeof(QCowSnapshot));
- g_free(s->snapshots);
+ memcpy(new_snapshot_list, s->snapshots,
+ s->nb_snapshots * sizeof(QCowSnapshot));
+ old_snapshot_list = s->snapshots;
}
- s->snapshots = snapshots1;
+ s->snapshots = new_snapshot_list;
s->snapshots[s->nb_snapshots++] = *sn;
- if (qcow2_write_snapshots(bs) < 0)
+ ret = qcow2_write_snapshots(bs);
+ if (ret < 0) {
+ g_free(s->snapshots);
+ s->snapshots = old_snapshot_list;
goto fail;
+ }
+
+ g_free(old_snapshot_list);
+
#ifdef DEBUG_ALLOC
{
BdrvCheckResult result = {0};
@@ -309,10 +400,13 @@ int qcow2_snapshot_create(BlockDriverState *bs, QEMUSnapshotInfo *sn_info)
}
#endif
return 0;
- fail:
+
+fail:
+ g_free(sn->id_str);
g_free(sn->name);
g_free(l1_table);
- return -1;
+
+ return ret;
}
/* copy the snapshot 'snapshot_name' into the current disk image */
@@ -322,38 +416,92 @@ int qcow2_snapshot_goto(BlockDriverState *bs, const char *snapshot_id)
QCowSnapshot *sn;
int i, snapshot_index;
int cur_l1_bytes, sn_l1_bytes;
+ int ret;
+ uint64_t *sn_l1_table = NULL;
+ /* Search the snapshot */
snapshot_index = find_snapshot_by_id_or_name(bs, snapshot_id);
- if (snapshot_index < 0)
+ if (snapshot_index < 0) {
return -ENOENT;
+ }
sn = &s->snapshots[snapshot_index];
- if (qcow2_update_snapshot_refcount(bs, s->l1_table_offset, s->l1_size, -1) < 0)
- goto fail;
-
- if (qcow2_grow_l1_table(bs, sn->l1_size, true) < 0)
+ /*
+ * Make sure that the current L1 table is big enough to contain the whole
+ * L1 table of the snapshot. If the snapshot L1 table is smaller, the
+ * current one must be padded with zeros.
+ */
+ ret = qcow2_grow_l1_table(bs, sn->l1_size, true);
+ if (ret < 0) {
goto fail;
+ }
cur_l1_bytes = s->l1_size * sizeof(uint64_t);
sn_l1_bytes = sn->l1_size * sizeof(uint64_t);
- if (cur_l1_bytes > sn_l1_bytes) {
- memset(s->l1_table + sn->l1_size, 0, cur_l1_bytes - sn_l1_bytes);
+ /*
+ * Copy the snapshot L1 table to the current L1 table.
+ *
+ * Before overwriting the old current L1 table on disk, make sure to
+ * increase all refcounts for the clusters referenced by the new one.
+ * Decrease the refcount referenced by the old one only when the L1
+ * table is overwritten.
+ */
+ sn_l1_table = g_malloc0(cur_l1_bytes);
+
+ ret = bdrv_pread(bs->file, sn->l1_table_offset, sn_l1_table, sn_l1_bytes);
+ if (ret < 0) {
+ goto fail;
}
- /* copy the snapshot l1 table to the current l1 table */
- if (bdrv_pread(bs->file, sn->l1_table_offset,
- s->l1_table, sn_l1_bytes) < 0)
+ ret = qcow2_update_snapshot_refcount(bs, sn->l1_table_offset,
+ sn->l1_size, 1);
+ if (ret < 0) {
goto fail;
- if (bdrv_pwrite_sync(bs->file, s->l1_table_offset,
- s->l1_table, cur_l1_bytes) < 0)
+ }
+
+ ret = bdrv_pwrite_sync(bs->file, s->l1_table_offset, sn_l1_table,
+ cur_l1_bytes);
+ if (ret < 0) {
goto fail;
+ }
+
+ /*
+ * Decrease refcount of clusters of current L1 table.
+ *
+ * At this point, the in-memory s->l1_table points to the old L1 table,
+ * whereas on disk we already have the new one.
+ *
+ * qcow2_update_snapshot_refcount special cases the current L1 table to use
+ * the in-memory data instead of really using the offset to load a new one,
+ * which is why this works.
+ */
+ ret = qcow2_update_snapshot_refcount(bs, s->l1_table_offset,
+ s->l1_size, -1);
+
+ /*
+ * Now update the in-memory L1 table to be in sync with the on-disk one. We
+ * need to do this even if updating refcounts failed.
+ */
for(i = 0;i < s->l1_size; i++) {
- be64_to_cpus(&s->l1_table[i]);
+ s->l1_table[i] = be64_to_cpu(sn_l1_table[i]);
}
- if (qcow2_update_snapshot_refcount(bs, s->l1_table_offset, s->l1_size, 1) < 0)
+ if (ret < 0) {
+ goto fail;
+ }
+
+ g_free(sn_l1_table);
+ sn_l1_table = NULL;
+
+ /*
+ * Update QCOW_OFLAG_COPIED in the active L1 table (it may have changed
+ * when we decreased the refcount of the old snapshot.
+ */
+ ret = qcow2_update_snapshot_refcount(bs, s->l1_table_offset, s->l1_size, 0);
+ if (ret < 0) {
goto fail;
+ }
#ifdef DEBUG_ALLOC
{
@@ -362,39 +510,59 @@ int qcow2_snapshot_goto(BlockDriverState *bs, const char *snapshot_id)
}
#endif
return 0;
- fail:
- return -EIO;
+
+fail:
+ g_free(sn_l1_table);
+ return ret;
}
int qcow2_snapshot_delete(BlockDriverState *bs, const char *snapshot_id)
{
BDRVQcowState *s = bs->opaque;
- QCowSnapshot *sn;
+ QCowSnapshot sn;
int snapshot_index, ret;
+ /* Search the snapshot */
snapshot_index = find_snapshot_by_id_or_name(bs, snapshot_id);
- if (snapshot_index < 0)
+ if (snapshot_index < 0) {
return -ENOENT;
- sn = &s->snapshots[snapshot_index];
+ }
+ sn = s->snapshots[snapshot_index];
- ret = qcow2_update_snapshot_refcount(bs, sn->l1_table_offset, sn->l1_size, -1);
- if (ret < 0)
+ /* Remove it from the snapshot list */
+ memmove(s->snapshots + snapshot_index,
+ s->snapshots + snapshot_index + 1,
+ (s->nb_snapshots - snapshot_index - 1) * sizeof(sn));
+ s->nb_snapshots--;
+ ret = qcow2_write_snapshots(bs);
+ if (ret < 0) {
return ret;
- /* must update the copied flag on the current cluster offsets */
- ret = qcow2_update_snapshot_refcount(bs, s->l1_table_offset, s->l1_size, 0);
- if (ret < 0)
+ }
+
+ /*
+ * The snapshot is now unused, clean up. If we fail after this point, we
+ * won't recover but just leak clusters.
+ */
+ g_free(sn.id_str);
+ g_free(sn.name);
+
+ /*
+ * Now decrease the refcounts of clusters referenced by the snapshot and
+ * free the L1 table.
+ */
+ ret = qcow2_update_snapshot_refcount(bs, sn.l1_table_offset,
+ sn.l1_size, -1);
+ if (ret < 0) {
return ret;
- qcow2_free_clusters(bs, sn->l1_table_offset, sn->l1_size * sizeof(uint64_t));
+ }
+ qcow2_free_clusters(bs, sn.l1_table_offset, sn.l1_size * sizeof(uint64_t));
- g_free(sn->id_str);
- g_free(sn->name);
- memmove(sn, sn + 1, (s->nb_snapshots - snapshot_index - 1) * sizeof(*sn));
- s->nb_snapshots--;
- ret = qcow2_write_snapshots(bs);
+ /* must update the copied flag on the current cluster offsets */
+ ret = qcow2_update_snapshot_refcount(bs, s->l1_table_offset, s->l1_size, 0);
if (ret < 0) {
- /* XXX: restore snapshot if error ? */
return ret;
}
+
#ifdef DEBUG_ALLOC
{
BdrvCheckResult result = {0};
@@ -435,32 +603,42 @@ int qcow2_snapshot_list(BlockDriverState *bs, QEMUSnapshotInfo **psn_tab)
int qcow2_snapshot_load_tmp(BlockDriverState *bs, const char *snapshot_name)
{
- int i, snapshot_index, l1_size2;
+ int i, snapshot_index;
BDRVQcowState *s = bs->opaque;
QCowSnapshot *sn;
+ uint64_t *new_l1_table;
+ int new_l1_bytes;
+ int ret;
+ assert(bs->read_only);
+
+ /* Search the snapshot */
snapshot_index = find_snapshot_by_id_or_name(bs, snapshot_name);
if (snapshot_index < 0) {
return -ENOENT;
}
-
sn = &s->snapshots[snapshot_index];
- s->l1_size = sn->l1_size;
- l1_size2 = s->l1_size * sizeof(uint64_t);
- if (s->l1_table != NULL) {
- g_free(s->l1_table);
- }
- s->l1_table_offset = sn->l1_table_offset;
- s->l1_table = g_malloc0(align_offset(l1_size2, 512));
+ /* Allocate and read in the snapshot's L1 table */
+ new_l1_bytes = s->l1_size * sizeof(uint64_t);
+ new_l1_table = g_malloc0(align_offset(new_l1_bytes, 512));
- if (bdrv_pread(bs->file, sn->l1_table_offset,
- s->l1_table, l1_size2) != l1_size2) {
- return -1;
+ ret = bdrv_pread(bs->file, sn->l1_table_offset, new_l1_table, new_l1_bytes);
+ if (ret < 0) {
+ g_free(new_l1_table);
+ return ret;
}
+ /* Switch the L1 table */
+ g_free(s->l1_table);
+
+ s->l1_size = sn->l1_size;
+ s->l1_table_offset = sn->l1_table_offset;
+ s->l1_table = new_l1_table;
+
for(i = 0;i < s->l1_size; i++) {
be64_to_cpus(&s->l1_table[i]);
}
+
return 0;
}
diff --git a/block/qcow2.c b/block/qcow2.c
index d7805ce943..aa32e8d01a 100644
--- a/block/qcow2.c
+++ b/block/qcow2.c
@@ -92,7 +92,7 @@ static int qcow2_read_extensions(BlockDriverState *bs, uint64_t start_offset,
if (offset > s->cluster_size)
printf("qcow2_read_extension: suspicious offset %lu\n", offset);
- printf("attemting to read extended header in offset %lu\n", offset);
+ printf("attempting to read extended header in offset %lu\n", offset);
#endif
if (bdrv_pread(bs->file, offset, &ext, sizeof(ext)) != sizeof(ext)) {
@@ -273,8 +273,9 @@ static int qcow2_open(BlockDriverState *bs, int flags)
}
bs->backing_file[len] = '\0';
}
- if (qcow2_read_snapshots(bs) < 0) {
- ret = -EINVAL;
+
+ ret = qcow2_read_snapshots(bs);
+ if (ret < 0) {
goto fail;
}
@@ -343,16 +344,19 @@ static int qcow2_set_key(BlockDriverState *bs, const char *key)
return 0;
}
-static int qcow2_is_allocated(BlockDriverState *bs, int64_t sector_num,
- int nb_sectors, int *pnum)
+static int coroutine_fn qcow2_co_is_allocated(BlockDriverState *bs,
+ int64_t sector_num, int nb_sectors, int *pnum)
{
+ BDRVQcowState *s = bs->opaque;
uint64_t cluster_offset;
int ret;
*pnum = nb_sectors;
- /* FIXME We can get errors here, but the bdrv_is_allocated interface can't
- * pass them on today */
+ /* FIXME We can get errors here, but the bdrv_co_is_allocated interface
+ * can't pass them on today */
+ qemu_co_mutex_lock(&s->lock);
ret = qcow2_get_cluster_offset(bs, sector_num << 9, pnum, &cluster_offset);
+ qemu_co_mutex_unlock(&s->lock);
if (ret < 0) {
*pnum = 0;
}
@@ -377,7 +381,7 @@ int qcow2_backing_read1(BlockDriverState *bs, QEMUIOVector *qiov,
return n1;
}
-static int qcow2_co_readv(BlockDriverState *bs, int64_t sector_num,
+static coroutine_fn int qcow2_co_readv(BlockDriverState *bs, int64_t sector_num,
int remaining_sectors, QEMUIOVector *qiov)
{
BDRVQcowState *s = bs->opaque;
@@ -512,12 +516,12 @@ static void run_dependent_requests(BDRVQcowState *s, QCowL2Meta *m)
/* Restart all dependent requests */
if (!qemu_co_queue_empty(&m->dependent_requests)) {
qemu_co_mutex_unlock(&s->lock);
- while(qemu_co_queue_next(&m->dependent_requests));
+ qemu_co_queue_restart_all(&m->dependent_requests);
qemu_co_mutex_lock(&s->lock);
}
}
-static int qcow2_co_writev(BlockDriverState *bs,
+static coroutine_fn int qcow2_co_writev(BlockDriverState *bs,
int64_t sector_num,
int remaining_sectors,
QEMUIOVector *qiov)
@@ -631,6 +635,7 @@ static void qcow2_close(BlockDriverState *bs)
g_free(s->cluster_cache);
qemu_vfree(s->cluster_data);
qcow2_refcount_close(bs);
+ qcow2_free_snapshots(bs);
}
static void qcow2_invalidate_cache(BlockDriverState *bs)
@@ -821,7 +826,7 @@ static int qcow2_create2(const char *filename, int64_t total_size,
int flags, size_t cluster_size, int prealloc,
QEMUOptionParameter *options)
{
- /* Calulate cluster_bits */
+ /* Calculate cluster_bits */
int cluster_bits;
cluster_bits = ffs(cluster_size) - 1;
if (cluster_bits < MIN_CLUSTER_BITS || cluster_bits > MAX_CLUSTER_BITS ||
@@ -1137,7 +1142,7 @@ fail:
return ret;
}
-static int qcow2_co_flush_to_os(BlockDriverState *bs)
+static coroutine_fn int qcow2_co_flush_to_os(BlockDriverState *bs)
{
BDRVQcowState *s = bs->opaque;
int ret;
@@ -1159,7 +1164,7 @@ static int qcow2_co_flush_to_os(BlockDriverState *bs)
return 0;
}
-static int qcow2_co_flush_to_disk(BlockDriverState *bs)
+static coroutine_fn int qcow2_co_flush_to_disk(BlockDriverState *bs)
{
return bdrv_co_flush(bs->file);
}
@@ -1276,7 +1281,7 @@ static BlockDriver bdrv_qcow2 = {
.bdrv_open = qcow2_open,
.bdrv_close = qcow2_close,
.bdrv_create = qcow2_create,
- .bdrv_is_allocated = qcow2_is_allocated,
+ .bdrv_co_is_allocated = qcow2_co_is_allocated,
.bdrv_set_key = qcow2_set_key,
.bdrv_make_empty = qcow2_make_empty,
diff --git a/block/qcow2.h b/block/qcow2.h
index 4e44eea5ef..99e45361f5 100644
--- a/block/qcow2.h
+++ b/block/qcow2.h
@@ -78,7 +78,7 @@ typedef struct QCowSnapshot {
uint32_t l1_size;
char *id_str;
char *name;
- uint32_t vm_state_size;
+ uint64_t vm_state_size;
uint32_t date_sec;
uint32_t date_nsec;
uint64_t vm_clock_nsec;
diff --git a/block/qed-table.c b/block/qed-table.c
index f31f9ff069..ce07b05549 100644
--- a/block/qed-table.c
+++ b/block/qed-table.c
@@ -29,7 +29,7 @@ static void qed_read_table_cb(void *opaque, int ret)
{
QEDReadTableCB *read_table_cb = opaque;
QEDTable *table = read_table_cb->table;
- int noffsets = read_table_cb->iov.iov_len / sizeof(uint64_t);
+ int noffsets = read_table_cb->qiov.size / sizeof(uint64_t);
int i;
/* Handle I/O error */
@@ -54,7 +54,6 @@ static void qed_read_table(BDRVQEDState *s, uint64_t offset, QEDTable *table,
QEDReadTableCB *read_table_cb = gencb_alloc(sizeof(*read_table_cb),
cb, opaque);
QEMUIOVector *qiov = &read_table_cb->qiov;
- BlockDriverAIOCB *aiocb;
trace_qed_read_table(s, offset, table);
@@ -64,12 +63,9 @@ static void qed_read_table(BDRVQEDState *s, uint64_t offset, QEDTable *table,
read_table_cb->iov.iov_len = s->header.cluster_size * s->header.table_size,
qemu_iovec_init_external(qiov, &read_table_cb->iov, 1);
- aiocb = bdrv_aio_readv(s->bs->file, offset / BDRV_SECTOR_SIZE, qiov,
- read_table_cb->iov.iov_len / BDRV_SECTOR_SIZE,
- qed_read_table_cb, read_table_cb);
- if (!aiocb) {
- qed_read_table_cb(read_table_cb, -EIO);
- }
+ bdrv_aio_readv(s->bs->file, offset / BDRV_SECTOR_SIZE, qiov,
+ qiov->size / BDRV_SECTOR_SIZE,
+ qed_read_table_cb, read_table_cb);
}
typedef struct {
@@ -127,7 +123,6 @@ static void qed_write_table(BDRVQEDState *s, uint64_t offset, QEDTable *table,
BlockDriverCompletionFunc *cb, void *opaque)
{
QEDWriteTableCB *write_table_cb;
- BlockDriverAIOCB *aiocb;
unsigned int sector_mask = BDRV_SECTOR_SIZE / sizeof(uint64_t) - 1;
unsigned int start, end, i;
size_t len_bytes;
@@ -158,13 +153,10 @@ static void qed_write_table(BDRVQEDState *s, uint64_t offset, QEDTable *table,
/* Adjust for offset into table */
offset += start * sizeof(uint64_t);
- aiocb = bdrv_aio_writev(s->bs->file, offset / BDRV_SECTOR_SIZE,
- &write_table_cb->qiov,
- write_table_cb->iov.iov_len / BDRV_SECTOR_SIZE,
- qed_write_table_cb, write_table_cb);
- if (!aiocb) {
- qed_write_table_cb(write_table_cb, -EIO);
- }
+ bdrv_aio_writev(s->bs->file, offset / BDRV_SECTOR_SIZE,
+ &write_table_cb->qiov,
+ write_table_cb->qiov.size / BDRV_SECTOR_SIZE,
+ qed_write_table_cb, write_table_cb);
}
/**
diff --git a/block/qed.c b/block/qed.c
index 7e22e77b9d..8da3ebe9d4 100644
--- a/block/qed.c
+++ b/block/qed.c
@@ -123,7 +123,6 @@ static void qed_write_header_read_cb(void *opaque, int ret)
{
QEDWriteHeaderCB *write_header_cb = opaque;
BDRVQEDState *s = write_header_cb->s;
- BlockDriverAIOCB *acb;
if (ret) {
qed_write_header_cb(write_header_cb, ret);
@@ -133,12 +132,9 @@ static void qed_write_header_read_cb(void *opaque, int ret)
/* Update header */
qed_header_cpu_to_le(&s->header, (QEDHeader *)write_header_cb->buf);
- acb = bdrv_aio_writev(s->bs->file, 0, &write_header_cb->qiov,
- write_header_cb->nsectors, qed_write_header_cb,
- write_header_cb);
- if (!acb) {
- qed_write_header_cb(write_header_cb, -EIO);
- }
+ bdrv_aio_writev(s->bs->file, 0, &write_header_cb->qiov,
+ write_header_cb->nsectors, qed_write_header_cb,
+ write_header_cb);
}
/**
@@ -156,7 +152,6 @@ static void qed_write_header(BDRVQEDState *s, BlockDriverCompletionFunc cb,
* them, and write back.
*/
- BlockDriverAIOCB *acb;
int nsectors = (sizeof(QEDHeader) + BDRV_SECTOR_SIZE - 1) /
BDRV_SECTOR_SIZE;
size_t len = nsectors * BDRV_SECTOR_SIZE;
@@ -170,11 +165,8 @@ static void qed_write_header(BDRVQEDState *s, BlockDriverCompletionFunc cb,
write_header_cb->iov.iov_len = len;
qemu_iovec_init_external(&write_header_cb->qiov, &write_header_cb->iov, 1);
- acb = bdrv_aio_readv(s->bs->file, 0, &write_header_cb->qiov, nsectors,
- qed_write_header_read_cb, write_header_cb);
- if (!acb) {
- qed_write_header_cb(write_header_cb, -EIO);
- }
+ bdrv_aio_readv(s->bs->file, 0, &write_header_cb->qiov, nsectors,
+ qed_write_header_read_cb, write_header_cb);
}
static uint64_t qed_max_image_size(uint32_t cluster_size, uint32_t table_size)
@@ -661,6 +653,7 @@ static int bdrv_qed_create(const char *filename, QEMUOptionParameter *options)
}
typedef struct {
+ Coroutine *co;
int is_allocated;
int *pnum;
} QEDIsAllocatedCB;
@@ -670,10 +663,14 @@ static void qed_is_allocated_cb(void *opaque, int ret, uint64_t offset, size_t l
QEDIsAllocatedCB *cb = opaque;
*cb->pnum = len / BDRV_SECTOR_SIZE;
cb->is_allocated = (ret == QED_CLUSTER_FOUND || ret == QED_CLUSTER_ZERO);
+ if (cb->co) {
+ qemu_coroutine_enter(cb->co, NULL);
+ }
}
-static int bdrv_qed_is_allocated(BlockDriverState *bs, int64_t sector_num,
- int nb_sectors, int *pnum)
+static int coroutine_fn bdrv_qed_co_is_allocated(BlockDriverState *bs,
+ int64_t sector_num,
+ int nb_sectors, int *pnum)
{
BDRVQEDState *s = bs->opaque;
uint64_t pos = (uint64_t)sector_num * BDRV_SECTOR_SIZE;
@@ -686,8 +683,10 @@ static int bdrv_qed_is_allocated(BlockDriverState *bs, int64_t sector_num,
qed_find_cluster(s, &request, pos, len, qed_is_allocated_cb, &cb);
+ /* Now sleep if the callback wasn't invoked immediately */
while (cb.is_allocated == -1) {
- qemu_aio_wait();
+ cb.co = qemu_coroutine_self();
+ qemu_coroutine_yield();
}
qed_unref_l2_cache_entry(request.l2_table);
@@ -721,7 +720,6 @@ static void qed_read_backing_file(BDRVQEDState *s, uint64_t pos,
QEMUIOVector *qiov,
BlockDriverCompletionFunc *cb, void *opaque)
{
- BlockDriverAIOCB *aiocb;
uint64_t backing_length = 0;
size_t size;
@@ -753,11 +751,8 @@ static void qed_read_backing_file(BDRVQEDState *s, uint64_t pos,
size = MIN((uint64_t)backing_length - pos, qiov->size);
BLKDBG_EVENT(s->bs->file, BLKDBG_READ_BACKING);
- aiocb = bdrv_aio_readv(s->bs->backing_hd, pos / BDRV_SECTOR_SIZE,
- qiov, size / BDRV_SECTOR_SIZE, cb, opaque);
- if (!aiocb) {
- cb(opaque, -EIO);
- }
+ bdrv_aio_readv(s->bs->backing_hd, pos / BDRV_SECTOR_SIZE,
+ qiov, size / BDRV_SECTOR_SIZE, cb, opaque);
}
typedef struct {
@@ -779,7 +774,6 @@ static void qed_copy_from_backing_file_write(void *opaque, int ret)
{
CopyFromBackingFileCB *copy_cb = opaque;
BDRVQEDState *s = copy_cb->s;
- BlockDriverAIOCB *aiocb;
if (ret) {
qed_copy_from_backing_file_cb(copy_cb, ret);
@@ -787,13 +781,9 @@ static void qed_copy_from_backing_file_write(void *opaque, int ret)
}
BLKDBG_EVENT(s->bs->file, BLKDBG_COW_WRITE);
- aiocb = bdrv_aio_writev(s->bs->file, copy_cb->offset / BDRV_SECTOR_SIZE,
- &copy_cb->qiov,
- copy_cb->qiov.size / BDRV_SECTOR_SIZE,
- qed_copy_from_backing_file_cb, copy_cb);
- if (!aiocb) {
- qed_copy_from_backing_file_cb(copy_cb, -EIO);
- }
+ bdrv_aio_writev(s->bs->file, copy_cb->offset / BDRV_SECTOR_SIZE,
+ &copy_cb->qiov, copy_cb->qiov.size / BDRV_SECTOR_SIZE,
+ qed_copy_from_backing_file_cb, copy_cb);
}
/**
@@ -1015,7 +1005,6 @@ static void qed_aio_write_main(void *opaque, int ret)
uint64_t offset = acb->cur_cluster +
qed_offset_into_cluster(s, acb->cur_pos);
BlockDriverCompletionFunc *next_fn;
- BlockDriverAIOCB *file_acb;
trace_qed_aio_write_main(s, acb, ret, offset, acb->cur_qiov.size);
@@ -1035,13 +1024,9 @@ static void qed_aio_write_main(void *opaque, int ret)
}
BLKDBG_EVENT(s->bs->file, BLKDBG_WRITE_AIO);
- file_acb = bdrv_aio_writev(s->bs->file, offset / BDRV_SECTOR_SIZE,
- &acb->cur_qiov,
- acb->cur_qiov.size / BDRV_SECTOR_SIZE,
- next_fn, acb);
- if (!file_acb) {
- qed_aio_complete(acb, -EIO);
- }
+ bdrv_aio_writev(s->bs->file, offset / BDRV_SECTOR_SIZE,
+ &acb->cur_qiov, acb->cur_qiov.size / BDRV_SECTOR_SIZE,
+ next_fn, acb);
}
/**
@@ -1208,7 +1193,6 @@ static void qed_aio_read_data(void *opaque, int ret,
QEDAIOCB *acb = opaque;
BDRVQEDState *s = acb_to_s(acb);
BlockDriverState *bs = acb->common.bs;
- BlockDriverAIOCB *file_acb;
/* Adjust offset into cluster */
offset += qed_offset_into_cluster(s, acb->cur_pos);
@@ -1233,14 +1217,9 @@ static void qed_aio_read_data(void *opaque, int ret,
}
BLKDBG_EVENT(bs->file, BLKDBG_READ_AIO);
- file_acb = bdrv_aio_readv(bs->file, offset / BDRV_SECTOR_SIZE,
- &acb->cur_qiov,
- acb->cur_qiov.size / BDRV_SECTOR_SIZE,
- qed_aio_next_io, acb);
- if (!file_acb) {
- ret = -EIO;
- goto err;
- }
+ bdrv_aio_readv(bs->file, offset / BDRV_SECTOR_SIZE,
+ &acb->cur_qiov, acb->cur_qiov.size / BDRV_SECTOR_SIZE,
+ qed_aio_next_io, acb);
return;
err:
@@ -1485,7 +1464,7 @@ static BlockDriver bdrv_qed = {
.bdrv_open = bdrv_qed_open,
.bdrv_close = bdrv_qed_close,
.bdrv_create = bdrv_qed_create,
- .bdrv_is_allocated = bdrv_qed_is_allocated,
+ .bdrv_co_is_allocated = bdrv_qed_co_is_allocated,
.bdrv_make_empty = bdrv_qed_make_empty,
.bdrv_aio_readv = bdrv_qed_aio_readv,
.bdrv_aio_writev = bdrv_qed_aio_writev,
diff --git a/block/raw-posix.c b/block/raw-posix.c
index a3de373586..2ee5d690e9 100644
--- a/block/raw-posix.c
+++ b/block/raw-posix.c
@@ -1153,7 +1153,7 @@ static int cdrom_open(BlockDriverState *bs, const char *filename, int flags)
if (ret)
return ret;
- /* make sure the door isnt locked at this time */
+ /* make sure the door isn't locked at this time */
ioctl(s->fd, CDIOCALLOW);
return 0;
}
@@ -1184,7 +1184,7 @@ static int cdrom_reopen(BlockDriverState *bs)
}
s->fd = fd;
- /* make sure the door isnt locked at this time */
+ /* make sure the door isn't locked at this time */
ioctl(s->fd, CDIOCALLOW);
return 0;
}
diff --git a/block/rbd.c b/block/rbd.c
index 9088c52d24..7a2384c8f9 100644
--- a/block/rbd.c
+++ b/block/rbd.c
@@ -632,9 +632,6 @@ static BlockDriverAIOCB *rbd_aio_rw_vector(BlockDriverState *bs,
BDRVRBDState *s = bs->opaque;
acb = qemu_aio_get(&rbd_aio_pool, bs, cb, opaque);
- if (!acb) {
- return NULL;
- }
acb->write = write;
acb->qiov = qiov;
acb->bounce = qemu_blockalign(bs, qiov->size);
@@ -808,7 +805,7 @@ static int qemu_rbd_snap_list(BlockDriverState *bs,
} while (snap_count == -ERANGE);
if (snap_count <= 0) {
- return snap_count;
+ goto done;
}
sn_tab = g_malloc0(snap_count * sizeof(QEMUSnapshotInfo));
@@ -827,6 +824,7 @@ static int qemu_rbd_snap_list(BlockDriverState *bs,
}
rbd_snap_list_end(snaps);
+ done:
*psn_tab = sn_tab;
return snap_count;
}
diff --git a/block/sheepdog.c b/block/sheepdog.c
index 9f8060960f..aa9707f2ae 100644
--- a/block/sheepdog.c
+++ b/block/sheepdog.c
@@ -1116,6 +1116,7 @@ static int coroutine_fn add_aio_request(BDRVSheepdogState *s, AIOReq *aio_req,
/* send a header */
ret = do_write(s->fd, &hdr, sizeof(hdr));
if (ret) {
+ qemu_co_mutex_unlock(&s->lock);
error_report("failed to send a req, %s", strerror(errno));
return -EIO;
}
@@ -1123,6 +1124,7 @@ static int coroutine_fn add_aio_request(BDRVSheepdogState *s, AIOReq *aio_req,
if (wlen) {
ret = do_writev(s->fd, iov, wlen, aio_req->iov_offset);
if (ret) {
+ qemu_co_mutex_unlock(&s->lock);
error_report("failed to send a data, %s", strerror(errno));
return -EIO;
}
@@ -1713,7 +1715,7 @@ out:
return 1;
}
-static int sd_co_writev(BlockDriverState *bs, int64_t sector_num,
+static coroutine_fn int sd_co_writev(BlockDriverState *bs, int64_t sector_num,
int nb_sectors, QEMUIOVector *qiov)
{
SheepdogAIOCB *acb;
@@ -1742,7 +1744,7 @@ static int sd_co_writev(BlockDriverState *bs, int64_t sector_num,
return acb->ret;
}
-static int sd_co_readv(BlockDriverState *bs, int64_t sector_num,
+static coroutine_fn int sd_co_readv(BlockDriverState *bs, int64_t sector_num,
int nb_sectors, QEMUIOVector *qiov)
{
SheepdogAIOCB *acb;
diff --git a/block/vdi.c b/block/vdi.c
index 684a4a87b6..31cdfabdea 100644
--- a/block/vdi.c
+++ b/block/vdi.c
@@ -52,6 +52,7 @@
#include "qemu-common.h"
#include "block_int.h"
#include "module.h"
+#include "migration.h"
#if defined(CONFIG_UUID)
#include <uuid/uuid.h>
@@ -203,6 +204,8 @@ typedef struct {
uint32_t bmap_sector;
/* VDI header (converted to host endianness). */
VdiHeader header;
+
+ Error *migration_blocker;
} BDRVVdiState;
/* Change UUID from little endian (IPRT = VirtualBox format) to big endian
@@ -454,6 +457,12 @@ static int vdi_open(BlockDriverState *bs, int flags)
goto fail_free_bmap;
}
+ /* Disable migration when vdi images are used */
+ error_set(&s->migration_blocker,
+ QERR_BLOCK_FORMAT_FEATURE_NOT_SUPPORTED,
+ "vdi", bs->device_name, "live migration");
+ migrate_add_blocker(s->migration_blocker);
+
return 0;
fail_free_bmap:
@@ -463,8 +472,8 @@ static int vdi_open(BlockDriverState *bs, int flags)
return -1;
}
-static int vdi_is_allocated(BlockDriverState *bs, int64_t sector_num,
- int nb_sectors, int *pnum)
+static int coroutine_fn vdi_co_is_allocated(BlockDriverState *bs,
+ int64_t sector_num, int nb_sectors, int *pnum)
{
/* TODO: Check for too large sector_num (in bdrv_is_allocated or here). */
BDRVVdiState *s = (BDRVVdiState *)bs->opaque;
@@ -506,28 +515,26 @@ static VdiAIOCB *vdi_aio_setup(BlockDriverState *bs, int64_t sector_num,
bs, sector_num, qiov, nb_sectors, cb, opaque, is_write);
acb = qemu_aio_get(&vdi_aio_pool, bs, cb, opaque);
- if (acb) {
- acb->hd_aiocb = NULL;
- acb->sector_num = sector_num;
- acb->qiov = qiov;
- acb->is_write = is_write;
-
- if (qiov->niov > 1) {
- acb->buf = qemu_blockalign(bs, qiov->size);
- acb->orig_buf = acb->buf;
- if (is_write) {
- qemu_iovec_to_buffer(qiov, acb->buf);
- }
- } else {
- acb->buf = (uint8_t *)qiov->iov->iov_base;
+ acb->hd_aiocb = NULL;
+ acb->sector_num = sector_num;
+ acb->qiov = qiov;
+ acb->is_write = is_write;
+
+ if (qiov->niov > 1) {
+ acb->buf = qemu_blockalign(bs, qiov->size);
+ acb->orig_buf = acb->buf;
+ if (is_write) {
+ qemu_iovec_to_buffer(qiov, acb->buf);
}
- acb->nb_sectors = nb_sectors;
- acb->n_sectors = 0;
- acb->bmap_first = VDI_UNALLOCATED;
- acb->bmap_last = VDI_UNALLOCATED;
- acb->block_buffer = NULL;
- acb->header_modified = 0;
- }
+ } else {
+ acb->buf = (uint8_t *)qiov->iov->iov_base;
+ }
+ acb->nb_sectors = nb_sectors;
+ acb->n_sectors = 0;
+ acb->bmap_first = VDI_UNALLOCATED;
+ acb->bmap_last = VDI_UNALLOCATED;
+ acb->block_buffer = NULL;
+ acb->header_modified = 0;
return acb;
}
@@ -624,10 +631,6 @@ static void vdi_aio_read_cb(void *opaque, int ret)
qemu_iovec_init_external(&acb->hd_qiov, &acb->hd_iov, 1);
acb->hd_aiocb = bdrv_aio_readv(bs->file, offset, &acb->hd_qiov,
n_sectors, vdi_aio_read_cb, acb);
- if (acb->hd_aiocb == NULL) {
- ret = -EIO;
- goto done;
- }
}
return;
done:
@@ -648,10 +651,6 @@ static BlockDriverAIOCB *vdi_aio_readv(BlockDriverState *bs,
logout("\n");
acb = vdi_aio_setup(bs, sector_num, qiov, nb_sectors, cb, opaque, 0);
- if (!acb) {
- return NULL;
- }
-
ret = vdi_schedule_bh(vdi_aio_rw_bh, acb);
if (ret < 0) {
if (acb->qiov->niov > 1) {
@@ -699,10 +698,6 @@ static void vdi_aio_write_cb(void *opaque, int ret)
qemu_iovec_init_external(&acb->hd_qiov, &acb->hd_iov, 1);
acb->hd_aiocb = bdrv_aio_writev(bs->file, 0, &acb->hd_qiov, 1,
vdi_aio_write_cb, acb);
- if (acb->hd_aiocb == NULL) {
- ret = -EIO;
- goto done;
- }
return;
} else if (VDI_IS_ALLOCATED(acb->bmap_first)) {
/* One or more new blocks were allocated. */
@@ -729,10 +724,6 @@ static void vdi_aio_write_cb(void *opaque, int ret)
n_sectors, bmap_first);
acb->hd_aiocb = bdrv_aio_writev(bs->file, offset, &acb->hd_qiov,
n_sectors, vdi_aio_write_cb, acb);
- if (acb->hd_aiocb == NULL) {
- ret = -EIO;
- goto done;
- }
return;
}
ret = 0;
@@ -780,10 +771,6 @@ static void vdi_aio_write_cb(void *opaque, int ret)
acb->hd_aiocb = bdrv_aio_writev(bs->file, offset,
&acb->hd_qiov, s->block_sectors,
vdi_aio_write_cb, acb);
- if (acb->hd_aiocb == NULL) {
- ret = -EIO;
- goto done;
- }
} else {
uint64_t offset = s->header.offset_data / SECTOR_SIZE +
(uint64_t)bmap_entry * s->block_sectors +
@@ -793,10 +780,6 @@ static void vdi_aio_write_cb(void *opaque, int ret)
qemu_iovec_init_external(&acb->hd_qiov, &acb->hd_iov, 1);
acb->hd_aiocb = bdrv_aio_writev(bs->file, offset, &acb->hd_qiov,
n_sectors, vdi_aio_write_cb, acb);
- if (acb->hd_aiocb == NULL) {
- ret = -EIO;
- goto done;
- }
}
return;
@@ -818,10 +801,6 @@ static BlockDriverAIOCB *vdi_aio_writev(BlockDriverState *bs,
logout("\n");
acb = vdi_aio_setup(bs, sector_num, qiov, nb_sectors, cb, opaque, 1);
- if (!acb) {
- return NULL;
- }
-
ret = vdi_schedule_bh(vdi_aio_rw_bh, acb);
if (ret < 0) {
if (acb->qiov->niov > 1) {
@@ -939,6 +918,12 @@ static int vdi_create(const char *filename, QEMUOptionParameter *options)
static void vdi_close(BlockDriverState *bs)
{
+ BDRVVdiState *s = bs->opaque;
+
+ g_free(s->bmap);
+
+ migrate_del_blocker(s->migration_blocker);
+ error_free(s->migration_blocker);
}
static coroutine_fn int vdi_co_flush(BlockDriverState *bs)
@@ -981,7 +966,7 @@ static BlockDriver bdrv_vdi = {
.bdrv_close = vdi_close,
.bdrv_create = vdi_create,
.bdrv_co_flush_to_disk = vdi_co_flush,
- .bdrv_is_allocated = vdi_is_allocated,
+ .bdrv_co_is_allocated = vdi_co_is_allocated,
.bdrv_make_empty = vdi_make_empty,
.bdrv_aio_readv = vdi_aio_readv,
diff --git a/block/vmdk.c b/block/vmdk.c
index 96f7d5d90f..5623ac10cd 100644
--- a/block/vmdk.c
+++ b/block/vmdk.c
@@ -26,6 +26,7 @@
#include "qemu-common.h"
#include "block_int.h"
#include "module.h"
+#include "migration.h"
#include <zlib.h>
#define VMDK3_MAGIC (('C' << 24) | ('O' << 16) | ('W' << 8) | 'D')
@@ -97,6 +98,7 @@ typedef struct BDRVVmdkState {
int num_extents;
/* Extent array with num_extents entries, ascend ordered by address */
VmdkExtent *extents;
+ Error *migration_blocker;
} BDRVVmdkState;
typedef struct VmdkMetaData {
@@ -659,7 +661,14 @@ static int vmdk_open(BlockDriverState *bs, int flags)
}
s->parent_cid = vmdk_read_cid(bs, 1);
qemu_co_mutex_init(&s->lock);
- return ret;
+
+ /* Disable migration when VMDK images are used */
+ error_set(&s->migration_blocker,
+ QERR_BLOCK_FORMAT_FEATURE_NOT_SUPPORTED,
+ "vmdk", bs->device_name, "live migration");
+ migrate_add_blocker(s->migration_blocker);
+
+ return 0;
fail:
vmdk_free_extents(bs);
@@ -852,8 +861,8 @@ static VmdkExtent *find_extent(BDRVVmdkState *s,
return NULL;
}
-static int vmdk_is_allocated(BlockDriverState *bs, int64_t sector_num,
- int nb_sectors, int *pnum)
+static int coroutine_fn vmdk_co_is_allocated(BlockDriverState *bs,
+ int64_t sector_num, int nb_sectors, int *pnum)
{
BDRVVmdkState *s = bs->opaque;
int64_t index_in_cluster, n, ret;
@@ -864,8 +873,10 @@ static int vmdk_is_allocated(BlockDriverState *bs, int64_t sector_num,
if (!extent) {
return 0;
}
+ qemu_co_mutex_lock(&s->lock);
ret = get_cluster_offset(bs, extent, NULL,
sector_num * 512, 0, &offset);
+ qemu_co_mutex_unlock(&s->lock);
/* get_cluster_offset returning 0 means success */
ret = !ret;
@@ -1504,7 +1515,12 @@ exit:
static void vmdk_close(BlockDriverState *bs)
{
+ BDRVVmdkState *s = bs->opaque;
+
vmdk_free_extents(bs);
+
+ migrate_del_blocker(s->migration_blocker);
+ error_free(s->migration_blocker);
}
static coroutine_fn int vmdk_co_flush(BlockDriverState *bs)
@@ -1582,7 +1598,7 @@ static BlockDriver bdrv_vmdk = {
.bdrv_close = vmdk_close,
.bdrv_create = vmdk_create,
.bdrv_co_flush_to_disk = vmdk_co_flush,
- .bdrv_is_allocated = vmdk_is_allocated,
+ .bdrv_co_is_allocated = vmdk_co_is_allocated,
.bdrv_get_allocated_file_size = vmdk_get_allocated_file_size,
.create_options = vmdk_create_options,
diff --git a/block/vpc.c b/block/vpc.c
index 39a324705d..89a5ee2668 100644
--- a/block/vpc.c
+++ b/block/vpc.c
@@ -25,6 +25,7 @@
#include "qemu-common.h"
#include "block_int.h"
#include "module.h"
+#include "migration.h"
/**************************************************************/
@@ -128,6 +129,8 @@ typedef struct BDRVVPCState {
uint64_t last_bitmap;
#endif
+
+ Error *migration_blocker;
} BDRVVPCState;
static uint32_t vpc_checksum(uint8_t* buf, size_t size)
@@ -228,6 +231,13 @@ static int vpc_open(BlockDriverState *bs, int flags)
#endif
qemu_co_mutex_init(&s->lock);
+
+ /* Disable migration when VHD images are used */
+ error_set(&s->migration_blocker,
+ QERR_BLOCK_FORMAT_FEATURE_NOT_SUPPORTED,
+ "vpc", bs->device_name, "live migration");
+ migrate_add_blocker(s->migration_blocker);
+
return 0;
fail:
return err;
@@ -352,8 +362,11 @@ static int64_t alloc_block(BlockDriverState* bs, int64_t sector_num)
// Initialize the block's bitmap
memset(bitmap, 0xff, s->bitmap_size);
- bdrv_pwrite_sync(bs->file, s->free_data_block_offset, bitmap,
+ ret = bdrv_pwrite_sync(bs->file, s->free_data_block_offset, bitmap,
s->bitmap_size);
+ if (ret < 0) {
+ return ret;
+ }
// Write new footer (the old one will be overwritten)
s->free_data_block_offset += s->block_size + s->bitmap_size;
@@ -651,6 +664,9 @@ static void vpc_close(BlockDriverState *bs)
#ifdef CACHE
g_free(s->pageentry_u8);
#endif
+
+ migrate_del_blocker(s->migration_blocker);
+ error_free(s->migration_blocker);
}
static QEMUOptionParameter vpc_create_options[] = {
diff --git a/block/vvfat.c b/block/vvfat.c
index 131680f6f4..eeffc4a4a8 100644
--- a/block/vvfat.c
+++ b/block/vvfat.c
@@ -27,6 +27,7 @@
#include "qemu-common.h"
#include "block_int.h"
#include "module.h"
+#include "migration.h"
#ifndef S_IWGRP
#define S_IWGRP 0
@@ -350,6 +351,8 @@ typedef struct BDRVVVFATState {
array_t commits;
const char* path;
int downcase_short_names;
+
+ Error *migration_blocker;
} BDRVVVFATState;
/* take the sector position spos and convert it to Cylinder/Head/Sector position
@@ -1073,6 +1076,15 @@ DLOG(if (stderr == NULL) {
// assert(is_consistent(s));
qemu_co_mutex_init(&s->lock);
+
+ /* Disable migration when vvfat is used rw */
+ if (s->qcow) {
+ error_set(&s->migration_blocker,
+ QERR_BLOCK_FORMAT_FEATURE_NOT_SUPPORTED,
+ "vvfat (rw)", bs->device_name, "live migration");
+ migrate_add_blocker(s->migration_blocker);
+ }
+
return 0;
}
@@ -2746,7 +2758,7 @@ static coroutine_fn int vvfat_co_write(BlockDriverState *bs, int64_t sector_num,
return ret;
}
-static int vvfat_is_allocated(BlockDriverState *bs,
+static int coroutine_fn vvfat_co_is_allocated(BlockDriverState *bs,
int64_t sector_num, int nb_sectors, int* n)
{
BDRVVVFATState* s = bs->opaque;
@@ -2829,6 +2841,11 @@ static void vvfat_close(BlockDriverState *bs)
array_free(&(s->directory));
array_free(&(s->mapping));
g_free(s->cluster_buffer);
+
+ if (s->qcow) {
+ migrate_del_blocker(s->migration_blocker);
+ error_free(s->migration_blocker);
+ }
}
static BlockDriver bdrv_vvfat = {
@@ -2838,7 +2855,7 @@ static BlockDriver bdrv_vvfat = {
.bdrv_read = vvfat_co_read,
.bdrv_write = vvfat_co_write,
.bdrv_close = vvfat_close,
- .bdrv_is_allocated = vvfat_is_allocated,
+ .bdrv_co_is_allocated = vvfat_co_is_allocated,
.protocol_name = "fat",
};
generated by cgit v1.2.3 (git 2.26.2) at 2024-12-30 19:51:12 +0000