static int coroutine_fn
raw_co_writev_flags(BlockDriverState *bs, int64_t sector_num, int nb_sectors,
QEMUIOVector *qiov, int flags)
{
void *buf = NULL;
BlockDriver *drv;
QEMUIOVector local_qiov;
int ret;
if (bs->probed && sector_num == 0) {
/* As long as these conditions are true, we can't get partial writes to
* the probe buffer and can just directly check the request. */
QEMU_BUILD_BUG_ON(BLOCK_PROBE_BUF_SIZE != 512);
QEMU_BUILD_BUG_ON(BDRV_SECTOR_SIZE != 512);
if (nb_sectors == 0) {
/* qemu_iovec_to_buf() would fail, but we want to return success
* instead of -EINVAL in this case. */
return 0;
}
buf = qemu_try_blockalign(bs->file->bs, 512);
if (!buf) {
ret = -ENOMEM;
goto fail;
}
ret = qemu_iovec_to_buf(qiov, 0, buf, 512);
if (ret != 512) {
ret = -EINVAL;
goto fail;
}
drv = bdrv_probe_all(buf, 512, NULL);
if (drv != bs->drv) {
ret = -EPERM;
goto fail;
}
/* Use the checked buffer, a malicious guest might be overwriting its
* original buffer in the background. */
qemu_iovec_init(&local_qiov, qiov->niov + 1);
qemu_iovec_add(&local_qiov, buf, 512);
qemu_iovec_concat(&local_qiov, qiov, 512, qiov->size - 512);
qiov = &local_qiov;
}
BLKDBG_EVENT(bs->file, BLKDBG_WRITE_AIO);
ret = bdrv_co_do_pwritev(bs->file->bs, sector_num * BDRV_SECTOR_SIZE,
nb_sectors * BDRV_SECTOR_SIZE, qiov, flags);
fail:
if (qiov == &local_qiov) {
qemu_iovec_destroy(&local_qiov);
}
qemu_vfree(buf);
return ret;
}
int qcow2_snapshot_load_tmp(BlockDriverState *bs,
const char *snapshot_id,
const char *name,
Error **errp)
{
int i, snapshot_index;
BDRVQcow2State *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_and_name(bs, snapshot_id, name);
if (snapshot_index < 0) {
error_setg(errp,
"Can't find snapshot");
return -ENOENT;
}
sn = &s->snapshots[snapshot_index];
/* Allocate and read in the snapshot's L1 table */
if (sn->l1_size > QCOW_MAX_L1_SIZE / sizeof(uint64_t)) {
error_setg(errp, "Snapshot L1 table too large");
return -EFBIG;
}
new_l1_bytes = sn->l1_size * sizeof(uint64_t);
new_l1_table = qemu_try_blockalign(bs->file->bs,
align_offset(new_l1_bytes, 512));
if (new_l1_table == NULL) {
return -ENOMEM;
}
ret = bdrv_pread(bs->file->bs, sn->l1_table_offset,
new_l1_table, new_l1_bytes);
if (ret < 0) {
error_setg(errp, "Failed to read l1 table for snapshot");
qemu_vfree(new_l1_table);
return ret;
}
/* Switch the L1 table */
qemu_vfree(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;
}
BlockDriverAIOCB *win32_aio_submit(BlockDriverState *bs,
QEMUWin32AIOState *aio, HANDLE hfile,
int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
BlockDriverCompletionFunc *cb, void *opaque, int type)
{
struct QEMUWin32AIOCB *waiocb;
uint64_t offset = sector_num * 512;
DWORD rc;
waiocb = qemu_aio_get(&win32_aiocb_info, bs, cb, opaque);
waiocb->nbytes = nb_sectors * 512;
waiocb->qiov = qiov;
waiocb->is_read = (type == QEMU_AIO_READ);
if (qiov->niov > 1) {
waiocb->buf = qemu_try_blockalign(bs, qiov->size);
if (waiocb->buf == NULL) {
goto out;
}
if (type & QEMU_AIO_WRITE) {
iov_to_buf(qiov->iov, qiov->niov, 0, waiocb->buf, qiov->size);
}
waiocb->is_linear = false;
} else {
waiocb->buf = qiov->iov[0].iov_base;
waiocb->is_linear = true;
}
memset(&waiocb->ov, 0, sizeof(waiocb->ov));
waiocb->ov.Offset = (DWORD)offset;
waiocb->ov.OffsetHigh = (DWORD)(offset >> 32);
waiocb->ov.hEvent = event_notifier_get_handle(&aio->e);
aio->count++;
if (type & QEMU_AIO_READ) {
rc = ReadFile(hfile, waiocb->buf, waiocb->nbytes, NULL, &waiocb->ov);
} else {
rc = WriteFile(hfile, waiocb->buf, waiocb->nbytes, NULL, &waiocb->ov);
}
if(rc == 0 && GetLastError() != ERROR_IO_PENDING) {
goto out_dec_count;
}
return &waiocb->common;
out_dec_count:
aio->count--;
out:
qemu_aio_release(waiocb);
return NULL;
}
Qcow2Cache *qcow2_cache_create(BlockDriverState *bs, int num_tables)
{
BDRVQcowState *s = bs->opaque;
Qcow2Cache *c;
c = g_new0(Qcow2Cache, 1);
c->size = num_tables;
c->entries = g_try_new0(Qcow2CachedTable, num_tables);
c->table_array = qemu_try_blockalign(bs->file,
(size_t) num_tables * s->cluster_size);
if (!c->entries || !c->table_array) {
qemu_vfree(c->table_array);
g_free(c->entries);
g_free(c);
c = NULL;
}
return c;
}
static void coroutine_fn mirror_run(void *opaque)
{
MirrorBlockJob *s = opaque;
MirrorExitData *data;
BlockDriverState *bs = s->source;
BlockDriverState *target_bs = blk_bs(s->target);
bool need_drain = true;
int64_t length;
BlockDriverInfo bdi;
char backing_filename[2]; /* we only need 2 characters because we are only
checking for a NULL string */
int ret = 0;
if (block_job_is_cancelled(&s->common)) {
goto immediate_exit;
}
s->bdev_length = bdrv_getlength(bs);
if (s->bdev_length < 0) {
ret = s->bdev_length;
goto immediate_exit;
}
/* Active commit must resize the base image if its size differs from the
* active layer. */
if (s->base == blk_bs(s->target)) {
int64_t base_length;
base_length = blk_getlength(s->target);
if (base_length < 0) {
ret = base_length;
goto immediate_exit;
}
if (s->bdev_length > base_length) {
ret = blk_truncate(s->target, s->bdev_length, PREALLOC_MODE_OFF,
NULL);
if (ret < 0) {
goto immediate_exit;
}
}
}
if (s->bdev_length == 0) {
/* Report BLOCK_JOB_READY and wait for complete. */
block_job_event_ready(&s->common);
s->synced = true;
while (!block_job_is_cancelled(&s->common) && !s->should_complete) {
block_job_yield(&s->common);
}
s->common.cancelled = false;
goto immediate_exit;
}
length = DIV_ROUND_UP(s->bdev_length, s->granularity);
s->in_flight_bitmap = bitmap_new(length);
/* If we have no backing file yet in the destination, we cannot let
* the destination do COW. Instead, we copy sectors around the
* dirty data if needed. We need a bitmap to do that.
*/
bdrv_get_backing_filename(target_bs, backing_filename,
sizeof(backing_filename));
if (!bdrv_get_info(target_bs, &bdi) && bdi.cluster_size) {
s->target_cluster_size = bdi.cluster_size;
} else {
s->target_cluster_size = BDRV_SECTOR_SIZE;
}
if (backing_filename[0] && !target_bs->backing &&
s->granularity < s->target_cluster_size) {
s->buf_size = MAX(s->buf_size, s->target_cluster_size);
s->cow_bitmap = bitmap_new(length);
}
s->max_iov = MIN(bs->bl.max_iov, target_bs->bl.max_iov);
s->buf = qemu_try_blockalign(bs, s->buf_size);
if (s->buf == NULL) {
ret = -ENOMEM;
goto immediate_exit;
}
mirror_free_init(s);
s->last_pause_ns = qemu_clock_get_ns(QEMU_CLOCK_REALTIME);
if (!s->is_none_mode) {
ret = mirror_dirty_init(s);
if (ret < 0 || block_job_is_cancelled(&s->common)) {
goto immediate_exit;
}
}
assert(!s->dbi);
s->dbi = bdrv_dirty_iter_new(s->dirty_bitmap);
for (;;) {
uint64_t delay_ns = 0;
int64_t cnt, delta;
bool should_complete;
if (s->ret < 0) {
ret = s->ret;
goto immediate_exit;
}
block_job_pause_point(&s->common);
cnt = bdrv_get_dirty_count(s->dirty_bitmap);
/* cnt is the number of dirty bytes remaining and s->bytes_in_flight is
* the number of bytes currently being processed; together those are
* the current remaining operation length */
block_job_progress_set_remaining(&s->common, s->bytes_in_flight + cnt);
/* Note that even when no rate limit is applied we need to yield
* periodically with no pending I/O so that bdrv_drain_all() returns.
* We do so every BLKOCK_JOB_SLICE_TIME nanoseconds, or when there is
* an error, or when the source is clean, whichever comes first. */
delta = qemu_clock_get_ns(QEMU_CLOCK_REALTIME) - s->last_pause_ns;
if (delta < BLOCK_JOB_SLICE_TIME &&
s->common.iostatus == BLOCK_DEVICE_IO_STATUS_OK) {
if (s->in_flight >= MAX_IN_FLIGHT || s->buf_free_count == 0 ||
(cnt == 0 && s->in_flight > 0)) {
trace_mirror_yield(s, cnt, s->buf_free_count, s->in_flight);
mirror_wait_for_io(s);
continue;
} else if (cnt != 0) {
delay_ns = mirror_iteration(s);
}
}
should_complete = false;
if (s->in_flight == 0 && cnt == 0) {
trace_mirror_before_flush(s);
if (!s->synced) {
if (mirror_flush(s) < 0) {
/* Go check s->ret. */
continue;
}
/* We're out of the streaming phase. From now on, if the job
* is cancelled we will actually complete all pending I/O and
* report completion. This way, block-job-cancel will leave
* the target in a consistent state.
*/
block_job_event_ready(&s->common);
s->synced = true;
}
should_complete = s->should_complete ||
block_job_is_cancelled(&s->common);
cnt = bdrv_get_dirty_count(s->dirty_bitmap);
}
if (cnt == 0 && should_complete) {
/* The dirty bitmap is not updated while operations are pending.
* If we're about to exit, wait for pending operations before
* calling bdrv_get_dirty_count(bs), or we may exit while the
* source has dirty data to copy!
*
* Note that I/O can be submitted by the guest while
* mirror_populate runs, so pause it now. Before deciding
* whether to switch to target check one last time if I/O has
* come in the meanwhile, and if not flush the data to disk.
*/
trace_mirror_before_drain(s, cnt);
bdrv_drained_begin(bs);
cnt = bdrv_get_dirty_count(s->dirty_bitmap);
if (cnt > 0 || mirror_flush(s) < 0) {
bdrv_drained_end(bs);
continue;
}
/* The two disks are in sync. Exit and report successful
* completion.
*/
assert(QLIST_EMPTY(&bs->tracked_requests));
s->common.cancelled = false;
need_drain = false;
break;
}
ret = 0;
if (s->synced && !should_complete) {
delay_ns = (s->in_flight == 0 &&
cnt == 0 ? BLOCK_JOB_SLICE_TIME : 0);
}
trace_mirror_before_sleep(s, cnt, s->synced, delay_ns);
block_job_sleep_ns(&s->common, delay_ns);
if (block_job_is_cancelled(&s->common) &&
(!s->synced || s->common.force))
{
break;
}
s->last_pause_ns = qemu_clock_get_ns(QEMU_CLOCK_REALTIME);
}
immediate_exit:
if (s->in_flight > 0) {
/* We get here only if something went wrong. Either the job failed,
* or it was cancelled prematurely so that we do not guarantee that
* the target is a copy of the source.
*/
assert(ret < 0 || ((s->common.force || !s->synced) &&
block_job_is_cancelled(&s->common)));
assert(need_drain);
mirror_wait_for_all_io(s);
}
assert(s->in_flight == 0);
qemu_vfree(s->buf);
g_free(s->cow_bitmap);
g_free(s->in_flight_bitmap);
bdrv_dirty_iter_free(s->dbi);
data = g_malloc(sizeof(*data));
data->ret = ret;
if (need_drain) {
bdrv_drained_begin(bs);
}
block_job_defer_to_main_loop(&s->common, mirror_exit, data);
}
static void coroutine_fn mirror_run(void *opaque)
{
MirrorBlockJob *s = opaque;
MirrorExitData *data;
BlockDriverState *bs = s->common.bs;
int64_t sector_num, end, length;
uint64_t last_pause_ns;
BlockDriverInfo bdi;
char backing_filename[2]; /* we only need 2 characters because we are only
checking for a NULL string */
int ret = 0;
int n;
if (block_job_is_cancelled(&s->common)) {
goto immediate_exit;
}
s->bdev_length = bdrv_getlength(bs);
if (s->bdev_length < 0) {
ret = s->bdev_length;
goto immediate_exit;
} else if (s->bdev_length == 0) {
/* Report BLOCK_JOB_READY and wait for complete. */
block_job_event_ready(&s->common);
s->synced = true;
while (!block_job_is_cancelled(&s->common) && !s->should_complete) {
block_job_yield(&s->common);
}
s->common.cancelled = false;
goto immediate_exit;
}
length = DIV_ROUND_UP(s->bdev_length, s->granularity);
s->in_flight_bitmap = bitmap_new(length);
/* If we have no backing file yet in the destination, we cannot let
* the destination do COW. Instead, we copy sectors around the
* dirty data if needed. We need a bitmap to do that.
*/
bdrv_get_backing_filename(s->target, backing_filename,
sizeof(backing_filename));
if (backing_filename[0] && !s->target->backing) {
ret = bdrv_get_info(s->target, &bdi);
if (ret < 0) {
goto immediate_exit;
}
if (s->granularity < bdi.cluster_size) {
s->buf_size = MAX(s->buf_size, bdi.cluster_size);
s->cow_bitmap = bitmap_new(length);
}
}
end = s->bdev_length / BDRV_SECTOR_SIZE;
s->buf = qemu_try_blockalign(bs, s->buf_size);
if (s->buf == NULL) {
ret = -ENOMEM;
goto immediate_exit;
}
mirror_free_init(s);
last_pause_ns = qemu_clock_get_ns(QEMU_CLOCK_REALTIME);
if (!s->is_none_mode) {
/* First part, loop on the sectors and initialize the dirty bitmap. */
BlockDriverState *base = s->base;
bool mark_all_dirty = s->base == NULL && !bdrv_has_zero_init(s->target);
for (sector_num = 0; sector_num < end; ) {
/* Just to make sure we are not exceeding int limit. */
int nb_sectors = MIN(INT_MAX >> BDRV_SECTOR_BITS,
end - sector_num);
int64_t now = qemu_clock_get_ns(QEMU_CLOCK_REALTIME);
if (now - last_pause_ns > SLICE_TIME) {
last_pause_ns = now;
block_job_sleep_ns(&s->common, QEMU_CLOCK_REALTIME, 0);
}
if (block_job_is_cancelled(&s->common)) {
goto immediate_exit;
}
ret = bdrv_is_allocated_above(bs, base, sector_num, nb_sectors, &n);
if (ret < 0) {
goto immediate_exit;
}
assert(n > 0);
if (ret == 1 || mark_all_dirty) {
bdrv_set_dirty_bitmap(s->dirty_bitmap, sector_num, n);
}
sector_num += n;
}
}
static BlockAIOCB *rbd_start_aio(BlockDriverState *bs,
int64_t off,
QEMUIOVector *qiov,
int64_t size,
BlockCompletionFunc *cb,
void *opaque,
RBDAIOCmd cmd)
{
RBDAIOCB *acb;
RADOSCB *rcb = NULL;
rbd_completion_t c;
int r;
BDRVRBDState *s = bs->opaque;
acb = qemu_aio_get(&rbd_aiocb_info, bs, cb, opaque);
acb->cmd = cmd;
acb->qiov = qiov;
assert(!qiov || qiov->size == size);
rcb = g_new(RADOSCB, 1);
if (!LIBRBD_USE_IOVEC) {
if (cmd == RBD_AIO_DISCARD || cmd == RBD_AIO_FLUSH) {
acb->bounce = NULL;
} else {
acb->bounce = qemu_try_blockalign(bs, qiov->size);
if (acb->bounce == NULL) {
goto failed;
}
}
if (cmd == RBD_AIO_WRITE) {
qemu_iovec_to_buf(acb->qiov, 0, acb->bounce, qiov->size);
}
rcb->buf = acb->bounce;
}
acb->ret = 0;
acb->error = 0;
acb->s = s;
rcb->acb = acb;
rcb->s = acb->s;
rcb->size = size;
r = rbd_aio_create_completion(rcb, (rbd_callback_t) rbd_finish_aiocb, &c);
if (r < 0) {
goto failed;
}
switch (cmd) {
case RBD_AIO_WRITE:
#ifdef LIBRBD_SUPPORTS_IOVEC
r = rbd_aio_writev(s->image, qiov->iov, qiov->niov, off, c);
#else
r = rbd_aio_write(s->image, off, size, rcb->buf, c);
#endif
break;
case RBD_AIO_READ:
#ifdef LIBRBD_SUPPORTS_IOVEC
r = rbd_aio_readv(s->image, qiov->iov, qiov->niov, off, c);
#else
r = rbd_aio_read(s->image, off, size, rcb->buf, c);
#endif
break;
case RBD_AIO_DISCARD:
r = rbd_aio_discard_wrapper(s->image, off, size, c);
break;
case RBD_AIO_FLUSH:
r = rbd_aio_flush_wrapper(s->image, c);
break;
default:
r = -EINVAL;
}
if (r < 0) {
goto failed_completion;
}
return &acb->common;
failed_completion:
rbd_aio_release(c);
failed:
g_free(rcb);
if (!LIBRBD_USE_IOVEC) {
qemu_vfree(acb->bounce);
}
qemu_aio_unref(acb);
return NULL;
}
static int dmg_open(BlockDriverState *bs, QDict *options, int flags,
Error **errp)
{
BDRVDMGState *s = bs->opaque;
DmgHeaderState ds;
uint64_t rsrc_fork_offset, rsrc_fork_length;
uint64_t plist_xml_offset, plist_xml_length;
int64_t offset;
int ret;
block_module_load_one("dmg-bz2");
bs->read_only = true;
s->n_chunks = 0;
s->offsets = s->lengths = s->sectors = s->sectorcounts = NULL;
/* used by dmg_read_mish_block to keep track of the current I/O position */
ds.data_fork_offset = 0;
ds.max_compressed_size = 1;
ds.max_sectors_per_chunk = 1;
/* locate the UDIF trailer */
offset = dmg_find_koly_offset(bs->file, errp);
if (offset < 0) {
ret = offset;
goto fail;
}
/* offset of data fork (DataForkOffset) */
ret = read_uint64(bs, offset + 0x18, &ds.data_fork_offset);
if (ret < 0) {
goto fail;
} else if (ds.data_fork_offset > offset) {
ret = -EINVAL;
goto fail;
}
/* offset of resource fork (RsrcForkOffset) */
ret = read_uint64(bs, offset + 0x28, &rsrc_fork_offset);
if (ret < 0) {
goto fail;
}
ret = read_uint64(bs, offset + 0x30, &rsrc_fork_length);
if (ret < 0) {
goto fail;
}
if (rsrc_fork_offset >= offset ||
rsrc_fork_length > offset - rsrc_fork_offset) {
ret = -EINVAL;
goto fail;
}
/* offset of property list (XMLOffset) */
ret = read_uint64(bs, offset + 0xd8, &plist_xml_offset);
if (ret < 0) {
goto fail;
}
ret = read_uint64(bs, offset + 0xe0, &plist_xml_length);
if (ret < 0) {
goto fail;
}
if (plist_xml_offset >= offset ||
plist_xml_length > offset - plist_xml_offset) {
ret = -EINVAL;
goto fail;
}
ret = read_uint64(bs, offset + 0x1ec, (uint64_t *)&bs->total_sectors);
if (ret < 0) {
goto fail;
}
if (bs->total_sectors < 0) {
ret = -EINVAL;
goto fail;
}
if (rsrc_fork_length != 0) {
ret = dmg_read_resource_fork(bs, &ds,
rsrc_fork_offset, rsrc_fork_length);
if (ret < 0) {
goto fail;
}
} else if (plist_xml_length != 0) {
ret = dmg_read_plist_xml(bs, &ds, plist_xml_offset, plist_xml_length);
if (ret < 0) {
goto fail;
}
} else {
ret = -EINVAL;
goto fail;
}
/* initialize zlib engine */
s->compressed_chunk = qemu_try_blockalign(bs->file->bs,
ds.max_compressed_size + 1);
s->uncompressed_chunk = qemu_try_blockalign(bs->file->bs,
512 * ds.max_sectors_per_chunk);
if (s->compressed_chunk == NULL || s->uncompressed_chunk == NULL) {
ret = -ENOMEM;
goto fail;
}
if (inflateInit(&s->zstream) != Z_OK) {
ret = -EINVAL;
goto fail;
}
s->current_chunk = s->n_chunks;
qemu_co_mutex_init(&s->lock);
return 0;
fail:
g_free(s->types);
g_free(s->offsets);
g_free(s->lengths);
g_free(s->sectors);
g_free(s->sectorcounts);
qemu_vfree(s->compressed_chunk);
qemu_vfree(s->uncompressed_chunk);
return ret;
}
static void coroutine_fn mirror_run(void *opaque)
{
MirrorBlockJob *s = opaque;
MirrorExitData *data;
BlockDriverState *bs = s->common.bs;
int64_t sector_num, end, sectors_per_chunk, length;
uint64_t last_pause_ns;
BlockDriverInfo bdi;
char backing_filename[2]; /* we only need 2 characters because we are only
checking for a NULL string */
int ret = 0;
int n;
if (block_job_is_cancelled(&s->common)) {
goto immediate_exit;
}
s->bdev_length = bdrv_getlength(bs);
if (s->bdev_length < 0) {
ret = s->bdev_length;
goto immediate_exit;
} else if (s->bdev_length == 0) {
/* Report BLOCK_JOB_READY and wait for complete. */
block_job_event_ready(&s->common);
s->synced = true;
while (!block_job_is_cancelled(&s->common) && !s->should_complete) {
block_job_yield(&s->common);
}
s->common.cancelled = false;
goto immediate_exit;
}
length = DIV_ROUND_UP(s->bdev_length, s->granularity);
s->in_flight_bitmap = bitmap_new(length);
/* If we have no backing file yet in the destination, we cannot let
* the destination do COW. Instead, we copy sectors around the
* dirty data if needed. We need a bitmap to do that.
*/
bdrv_get_backing_filename(s->target, backing_filename,
sizeof(backing_filename));
if (backing_filename[0] && !s->target->backing_hd) {
ret = bdrv_get_info(s->target, &bdi);
if (ret < 0) {
goto immediate_exit;
}
if (s->granularity < bdi.cluster_size) {
s->buf_size = MAX(s->buf_size, bdi.cluster_size);
s->cow_bitmap = bitmap_new(length);
}
}
end = s->bdev_length / BDRV_SECTOR_SIZE;
s->buf = qemu_try_blockalign(bs, s->buf_size);
if (s->buf == NULL) {
ret = -ENOMEM;
goto immediate_exit;
}
sectors_per_chunk = s->granularity >> BDRV_SECTOR_BITS;
mirror_free_init(s);
if (!s->is_none_mode) {
/* First part, loop on the sectors and initialize the dirty bitmap. */
BlockDriverState *base = s->base;
for (sector_num = 0; sector_num < end; ) {
int64_t next = (sector_num | (sectors_per_chunk - 1)) + 1;
ret = bdrv_is_allocated_above(bs, base,
sector_num, next - sector_num, &n);
if (ret < 0) {
goto immediate_exit;
}
assert(n > 0);
if (ret == 1) {
bdrv_set_dirty_bitmap(s->dirty_bitmap, sector_num, n);
sector_num = next;
} else {
sector_num += n;
}
}
}
bdrv_dirty_iter_init(s->dirty_bitmap, &s->hbi);
last_pause_ns = qemu_clock_get_ns(QEMU_CLOCK_REALTIME);
for (;;) {
uint64_t delay_ns = 0;
int64_t cnt;
bool should_complete;
if (s->ret < 0) {
ret = s->ret;
goto immediate_exit;
}
cnt = bdrv_get_dirty_count(s->dirty_bitmap);
/* s->common.offset contains the number of bytes already processed so
* far, cnt is the number of dirty sectors remaining and
* s->sectors_in_flight is the number of sectors currently being
* processed; together those are the current total operation length */
s->common.len = s->common.offset +
(cnt + s->sectors_in_flight) * BDRV_SECTOR_SIZE;
/* Note that even when no rate limit is applied we need to yield
* periodically with no pending I/O so that bdrv_drain_all() returns.
* We do so every SLICE_TIME nanoseconds, or when there is an error,
* or when the source is clean, whichever comes first.
*/
if (qemu_clock_get_ns(QEMU_CLOCK_REALTIME) - last_pause_ns < SLICE_TIME &&
s->common.iostatus == BLOCK_DEVICE_IO_STATUS_OK) {
if (s->in_flight == MAX_IN_FLIGHT || s->buf_free_count == 0 ||
(cnt == 0 && s->in_flight > 0)) {
trace_mirror_yield(s, s->in_flight, s->buf_free_count, cnt);
qemu_coroutine_yield();
continue;
} else if (cnt != 0) {
delay_ns = mirror_iteration(s);
}
}
should_complete = false;
if (s->in_flight == 0 && cnt == 0) {
trace_mirror_before_flush(s);
ret = bdrv_flush(s->target);
if (ret < 0) {
if (mirror_error_action(s, false, -ret) ==
BLOCK_ERROR_ACTION_REPORT) {
goto immediate_exit;
}
} else {
/* We're out of the streaming phase. From now on, if the job
* is cancelled we will actually complete all pending I/O and
* report completion. This way, block-job-cancel will leave
* the target in a consistent state.
*/
if (!s->synced) {
block_job_event_ready(&s->common);
s->synced = true;
}
should_complete = s->should_complete ||
block_job_is_cancelled(&s->common);
cnt = bdrv_get_dirty_count(s->dirty_bitmap);
}
}
if (cnt == 0 && should_complete) {
/* The dirty bitmap is not updated while operations are pending.
* If we're about to exit, wait for pending operations before
* calling bdrv_get_dirty_count(bs), or we may exit while the
* source has dirty data to copy!
*
* Note that I/O can be submitted by the guest while
* mirror_populate runs.
*/
trace_mirror_before_drain(s, cnt);
bdrv_drain(bs);
cnt = bdrv_get_dirty_count(s->dirty_bitmap);
}
ret = 0;
trace_mirror_before_sleep(s, cnt, s->synced, delay_ns);
if (!s->synced) {
block_job_sleep_ns(&s->common, QEMU_CLOCK_REALTIME, delay_ns);
if (block_job_is_cancelled(&s->common)) {
break;
}
} else if (!should_complete) {
delay_ns = (s->in_flight == 0 && cnt == 0 ? SLICE_TIME : 0);
block_job_sleep_ns(&s->common, QEMU_CLOCK_REALTIME, delay_ns);
} else if (cnt == 0) {
/* The two disks are in sync. Exit and report successful
* completion.
*/
assert(QLIST_EMPTY(&bs->tracked_requests));
s->common.cancelled = false;
break;
}
last_pause_ns = qemu_clock_get_ns(QEMU_CLOCK_REALTIME);
}
immediate_exit:
if (s->in_flight > 0) {
/* We get here only if something went wrong. Either the job failed,
* or it was cancelled prematurely so that we do not guarantee that
* the target is a copy of the source.
*/
assert(ret < 0 || (!s->synced && block_job_is_cancelled(&s->common)));
mirror_drain(s);
}
assert(s->in_flight == 0);
qemu_vfree(s->buf);
g_free(s->cow_bitmap);
g_free(s->in_flight_bitmap);
bdrv_release_dirty_bitmap(bs, s->dirty_bitmap);
bdrv_iostatus_disable(s->target);
data = g_malloc(sizeof(*data));
data->ret = ret;
block_job_defer_to_main_loop(&s->common, mirror_exit, data);
}
static coroutine_fn int
block_crypto_co_writev(BlockDriverState *bs, int64_t sector_num,
int remaining_sectors, QEMUIOVector *qiov)
{
BlockCrypto *crypto = bs->opaque;
int cur_nr_sectors; /* number of sectors in current iteration */
uint64_t bytes_done = 0;
uint8_t *cipher_data = NULL;
QEMUIOVector hd_qiov;
int ret = 0;
size_t payload_offset =
qcrypto_block_get_payload_offset(crypto->block) / 512;
qemu_iovec_init(&hd_qiov, qiov->niov);
/* Bounce buffer so we have a linear mem region for
* entire sector. XXX optimize so we avoid bounce
* buffer in case that qiov->niov == 1
*/
cipher_data =
qemu_try_blockalign(bs->file->bs, MIN(BLOCK_CRYPTO_MAX_SECTORS * 512,
qiov->size));
if (cipher_data == NULL) {
ret = -ENOMEM;
goto cleanup;
}
while (remaining_sectors) {
cur_nr_sectors = remaining_sectors;
if (cur_nr_sectors > BLOCK_CRYPTO_MAX_SECTORS) {
cur_nr_sectors = BLOCK_CRYPTO_MAX_SECTORS;
}
qemu_iovec_to_buf(qiov, bytes_done,
cipher_data, cur_nr_sectors * 512);
if (qcrypto_block_encrypt(crypto->block,
sector_num,
cipher_data, cur_nr_sectors * 512,
NULL) < 0) {
ret = -EIO;
goto cleanup;
}
qemu_iovec_reset(&hd_qiov);
qemu_iovec_add(&hd_qiov, cipher_data, cur_nr_sectors * 512);
ret = bdrv_co_writev(bs->file,
payload_offset + sector_num,
cur_nr_sectors, &hd_qiov);
if (ret < 0) {
goto cleanup;
}
remaining_sectors -= cur_nr_sectors;
sector_num += cur_nr_sectors;
bytes_done += cur_nr_sectors * 512;
}
cleanup:
qemu_iovec_destroy(&hd_qiov);
qemu_vfree(cipher_data);
return ret;
}
static BlockDriverAIOCB *rbd_start_aio(BlockDriverState *bs,
int64_t sector_num,
QEMUIOVector *qiov,
int nb_sectors,
BlockDriverCompletionFunc *cb,
void *opaque,
RBDAIOCmd cmd)
{
RBDAIOCB *acb;
RADOSCB *rcb = NULL;
rbd_completion_t c;
int64_t off, size;
char *buf;
int r;
BDRVRBDState *s = bs->opaque;
acb = qemu_aio_get(&rbd_aiocb_info, bs, cb, opaque);
acb->cmd = cmd;
acb->qiov = qiov;
if (cmd == RBD_AIO_DISCARD || cmd == RBD_AIO_FLUSH) {
acb->bounce = NULL;
} else {
acb->bounce = qemu_try_blockalign(bs, qiov->size);
if (acb->bounce == NULL) {
goto failed;
}
}
acb->ret = 0;
acb->error = 0;
acb->s = s;
acb->cancelled = 0;
acb->bh = NULL;
acb->status = -EINPROGRESS;
if (cmd == RBD_AIO_WRITE) {
qemu_iovec_to_buf(acb->qiov, 0, acb->bounce, qiov->size);
}
buf = acb->bounce;
off = sector_num * BDRV_SECTOR_SIZE;
size = nb_sectors * BDRV_SECTOR_SIZE;
rcb = g_new(RADOSCB, 1);
rcb->done = 0;
rcb->acb = acb;
rcb->buf = buf;
rcb->s = acb->s;
rcb->size = size;
r = rbd_aio_create_completion(rcb, (rbd_callback_t) rbd_finish_aiocb, &c);
if (r < 0) {
goto failed;
}
switch (cmd) {
case RBD_AIO_WRITE:
r = rbd_aio_write(s->image, off, size, buf, c);
break;
case RBD_AIO_READ:
r = rbd_aio_read(s->image, off, size, buf, c);
break;
case RBD_AIO_DISCARD:
r = rbd_aio_discard_wrapper(s->image, off, size, c);
break;
case RBD_AIO_FLUSH:
r = rbd_aio_flush_wrapper(s->image, c);
break;
default:
r = -EINVAL;
}
if (r < 0) {
goto failed_completion;
}
return &acb->common;
failed_completion:
rbd_aio_release(c);
failed:
g_free(rcb);
qemu_vfree(acb->bounce);
qemu_aio_release(acb);
return NULL;
}
static int vpc_open(BlockDriverState *bs, QDict *options, int flags,
Error **errp)
{
BDRVVPCState *s = bs->opaque;
int i;
VHDFooter *footer;
VHDDynDiskHeader *dyndisk_header;
uint8_t buf[HEADER_SIZE];
uint32_t checksum;
uint64_t computed_size;
uint64_t pagetable_size;
int disk_type = VHD_DYNAMIC;
int ret;
ret = bdrv_pread(bs->file->bs, 0, s->footer_buf, HEADER_SIZE);
if (ret < 0) {
goto fail;
}
footer = (VHDFooter *) s->footer_buf;
if (strncmp(footer->creator, "conectix", 8)) {
int64_t offset = bdrv_getlength(bs->file->bs);
if (offset < 0) {
ret = offset;
goto fail;
} else if (offset < HEADER_SIZE) {
ret = -EINVAL;
goto fail;
}
/* If a fixed disk, the footer is found only at the end of the file */
ret = bdrv_pread(bs->file->bs, offset-HEADER_SIZE, s->footer_buf,
HEADER_SIZE);
if (ret < 0) {
goto fail;
}
if (strncmp(footer->creator, "conectix", 8)) {
error_setg(errp, "invalid VPC image");
ret = -EINVAL;
goto fail;
}
disk_type = VHD_FIXED;
}
checksum = be32_to_cpu(footer->checksum);
footer->checksum = 0;
if (vpc_checksum(s->footer_buf, HEADER_SIZE) != checksum)
fprintf(stderr, "block-vpc: The header checksum of '%s' is "
"incorrect.\n", bs->filename);
/* Write 'checksum' back to footer, or else will leave it with zero. */
footer->checksum = cpu_to_be32(checksum);
// The visible size of a image in Virtual PC depends on the geometry
// rather than on the size stored in the footer (the size in the footer
// is too large usually)
bs->total_sectors = (int64_t)
be16_to_cpu(footer->cyls) * footer->heads * footer->secs_per_cyl;
/* Images that have exactly the maximum geometry are probably bigger and
* would be truncated if we adhered to the geometry for them. Rely on
* footer->current_size for them. */
if (bs->total_sectors == VHD_MAX_GEOMETRY) {
bs->total_sectors = be64_to_cpu(footer->current_size) /
BDRV_SECTOR_SIZE;
}
/* Allow a maximum disk size of approximately 2 TB */
if (bs->total_sectors >= VHD_MAX_SECTORS) {
ret = -EFBIG;
goto fail;
}
if (disk_type == VHD_DYNAMIC) {
ret = bdrv_pread(bs->file->bs, be64_to_cpu(footer->data_offset), buf,
HEADER_SIZE);
if (ret < 0) {
goto fail;
}
dyndisk_header = (VHDDynDiskHeader *) buf;
if (strncmp(dyndisk_header->magic, "cxsparse", 8)) {
ret = -EINVAL;
goto fail;
}
s->block_size = be32_to_cpu(dyndisk_header->block_size);
if (!is_power_of_2(s->block_size) || s->block_size < BDRV_SECTOR_SIZE) {
error_setg(errp, "Invalid block size %" PRIu32, s->block_size);
ret = -EINVAL;
goto fail;
}
s->bitmap_size = ((s->block_size / (8 * 512)) + 511) & ~511;
s->max_table_entries = be32_to_cpu(dyndisk_header->max_table_entries);
if ((bs->total_sectors * 512) / s->block_size > 0xffffffffU) {
ret = -EINVAL;
goto fail;
}
if (s->max_table_entries > (VHD_MAX_SECTORS * 512) / s->block_size) {
ret = -EINVAL;
goto fail;
}
computed_size = (uint64_t) s->max_table_entries * s->block_size;
if (computed_size < bs->total_sectors * 512) {
ret = -EINVAL;
goto fail;
}
if (s->max_table_entries > SIZE_MAX / 4 ||
s->max_table_entries > (int) INT_MAX / 4) {
error_setg(errp, "Max Table Entries too large (%" PRId32 ")",
s->max_table_entries);
ret = -EINVAL;
goto fail;
}
pagetable_size = (uint64_t) s->max_table_entries * 4;
s->pagetable = qemu_try_blockalign(bs->file->bs, pagetable_size);
if (s->pagetable == NULL) {
ret = -ENOMEM;
goto fail;
}
s->bat_offset = be64_to_cpu(dyndisk_header->table_offset);
ret = bdrv_pread(bs->file->bs, s->bat_offset, s->pagetable,
pagetable_size);
if (ret < 0) {
goto fail;
}
s->free_data_block_offset =
ROUND_UP(s->bat_offset + pagetable_size, 512);
for (i = 0; i < s->max_table_entries; i++) {
be32_to_cpus(&s->pagetable[i]);
if (s->pagetable[i] != 0xFFFFFFFF) {
int64_t next = (512 * (int64_t) s->pagetable[i]) +
s->bitmap_size + s->block_size;
if (next > s->free_data_block_offset) {
s->free_data_block_offset = next;
}
}
}
if (s->free_data_block_offset > bdrv_getlength(bs->file->bs)) {
error_setg(errp, "block-vpc: free_data_block_offset points after "
"the end of file. The image has been truncated.");
ret = -EINVAL;
goto fail;
}
s->last_bitmap_offset = (int64_t) -1;
#ifdef CACHE
s->pageentry_u8 = g_malloc(512);
s->pageentry_u32 = s->pageentry_u8;
s->pageentry_u16 = s->pageentry_u8;
s->last_pagetable = -1;
#endif
}
qemu_co_mutex_init(&s->lock);
/* Disable migration when VHD images are used */
error_setg(&s->migration_blocker, "The vpc format used by node '%s' "
"does not support live migration",
bdrv_get_device_or_node_name(bs));
migrate_add_blocker(s->migration_blocker);
return 0;
fail:
qemu_vfree(s->pagetable);
#ifdef CACHE
g_free(s->pageentry_u8);
#endif
return ret;
}
static int dmg_open(BlockDriverState *bs, QDict *options, int flags,
Error **errp)
{
BDRVDMGState *s = bs->opaque;
uint64_t info_begin, info_end, last_in_offset, last_out_offset;
uint32_t count, tmp;
uint32_t max_compressed_size = 1, max_sectors_per_chunk = 1, i;
int64_t offset;
int ret;
bs->read_only = 1;
s->n_chunks = 0;
s->offsets = s->lengths = s->sectors = s->sectorcounts = NULL;
/* read offset of info blocks */
offset = bdrv_getlength(bs->file);
if (offset < 0) {
ret = offset;
goto fail;
}
offset -= 0x1d8;
ret = read_uint64(bs, offset, &info_begin);
if (ret < 0) {
goto fail;
} else if (info_begin == 0) {
ret = -EINVAL;
goto fail;
}
ret = read_uint32(bs, info_begin, &tmp);
if (ret < 0) {
goto fail;
} else if (tmp != 0x100) {
ret = -EINVAL;
goto fail;
}
ret = read_uint32(bs, info_begin + 4, &count);
if (ret < 0) {
goto fail;
} else if (count == 0) {
ret = -EINVAL;
goto fail;
}
info_end = info_begin + count;
offset = info_begin + 0x100;
/* read offsets */
last_in_offset = last_out_offset = 0;
while (offset < info_end) {
uint32_t type;
ret = read_uint32(bs, offset, &count);
if (ret < 0) {
goto fail;
} else if (count == 0) {
ret = -EINVAL;
goto fail;
}
offset += 4;
ret = read_uint32(bs, offset, &type);
if (ret < 0) {
goto fail;
}
if (type == 0x6d697368 && count >= 244) {
size_t new_size;
uint32_t chunk_count;
offset += 4;
offset += 200;
chunk_count = (count - 204) / 40;
new_size = sizeof(uint64_t) * (s->n_chunks + chunk_count);
s->types = g_realloc(s->types, new_size / 2);
s->offsets = g_realloc(s->offsets, new_size);
s->lengths = g_realloc(s->lengths, new_size);
s->sectors = g_realloc(s->sectors, new_size);
s->sectorcounts = g_realloc(s->sectorcounts, new_size);
for (i = s->n_chunks; i < s->n_chunks + chunk_count; i++) {
ret = read_uint32(bs, offset, &s->types[i]);
if (ret < 0) {
goto fail;
}
offset += 4;
if (s->types[i] != 0x80000005 && s->types[i] != 1 &&
s->types[i] != 2) {
if (s->types[i] == 0xffffffff && i > 0) {
last_in_offset = s->offsets[i - 1] + s->lengths[i - 1];
last_out_offset = s->sectors[i - 1] +
s->sectorcounts[i - 1];
}
chunk_count--;
i--;
offset += 36;
continue;
}
offset += 4;
ret = read_uint64(bs, offset, &s->sectors[i]);
if (ret < 0) {
goto fail;
}
s->sectors[i] += last_out_offset;
offset += 8;
ret = read_uint64(bs, offset, &s->sectorcounts[i]);
if (ret < 0) {
goto fail;
}
offset += 8;
if (s->sectorcounts[i] > DMG_SECTORCOUNTS_MAX) {
error_report("sector count %" PRIu64 " for chunk %" PRIu32
" is larger than max (%u)",
s->sectorcounts[i], i, DMG_SECTORCOUNTS_MAX);
ret = -EINVAL;
goto fail;
}
ret = read_uint64(bs, offset, &s->offsets[i]);
if (ret < 0) {
goto fail;
}
s->offsets[i] += last_in_offset;
offset += 8;
ret = read_uint64(bs, offset, &s->lengths[i]);
if (ret < 0) {
goto fail;
}
offset += 8;
if (s->lengths[i] > DMG_LENGTHS_MAX) {
error_report("length %" PRIu64 " for chunk %" PRIu32
" is larger than max (%u)",
s->lengths[i], i, DMG_LENGTHS_MAX);
ret = -EINVAL;
goto fail;
}
update_max_chunk_size(s, i, &max_compressed_size,
&max_sectors_per_chunk);
}
s->n_chunks += chunk_count;
}
}
/* initialize zlib engine */
s->compressed_chunk = qemu_try_blockalign(bs->file,
max_compressed_size + 1);
s->uncompressed_chunk = qemu_try_blockalign(bs->file,
512 * max_sectors_per_chunk);
if (s->compressed_chunk == NULL || s->uncompressed_chunk == NULL) {
ret = -ENOMEM;
goto fail;
}
if (inflateInit(&s->zstream) != Z_OK) {
ret = -EINVAL;
goto fail;
}
s->current_chunk = s->n_chunks;
qemu_co_mutex_init(&s->lock);
return 0;
fail:
g_free(s->types);
g_free(s->offsets);
g_free(s->lengths);
g_free(s->sectors);
g_free(s->sectorcounts);
qemu_vfree(s->compressed_chunk);
qemu_vfree(s->uncompressed_chunk);
return ret;
}
/* Reads the log header, and subsequent descriptors (if any). This
* will allocate all the space for buffer, which must be NULL when
* passed into this function. Each descriptor will also be validated,
* and error returned if any are invalid. */
static int vhdx_log_read_desc(BlockDriverState *bs, BDRVVHDXState *s,
VHDXLogEntries *log, VHDXLogDescEntries **buffer,
bool convert_endian)
{
int ret = 0;
uint32_t desc_sectors;
uint32_t sectors_read;
VHDXLogEntryHeader hdr;
VHDXLogDescEntries *desc_entries = NULL;
VHDXLogDescriptor desc;
int i;
assert(*buffer == NULL);
ret = vhdx_log_peek_hdr(bs, log, &hdr);
if (ret < 0) {
goto exit;
}
if (vhdx_log_hdr_is_valid(log, &hdr, s) == false) {
ret = -EINVAL;
goto exit;
}
desc_sectors = vhdx_compute_desc_sectors(hdr.descriptor_count);
desc_entries = qemu_try_blockalign(bs->file->bs,
desc_sectors * VHDX_LOG_SECTOR_SIZE);
if (desc_entries == NULL) {
ret = -ENOMEM;
goto exit;
}
ret = vhdx_log_read_sectors(bs, log, §ors_read, desc_entries,
desc_sectors, false);
if (ret < 0) {
goto free_and_exit;
}
if (sectors_read != desc_sectors) {
ret = -EINVAL;
goto free_and_exit;
}
/* put in proper endianness, and validate each desc */
for (i = 0; i < hdr.descriptor_count; i++) {
desc = desc_entries->desc[i];
vhdx_log_desc_le_import(&desc);
if (convert_endian) {
desc_entries->desc[i] = desc;
}
if (vhdx_log_desc_is_valid(&desc, &hdr) == false) {
ret = -EINVAL;
goto free_and_exit;
}
}
if (convert_endian) {
desc_entries->hdr = hdr;
}
*buffer = desc_entries;
goto exit;
free_and_exit:
qemu_vfree(desc_entries);
exit:
return ret;
}
