/* commit COW file into the raw image */
int bdrv_commit(BlockDriverState *bs)
{
BlockBackend *src, *backing;
BlockDriverState *backing_file_bs = NULL;
BlockDriverState *commit_top_bs = NULL;
BlockDriver *drv = bs->drv;
int64_t offset, length, backing_length;
int ro;
int64_t n;
int ret = 0;
uint8_t *buf = NULL;
Error *local_err = NULL;
if (!drv)
return -ENOMEDIUM;
if (!bs->backing) {
return -ENOTSUP;
}
if (bdrv_op_is_blocked(bs, BLOCK_OP_TYPE_COMMIT_SOURCE, NULL) ||
bdrv_op_is_blocked(bs->backing->bs, BLOCK_OP_TYPE_COMMIT_TARGET, NULL)) {
return -EBUSY;
}
ro = bs->backing->bs->read_only;
if (ro) {
if (bdrv_reopen_set_read_only(bs->backing->bs, false, NULL)) {
return -EACCES;
}
}
src = blk_new(BLK_PERM_CONSISTENT_READ, BLK_PERM_ALL);
backing = blk_new(BLK_PERM_WRITE | BLK_PERM_RESIZE, BLK_PERM_ALL);
ret = blk_insert_bs(src, bs, &local_err);
if (ret < 0) {
error_report_err(local_err);
goto ro_cleanup;
}
/* Insert commit_top block node above backing, so we can write to it */
backing_file_bs = backing_bs(bs);
commit_top_bs = bdrv_new_open_driver(&bdrv_commit_top, NULL, BDRV_O_RDWR,
&local_err);
if (commit_top_bs == NULL) {
error_report_err(local_err);
goto ro_cleanup;
}
bdrv_set_aio_context(commit_top_bs, bdrv_get_aio_context(backing_file_bs));
bdrv_set_backing_hd(commit_top_bs, backing_file_bs, &error_abort);
bdrv_set_backing_hd(bs, commit_top_bs, &error_abort);
ret = blk_insert_bs(backing, backing_file_bs, &local_err);
if (ret < 0) {
error_report_err(local_err);
goto ro_cleanup;
}
length = blk_getlength(src);
if (length < 0) {
ret = length;
goto ro_cleanup;
}
backing_length = blk_getlength(backing);
if (backing_length < 0) {
ret = backing_length;
goto ro_cleanup;
}
/* If our top snapshot is larger than the backing file image,
* grow the backing file image if possible. If not possible,
* we must return an error */
if (length > backing_length) {
ret = blk_truncate(backing, length, PREALLOC_MODE_OFF, &local_err);
if (ret < 0) {
error_report_err(local_err);
goto ro_cleanup;
}
}
/* blk_try_blockalign() for src will choose an alignment that works for
* backing as well, so no need to compare the alignment manually. */
buf = blk_try_blockalign(src, COMMIT_BUF_SIZE);
if (buf == NULL) {
ret = -ENOMEM;
goto ro_cleanup;
}
for (offset = 0; offset < length; offset += n) {
ret = bdrv_is_allocated(bs, offset, COMMIT_BUF_SIZE, &n);
if (ret < 0) {
goto ro_cleanup;
}
if (ret) {
ret = blk_pread(src, offset, buf, n);
if (ret < 0) {
goto ro_cleanup;
}
ret = blk_pwrite(backing, offset, buf, n, 0);
if (ret < 0) {
goto ro_cleanup;
}
}
}
if (drv->bdrv_make_empty) {
ret = drv->bdrv_make_empty(bs);
if (ret < 0) {
goto ro_cleanup;
}
blk_flush(src);
}
/*
* Make sure all data we wrote to the backing device is actually
* stable on disk.
*/
blk_flush(backing);
ret = 0;
ro_cleanup:
qemu_vfree(buf);
blk_unref(backing);
if (backing_file_bs) {
bdrv_set_backing_hd(bs, backing_file_bs, &error_abort);
}
bdrv_unref(commit_top_bs);
blk_unref(src);
if (ro) {
/* ignoring error return here */
bdrv_reopen_set_read_only(bs->backing->bs, true, NULL);
}
return ret;
}
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 int coroutine_fn commit_run(Job *job, Error **errp)
{
CommitBlockJob *s = container_of(job, CommitBlockJob, common.job);
int64_t offset;
uint64_t delay_ns = 0;
int ret = 0;
int64_t n = 0; /* bytes */
void *buf = NULL;
int bytes_written = 0;
int64_t len, base_len;
ret = len = blk_getlength(s->top);
if (len < 0) {
goto out;
}
job_progress_set_remaining(&s->common.job, len);
ret = base_len = blk_getlength(s->base);
if (base_len < 0) {
goto out;
}
if (base_len < len) {
ret = blk_truncate(s->base, len, PREALLOC_MODE_OFF, NULL);
if (ret) {
goto out;
}
}
buf = blk_blockalign(s->top, COMMIT_BUFFER_SIZE);
for (offset = 0; offset < len; offset += n) {
bool copy;
/* Note that even when no rate limit is applied we need to yield
* with no pending I/O here so that bdrv_drain_all() returns.
*/
job_sleep_ns(&s->common.job, delay_ns);
if (job_is_cancelled(&s->common.job)) {
break;
}
/* Copy if allocated above the base */
ret = bdrv_is_allocated_above(blk_bs(s->top), blk_bs(s->base),
offset, COMMIT_BUFFER_SIZE, &n);
copy = (ret == 1);
trace_commit_one_iteration(s, offset, n, ret);
if (copy) {
ret = commit_populate(s->top, s->base, offset, n, buf);
bytes_written += n;
}
if (ret < 0) {
BlockErrorAction action =
block_job_error_action(&s->common, false, s->on_error, -ret);
if (action == BLOCK_ERROR_ACTION_REPORT) {
goto out;
} else {
n = 0;
continue;
}
}
/* Publish progress */
job_progress_update(&s->common.job, n);
if (copy) {
delay_ns = block_job_ratelimit_get_delay(&s->common, n);
} else {
delay_ns = 0;
}
}
ret = 0;
out:
qemu_vfree(buf);
return ret;
}
