static int coroutine_fn bdrv_co_do_write_zeroes(BlockDriverState *bs,
int64_t sector_num, int nb_sectors);
-static bool bdrv_exceed_bps_limits(BlockDriverState *bs, int nb_sectors,
- bool is_write, double elapsed_time, uint64_t *wait);
-static bool bdrv_exceed_iops_limits(BlockDriverState *bs, bool is_write,
- double elapsed_time, uint64_t *wait);
-static bool bdrv_exceed_io_limits(BlockDriverState *bs, int nb_sectors,
- bool is_write, int64_t *wait);
-
static QTAILQ_HEAD(, BlockDriverState) bdrv_states =
QTAILQ_HEAD_INITIALIZER(bdrv_states);
#endif
/* throttling disk I/O limits */
-void bdrv_io_limits_disable(BlockDriverState *bs)
+void bdrv_set_io_limits(BlockDriverState *bs,
+ ThrottleConfig *cfg)
{
- bs->io_limits_enabled = false;
+ int i;
+
+ throttle_config(&bs->throttle_state, cfg);
+
+ for (i = 0; i < 2; i++) {
+ qemu_co_enter_next(&bs->throttled_reqs[i]);
+ }
+}
- do {} while (qemu_co_enter_next(&bs->throttled_reqs));
+/* this function drain all the throttled IOs */
+static bool bdrv_start_throttled_reqs(BlockDriverState *bs)
+{
+ bool drained = false;
+ bool enabled = bs->io_limits_enabled;
+ int i;
- if (bs->block_timer) {
- qemu_del_timer(bs->block_timer);
- qemu_free_timer(bs->block_timer);
- bs->block_timer = NULL;
+ bs->io_limits_enabled = false;
+
+ for (i = 0; i < 2; i++) {
+ while (qemu_co_enter_next(&bs->throttled_reqs[i])) {
+ drained = true;
+ }
}
- bs->slice_start = 0;
- bs->slice_end = 0;
+ bs->io_limits_enabled = enabled;
+
+ return drained;
}
-static void bdrv_block_timer(void *opaque)
+void bdrv_io_limits_disable(BlockDriverState *bs)
{
- BlockDriverState *bs = opaque;
+ bs->io_limits_enabled = false;
+
+ bdrv_start_throttled_reqs(bs);
- qemu_co_enter_next(&bs->throttled_reqs);
+ throttle_destroy(&bs->throttle_state);
}
-void bdrv_io_limits_enable(BlockDriverState *bs)
+static void bdrv_throttle_read_timer_cb(void *opaque)
{
- qemu_co_queue_init(&bs->throttled_reqs);
- bs->block_timer = qemu_new_timer_ns(vm_clock, bdrv_block_timer, bs);
- bs->io_limits_enabled = true;
+ BlockDriverState *bs = opaque;
+ qemu_co_enter_next(&bs->throttled_reqs[0]);
}
-bool bdrv_io_limits_enabled(BlockDriverState *bs)
+static void bdrv_throttle_write_timer_cb(void *opaque)
+{
+ BlockDriverState *bs = opaque;
+ qemu_co_enter_next(&bs->throttled_reqs[1]);
+}
+
+/* should be called before bdrv_set_io_limits if a limit is set */
+void bdrv_io_limits_enable(BlockDriverState *bs)
{
- BlockIOLimit *io_limits = &bs->io_limits;
- return io_limits->bps[BLOCK_IO_LIMIT_READ]
- || io_limits->bps[BLOCK_IO_LIMIT_WRITE]
- || io_limits->bps[BLOCK_IO_LIMIT_TOTAL]
- || io_limits->iops[BLOCK_IO_LIMIT_READ]
- || io_limits->iops[BLOCK_IO_LIMIT_WRITE]
- || io_limits->iops[BLOCK_IO_LIMIT_TOTAL];
+ assert(!bs->io_limits_enabled);
+ throttle_init(&bs->throttle_state,
+ QEMU_CLOCK_VIRTUAL,
+ bdrv_throttle_read_timer_cb,
+ bdrv_throttle_write_timer_cb,
+ bs);
+ bs->io_limits_enabled = true;
}
+/* This function makes an IO wait if needed
+ *
+ * @nb_sectors: the number of sectors of the IO
+ * @is_write: is the IO a write
+ */
static void bdrv_io_limits_intercept(BlockDriverState *bs,
- bool is_write, int nb_sectors)
+ int nb_sectors,
+ bool is_write)
{
- int64_t wait_time = -1;
+ /* does this io must wait */
+ bool must_wait = throttle_schedule_timer(&bs->throttle_state, is_write);
- if (!qemu_co_queue_empty(&bs->throttled_reqs)) {
- qemu_co_queue_wait(&bs->throttled_reqs);
+ /* if must wait or any request of this type throttled queue the IO */
+ if (must_wait ||
+ !qemu_co_queue_empty(&bs->throttled_reqs[is_write])) {
+ qemu_co_queue_wait(&bs->throttled_reqs[is_write]);
}
- /* In fact, we hope to keep each request's timing, in FIFO mode. The next
- * throttled requests will not be dequeued until the current request is
- * allowed to be serviced. So if the current request still exceeds the
- * limits, it will be inserted to the head. All requests followed it will
- * be still in throttled_reqs queue.
- */
+ /* the IO will be executed, do the accounting */
+ throttle_account(&bs->throttle_state,
+ is_write,
+ nb_sectors * BDRV_SECTOR_SIZE);
- while (bdrv_exceed_io_limits(bs, nb_sectors, is_write, &wait_time)) {
- qemu_mod_timer(bs->block_timer,
- wait_time + qemu_get_clock_ns(vm_clock));
- qemu_co_queue_wait_insert_head(&bs->throttled_reqs);
+ /* if the next request must wait -> do nothing */
+ if (throttle_schedule_timer(&bs->throttle_state, is_write)) {
+ return;
}
- qemu_co_queue_next(&bs->throttled_reqs);
+ /* else queue next request for execution */
+ qemu_co_queue_next(&bs->throttled_reqs[is_write]);
}
/* check if the path starts with "<protocol>:" */
bdrv_iostatus_disable(bs);
notifier_list_init(&bs->close_notifiers);
notifier_with_return_list_init(&bs->before_write_notifiers);
+ qemu_co_queue_init(&bs->throttled_reqs[0]);
+ qemu_co_queue_init(&bs->throttled_reqs[1]);
+ bs->refcnt = 1;
return bs;
}
bs->open_flags = flags;
bs->buffer_alignment = 512;
+ bs->zero_beyond_eof = true;
open_flags = bdrv_open_flags(bs, flags);
bs->read_only = !(open_flags & BDRV_O_RDWR);
ret = -EINVAL;
goto free_and_fail;
}
- assert(file != NULL);
bs->file = file;
ret = drv->bdrv_open(bs, options, open_flags);
}
if (!bs->drv) {
QDECREF(bs->options);
}
- bdrv_delete(bs);
+ bdrv_unref(bs);
return ret;
}
*backing_filename ? backing_filename : NULL, options,
back_flags, back_drv);
if (ret < 0) {
- bdrv_delete(bs->backing_hd);
+ bdrv_unref(bs->backing_hd);
bs->backing_hd = NULL;
bs->open_flags |= BDRV_O_NO_BACKING;
return ret;
bs1 = bdrv_new("");
ret = bdrv_open(bs1, filename, NULL, 0, drv);
if (ret < 0) {
- bdrv_delete(bs1);
+ bdrv_unref(bs1);
goto fail;
}
total_size = bdrv_getlength(bs1) & BDRV_SECTOR_MASK;
- bdrv_delete(bs1);
+ bdrv_unref(bs1);
ret = get_tmp_filename(tmp_filename, sizeof(tmp_filename));
if (ret < 0) {
}
if (bs->file != file) {
- bdrv_delete(file);
+ bdrv_unref(file);
file = NULL;
}
bdrv_dev_change_media_cb(bs, true);
}
- /* throttling disk I/O limits */
- if (bs->io_limits_enabled) {
- bdrv_io_limits_enable(bs);
- }
-
return 0;
unlink_and_fail:
if (file != NULL) {
- bdrv_delete(file);
+ bdrv_unref(file);
}
if (bs->is_temporary) {
unlink(filename);
if (bs->drv) {
if (bs->backing_hd) {
- bdrv_delete(bs->backing_hd);
+ bdrv_unref(bs->backing_hd);
bs->backing_hd = NULL;
}
bs->drv->bdrv_close(bs);
bs->valid_key = 0;
bs->sg = 0;
bs->growable = 0;
+ bs->zero_beyond_eof = false;
QDECREF(bs->options);
bs->options = NULL;
if (bs->file != NULL) {
- bdrv_delete(bs->file);
+ bdrv_unref(bs->file);
bs->file = NULL;
}
}
}
}
+/* Check if any requests are in-flight (including throttled requests) */
+static bool bdrv_requests_pending(BlockDriverState *bs)
+{
+ if (!QLIST_EMPTY(&bs->tracked_requests)) {
+ return true;
+ }
+ if (!qemu_co_queue_empty(&bs->throttled_reqs[0])) {
+ return true;
+ }
+ if (!qemu_co_queue_empty(&bs->throttled_reqs[1])) {
+ return true;
+ }
+ if (bs->file && bdrv_requests_pending(bs->file)) {
+ return true;
+ }
+ if (bs->backing_hd && bdrv_requests_pending(bs->backing_hd)) {
+ return true;
+ }
+ return false;
+}
+
+static bool bdrv_requests_pending_all(void)
+{
+ BlockDriverState *bs;
+ QTAILQ_FOREACH(bs, &bdrv_states, list) {
+ if (bdrv_requests_pending(bs)) {
+ return true;
+ }
+ }
+ return false;
+}
+
/*
* Wait for pending requests to complete across all BlockDriverStates
*
*/
void bdrv_drain_all(void)
{
+ /* Always run first iteration so any pending completion BHs run */
+ bool busy = true;
BlockDriverState *bs;
- bool busy;
-
- do {
- busy = qemu_aio_wait();
+ while (busy) {
/* FIXME: We do not have timer support here, so this is effectively
* a busy wait.
*/
QTAILQ_FOREACH(bs, &bdrv_states, list) {
- while (qemu_co_enter_next(&bs->throttled_reqs)) {
+ if (bdrv_start_throttled_reqs(bs)) {
busy = true;
}
}
- } while (busy);
- /* If requests are still pending there is a bug somewhere */
- QTAILQ_FOREACH(bs, &bdrv_states, list) {
- assert(QLIST_EMPTY(&bs->tracked_requests));
- assert(qemu_co_queue_empty(&bs->throttled_reqs));
+ busy = bdrv_requests_pending_all();
+ busy |= aio_poll(qemu_get_aio_context(), busy);
}
}
bs_dest->enable_write_cache = bs_src->enable_write_cache;
- /* i/o timing parameters */
- bs_dest->slice_start = bs_src->slice_start;
- bs_dest->slice_end = bs_src->slice_end;
- bs_dest->slice_submitted = bs_src->slice_submitted;
- bs_dest->io_limits = bs_src->io_limits;
- bs_dest->throttled_reqs = bs_src->throttled_reqs;
- bs_dest->block_timer = bs_src->block_timer;
+ /* i/o throttled req */
+ memcpy(&bs_dest->throttle_state,
+ &bs_src->throttle_state,
+ sizeof(ThrottleState));
+ bs_dest->throttled_reqs[0] = bs_src->throttled_reqs[0];
+ bs_dest->throttled_reqs[1] = bs_src->throttled_reqs[1];
bs_dest->io_limits_enabled = bs_src->io_limits_enabled;
/* r/w error */
/* dirty bitmap */
bs_dest->dirty_bitmap = bs_src->dirty_bitmap;
+ /* reference count */
+ bs_dest->refcnt = bs_src->refcnt;
+
/* job */
bs_dest->in_use = bs_src->in_use;
bs_dest->job = bs_src->job;
assert(bs_new->dev == NULL);
assert(bs_new->in_use == 0);
assert(bs_new->io_limits_enabled == false);
- assert(bs_new->block_timer == NULL);
+ assert(!throttle_have_timer(&bs_new->throttle_state));
tmp = *bs_new;
*bs_new = *bs_old;
assert(bs_new->job == NULL);
assert(bs_new->in_use == 0);
assert(bs_new->io_limits_enabled == false);
- assert(bs_new->block_timer == NULL);
+ assert(!throttle_have_timer(&bs_new->throttle_state));
bdrv_rebind(bs_new);
bdrv_rebind(bs_old);
bs_new->drv ? bs_new->drv->format_name : "");
}
-void bdrv_delete(BlockDriverState *bs)
+static void bdrv_delete(BlockDriverState *bs)
{
assert(!bs->dev);
assert(!bs->job);
assert(!bs->in_use);
+ assert(!bs->refcnt);
bdrv_close(bs);
buf = g_malloc(COMMIT_BUF_SECTORS * BDRV_SECTOR_SIZE);
for (sector = 0; sector < total_sectors; sector += n) {
- if (bdrv_is_allocated(bs, sector, COMMIT_BUF_SECTORS, &n)) {
-
+ ret = bdrv_is_allocated(bs, sector, COMMIT_BUF_SECTORS, &n);
+ if (ret < 0) {
+ goto ro_cleanup;
+ }
+ if (ret) {
if (bdrv_read(bs, sector, buf, n) != 0) {
ret = -EIO;
goto ro_cleanup;
QSIMPLEQ_FOREACH_SAFE(intermediate_state, &states_to_delete, entry, next) {
/* so that bdrv_close() does not recursively close the chain */
intermediate_state->bs->backing_hd = NULL;
- bdrv_delete(intermediate_state->bs);
+ bdrv_unref(intermediate_state->bs);
}
ret = 0;
return -EIO;
}
- /* throttling disk read I/O */
- if (bs->io_limits_enabled) {
- bdrv_io_limits_intercept(bs, false, nb_sectors);
- }
-
if (bs->copy_on_read) {
flags |= BDRV_REQ_COPY_ON_READ;
}
wait_for_overlapping_requests(bs, sector_num, nb_sectors);
}
+ /* throttling disk I/O */
+ if (bs->io_limits_enabled) {
+ bdrv_io_limits_intercept(bs, nb_sectors, false);
+ }
+
tracked_request_begin(&req, bs, sector_num, nb_sectors, false);
if (flags & BDRV_REQ_COPY_ON_READ) {
int pnum;
- ret = bdrv_co_is_allocated(bs, sector_num, nb_sectors, &pnum);
+ ret = bdrv_is_allocated(bs, sector_num, nb_sectors, &pnum);
if (ret < 0) {
goto out;
}
}
}
- ret = drv->bdrv_co_readv(bs, sector_num, nb_sectors, qiov);
+ if (!(bs->zero_beyond_eof && bs->growable)) {
+ ret = drv->bdrv_co_readv(bs, sector_num, nb_sectors, qiov);
+ } else {
+ /* Read zeros after EOF of growable BDSes */
+ int64_t len, total_sectors, max_nb_sectors;
+
+ len = bdrv_getlength(bs);
+ if (len < 0) {
+ ret = len;
+ goto out;
+ }
+
+ total_sectors = len >> BDRV_SECTOR_BITS;
+ max_nb_sectors = MAX(0, total_sectors - sector_num);
+ if (max_nb_sectors > 0) {
+ ret = drv->bdrv_co_readv(bs, sector_num,
+ MIN(nb_sectors, max_nb_sectors), qiov);
+ } else {
+ ret = 0;
+ }
+
+ /* Reading beyond end of file is supposed to produce zeroes */
+ if (ret == 0 && total_sectors < sector_num + nb_sectors) {
+ uint64_t offset = MAX(0, total_sectors - sector_num);
+ uint64_t bytes = (sector_num + nb_sectors - offset) *
+ BDRV_SECTOR_SIZE;
+ qemu_iovec_memset(qiov, offset * BDRV_SECTOR_SIZE, 0, bytes);
+ }
+ }
out:
tracked_request_end(&req);
return -EIO;
}
- /* throttling disk write I/O */
- if (bs->io_limits_enabled) {
- bdrv_io_limits_intercept(bs, true, nb_sectors);
- }
-
if (bs->copy_on_read_in_flight) {
wait_for_overlapping_requests(bs, sector_num, nb_sectors);
}
+ /* throttling disk I/O */
+ if (bs->io_limits_enabled) {
+ bdrv_io_limits_intercept(bs, nb_sectors, true);
+ }
+
tracked_request_begin(&req, bs, sector_num, nb_sectors, true);
ret = notifier_with_return_list_notify(&bs->before_write_notifiers, &req);
if (bs->wr_highest_sector < sector_num + nb_sectors - 1) {
bs->wr_highest_sector = sector_num + nb_sectors - 1;
}
+ if (bs->growable && ret >= 0) {
+ bs->total_sectors = MAX(bs->total_sectors, sector_num + nb_sectors);
+ }
tracked_request_end(&req);
if (!drv)
return -ENOMEDIUM;
- if (bs->growable || bdrv_dev_has_removable_media(bs)) {
+ if (bdrv_dev_has_removable_media(bs)) {
if (drv->bdrv_getlength) {
return drv->bdrv_getlength(bs);
}
*nb_sectors_ptr = length;
}
-/* throttling disk io limits */
-void bdrv_set_io_limits(BlockDriverState *bs,
- BlockIOLimit *io_limits)
-{
- bs->io_limits = *io_limits;
- bs->io_limits_enabled = bdrv_io_limits_enabled(bs);
-}
-
void bdrv_set_on_error(BlockDriverState *bs, BlockdevOnError on_read_error,
BlockdevOnError on_write_error)
{
{
assert(bs->drv);
+ /* If BS is a copy on write image, it is initialized to
+ the contents of the base image, which may not be zeroes. */
+ if (bs->backing_hd) {
+ return 0;
+ }
if (bs->drv->bdrv_has_zero_init) {
return bs->drv->bdrv_has_zero_init(bs);
}
return 0;
}
-typedef struct BdrvCoIsAllocatedData {
+typedef struct BdrvCoGetBlockStatusData {
BlockDriverState *bs;
BlockDriverState *base;
int64_t sector_num;
int nb_sectors;
int *pnum;
- int ret;
+ int64_t ret;
bool done;
-} BdrvCoIsAllocatedData;
+} BdrvCoGetBlockStatusData;
/*
* Returns true iff the specified sector is present in the disk image. Drivers
* 'nb_sectors' is the max value 'pnum' should be set to. If nb_sectors goes
* beyond the end of the disk image it will be clamped.
*/
-int coroutine_fn bdrv_co_is_allocated(BlockDriverState *bs, int64_t sector_num,
- int nb_sectors, int *pnum)
+static int64_t coroutine_fn bdrv_co_get_block_status(BlockDriverState *bs,
+ int64_t sector_num,
+ int nb_sectors, int *pnum)
{
+ int64_t length;
int64_t n;
+ int64_t ret, ret2;
+
+ length = bdrv_getlength(bs);
+ if (length < 0) {
+ return length;
+ }
- if (sector_num >= bs->total_sectors) {
+ if (sector_num >= (length >> BDRV_SECTOR_BITS)) {
*pnum = 0;
return 0;
}
nb_sectors = n;
}
- if (!bs->drv->bdrv_co_is_allocated) {
+ if (!bs->drv->bdrv_co_get_block_status) {
*pnum = nb_sectors;
- return 1;
+ ret = BDRV_BLOCK_DATA;
+ if (bs->drv->protocol_name) {
+ ret |= BDRV_BLOCK_OFFSET_VALID | (sector_num * BDRV_SECTOR_SIZE);
+ }
+ return ret;
+ }
+
+ ret = bs->drv->bdrv_co_get_block_status(bs, sector_num, nb_sectors, pnum);
+ if (ret < 0) {
+ return ret;
+ }
+
+ if (!(ret & BDRV_BLOCK_DATA)) {
+ if (bdrv_has_zero_init(bs)) {
+ ret |= BDRV_BLOCK_ZERO;
+ } else {
+ BlockDriverState *bs2 = bs->backing_hd;
+ int64_t length2 = bdrv_getlength(bs2);
+ if (length2 >= 0 && sector_num >= (length2 >> BDRV_SECTOR_BITS)) {
+ ret |= BDRV_BLOCK_ZERO;
+ }
+ }
}
- return bs->drv->bdrv_co_is_allocated(bs, sector_num, nb_sectors, pnum);
+ if (bs->file &&
+ (ret & BDRV_BLOCK_DATA) && !(ret & BDRV_BLOCK_ZERO) &&
+ (ret & BDRV_BLOCK_OFFSET_VALID)) {
+ ret2 = bdrv_co_get_block_status(bs->file, ret >> BDRV_SECTOR_BITS,
+ *pnum, pnum);
+ if (ret2 >= 0) {
+ /* Ignore errors. This is just providing extra information, it
+ * is useful but not necessary.
+ */
+ ret |= (ret2 & BDRV_BLOCK_ZERO);
+ }
+ }
+
+ return ret;
}
-/* Coroutine wrapper for bdrv_is_allocated() */
-static void coroutine_fn bdrv_is_allocated_co_entry(void *opaque)
+/* Coroutine wrapper for bdrv_get_block_status() */
+static void coroutine_fn bdrv_get_block_status_co_entry(void *opaque)
{
- BdrvCoIsAllocatedData *data = opaque;
+ BdrvCoGetBlockStatusData *data = opaque;
BlockDriverState *bs = data->bs;
- data->ret = bdrv_co_is_allocated(bs, data->sector_num, data->nb_sectors,
- data->pnum);
+ data->ret = bdrv_co_get_block_status(bs, data->sector_num, data->nb_sectors,
+ data->pnum);
data->done = true;
}
/*
- * Synchronous wrapper around bdrv_co_is_allocated().
+ * Synchronous wrapper around bdrv_co_get_block_status().
*
- * See bdrv_co_is_allocated() for details.
+ * See bdrv_co_get_block_status() for details.
*/
-int bdrv_is_allocated(BlockDriverState *bs, int64_t sector_num, int nb_sectors,
- int *pnum)
+int64_t bdrv_get_block_status(BlockDriverState *bs, int64_t sector_num,
+ int nb_sectors, int *pnum)
{
Coroutine *co;
- BdrvCoIsAllocatedData data = {
+ BdrvCoGetBlockStatusData data = {
.bs = bs,
.sector_num = sector_num,
.nb_sectors = nb_sectors,
.done = false,
};
- co = qemu_coroutine_create(bdrv_is_allocated_co_entry);
- qemu_coroutine_enter(co, &data);
- while (!data.done) {
- qemu_aio_wait();
+ if (qemu_in_coroutine()) {
+ /* Fast-path if already in coroutine context */
+ bdrv_get_block_status_co_entry(&data);
+ } else {
+ co = qemu_coroutine_create(bdrv_get_block_status_co_entry);
+ qemu_coroutine_enter(co, &data);
+ while (!data.done) {
+ qemu_aio_wait();
+ }
}
return data.ret;
}
+int coroutine_fn bdrv_is_allocated(BlockDriverState *bs, int64_t sector_num,
+ int nb_sectors, int *pnum)
+{
+ int64_t ret = bdrv_get_block_status(bs, sector_num, nb_sectors, pnum);
+ if (ret < 0) {
+ return ret;
+ }
+ return
+ (ret & BDRV_BLOCK_DATA) ||
+ ((ret & BDRV_BLOCK_ZERO) && !bdrv_has_zero_init(bs));
+}
+
/*
* Given an image chain: ... -> [BASE] -> [INTER1] -> [INTER2] -> [TOP]
*
* allocated/unallocated state.
*
*/
-int coroutine_fn bdrv_co_is_allocated_above(BlockDriverState *top,
- BlockDriverState *base,
- int64_t sector_num,
- int nb_sectors, int *pnum)
+int bdrv_is_allocated_above(BlockDriverState *top,
+ BlockDriverState *base,
+ int64_t sector_num,
+ int nb_sectors, int *pnum)
{
BlockDriverState *intermediate;
int ret, n = nb_sectors;
intermediate = top;
while (intermediate && intermediate != base) {
int pnum_inter;
- ret = bdrv_co_is_allocated(intermediate, sector_num, nb_sectors,
- &pnum_inter);
+ ret = bdrv_is_allocated(intermediate, sector_num, nb_sectors,
+ &pnum_inter);
if (ret < 0) {
return ret;
} else if (ret) {
return 0;
}
-/* Coroutine wrapper for bdrv_is_allocated_above() */
-static void coroutine_fn bdrv_is_allocated_above_co_entry(void *opaque)
-{
- BdrvCoIsAllocatedData *data = opaque;
- BlockDriverState *top = data->bs;
- BlockDriverState *base = data->base;
-
- data->ret = bdrv_co_is_allocated_above(top, base, data->sector_num,
- data->nb_sectors, data->pnum);
- data->done = true;
-}
-
-/*
- * Synchronous wrapper around bdrv_co_is_allocated_above().
- *
- * See bdrv_co_is_allocated_above() for details.
- */
-int bdrv_is_allocated_above(BlockDriverState *top, BlockDriverState *base,
- int64_t sector_num, int nb_sectors, int *pnum)
-{
- Coroutine *co;
- BdrvCoIsAllocatedData data = {
- .bs = top,
- .base = base,
- .sector_num = sector_num,
- .nb_sectors = nb_sectors,
- .pnum = pnum,
- .done = false,
- };
-
- co = qemu_coroutine_create(bdrv_is_allocated_above_co_entry);
- qemu_coroutine_enter(co, &data);
- while (!data.done) {
- qemu_aio_wait();
- }
- return data.ret;
-}
-
const char *bdrv_get_encrypted_filename(BlockDriverState *bs)
{
if (bs->backing_hd && bs->backing_hd->encrypted)
acb->aiocb_info->cancel(acb);
}
-/* block I/O throttling */
-static bool bdrv_exceed_bps_limits(BlockDriverState *bs, int nb_sectors,
- bool is_write, double elapsed_time, uint64_t *wait)
-{
- uint64_t bps_limit = 0;
- uint64_t extension;
- double bytes_limit, bytes_base, bytes_res;
- double slice_time, wait_time;
-
- if (bs->io_limits.bps[BLOCK_IO_LIMIT_TOTAL]) {
- bps_limit = bs->io_limits.bps[BLOCK_IO_LIMIT_TOTAL];
- } else if (bs->io_limits.bps[is_write]) {
- bps_limit = bs->io_limits.bps[is_write];
- } else {
- if (wait) {
- *wait = 0;
- }
-
- return false;
- }
-
- slice_time = bs->slice_end - bs->slice_start;
- slice_time /= (NANOSECONDS_PER_SECOND);
- bytes_limit = bps_limit * slice_time;
- bytes_base = bs->slice_submitted.bytes[is_write];
- if (bs->io_limits.bps[BLOCK_IO_LIMIT_TOTAL]) {
- bytes_base += bs->slice_submitted.bytes[!is_write];
- }
-
- /* bytes_base: the bytes of data which have been read/written; and
- * it is obtained from the history statistic info.
- * bytes_res: the remaining bytes of data which need to be read/written.
- * (bytes_base + bytes_res) / bps_limit: used to calcuate
- * the total time for completing reading/writting all data.
- */
- bytes_res = (unsigned) nb_sectors * BDRV_SECTOR_SIZE;
-
- if (bytes_base + bytes_res <= bytes_limit) {
- if (wait) {
- *wait = 0;
- }
-
- return false;
- }
-
- /* Calc approx time to dispatch */
- wait_time = (bytes_base + bytes_res) / bps_limit - elapsed_time;
-
- /* When the I/O rate at runtime exceeds the limits,
- * bs->slice_end need to be extended in order that the current statistic
- * info can be kept until the timer fire, so it is increased and tuned
- * based on the result of experiment.
- */
- extension = wait_time * NANOSECONDS_PER_SECOND;
- extension = DIV_ROUND_UP(extension, BLOCK_IO_SLICE_TIME) *
- BLOCK_IO_SLICE_TIME;
- bs->slice_end += extension;
- if (wait) {
- *wait = wait_time * NANOSECONDS_PER_SECOND;
- }
-
- return true;
-}
-
-static bool bdrv_exceed_iops_limits(BlockDriverState *bs, bool is_write,
- double elapsed_time, uint64_t *wait)
-{
- uint64_t iops_limit = 0;
- double ios_limit, ios_base;
- double slice_time, wait_time;
-
- if (bs->io_limits.iops[BLOCK_IO_LIMIT_TOTAL]) {
- iops_limit = bs->io_limits.iops[BLOCK_IO_LIMIT_TOTAL];
- } else if (bs->io_limits.iops[is_write]) {
- iops_limit = bs->io_limits.iops[is_write];
- } else {
- if (wait) {
- *wait = 0;
- }
-
- return false;
- }
-
- slice_time = bs->slice_end - bs->slice_start;
- slice_time /= (NANOSECONDS_PER_SECOND);
- ios_limit = iops_limit * slice_time;
- ios_base = bs->slice_submitted.ios[is_write];
- if (bs->io_limits.iops[BLOCK_IO_LIMIT_TOTAL]) {
- ios_base += bs->slice_submitted.ios[!is_write];
- }
-
- if (ios_base + 1 <= ios_limit) {
- if (wait) {
- *wait = 0;
- }
-
- return false;
- }
-
- /* Calc approx time to dispatch, in seconds */
- wait_time = (ios_base + 1) / iops_limit;
- if (wait_time > elapsed_time) {
- wait_time = wait_time - elapsed_time;
- } else {
- wait_time = 0;
- }
-
- /* Exceeded current slice, extend it by another slice time */
- bs->slice_end += BLOCK_IO_SLICE_TIME;
- if (wait) {
- *wait = wait_time * NANOSECONDS_PER_SECOND;
- }
-
- return true;
-}
-
-static bool bdrv_exceed_io_limits(BlockDriverState *bs, int nb_sectors,
- bool is_write, int64_t *wait)
-{
- int64_t now, max_wait;
- uint64_t bps_wait = 0, iops_wait = 0;
- double elapsed_time;
- int bps_ret, iops_ret;
-
- now = qemu_get_clock_ns(vm_clock);
- if (now > bs->slice_end) {
- bs->slice_start = now;
- bs->slice_end = now + BLOCK_IO_SLICE_TIME;
- memset(&bs->slice_submitted, 0, sizeof(bs->slice_submitted));
- }
-
- elapsed_time = now - bs->slice_start;
- elapsed_time /= (NANOSECONDS_PER_SECOND);
-
- bps_ret = bdrv_exceed_bps_limits(bs, nb_sectors,
- is_write, elapsed_time, &bps_wait);
- iops_ret = bdrv_exceed_iops_limits(bs, is_write,
- elapsed_time, &iops_wait);
- if (bps_ret || iops_ret) {
- max_wait = bps_wait > iops_wait ? bps_wait : iops_wait;
- if (wait) {
- *wait = max_wait;
- }
-
- now = qemu_get_clock_ns(vm_clock);
- if (bs->slice_end < now + max_wait) {
- bs->slice_end = now + max_wait;
- }
-
- return true;
- }
-
- if (wait) {
- *wait = 0;
- }
-
- bs->slice_submitted.bytes[is_write] += (int64_t)nb_sectors *
- BDRV_SECTOR_SIZE;
- bs->slice_submitted.ios[is_write]++;
-
- return false;
-}
-
/**************************************************************/
/* async block device emulation */
}
}
+/* Get a reference to bs */
+void bdrv_ref(BlockDriverState *bs)
+{
+ bs->refcnt++;
+}
+
+/* Release a previously grabbed reference to bs.
+ * If after releasing, reference count is zero, the BlockDriverState is
+ * deleted. */
+void bdrv_unref(BlockDriverState *bs)
+{
+ assert(bs->refcnt > 0);
+ if (--bs->refcnt == 0) {
+ bdrv_delete(bs);
+ }
+}
+
void bdrv_set_in_use(BlockDriverState *bs, int in_use)
{
assert(bs->in_use != in_use);
free_option_parameters(param);
if (bs) {
- bdrv_delete(bs);
+ bdrv_unref(bs);
}
}