throttle: Check for pending requests in throttle_group_unregister_bs()

throttle_group_unregister_bs() removes a BlockDriverState from its
throttling group and destroys the timers. This means that there must
be no pending throttled requests at that point (because it would be
impossible to complete them), so the caller has to drain them first.

At the moment throttle_group_unregister_bs() is only called from
bdrv_io_limits_disable(), which already takes care of draining the
requests, so there's nothing to worry about, but this patch makes
this invariant explicit in the documentation and adds the relevant
assertions.

Signed-off-by: Alberto Garcia <berto@igalia.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
This commit is contained in:
Alberto Garcia 2015-11-04 15:15:35 +02:00 committed by Kevin Wolf
parent 62547b8a1c
commit 5ac724184c

View file

@ -437,6 +437,9 @@ void throttle_group_register_bs(BlockDriverState *bs, const char *groupname)
* list, destroying the timers and setting the throttle_state pointer
* to NULL.
*
* The BlockDriverState must not have pending throttled requests, so
* the caller has to drain them first.
*
* The group will be destroyed if it's empty after this operation.
*
* @bs: the BlockDriverState to remove
@ -446,6 +449,10 @@ void throttle_group_unregister_bs(BlockDriverState *bs)
ThrottleGroup *tg = container_of(bs->throttle_state, ThrottleGroup, ts);
int i;
assert(bs->pending_reqs[0] == 0 && bs->pending_reqs[1] == 0);
assert(qemu_co_queue_empty(&bs->throttled_reqs[0]));
assert(qemu_co_queue_empty(&bs->throttled_reqs[1]));
qemu_mutex_lock(&tg->lock);
for (i = 0; i < 2; i++) {
if (tg->tokens[i] == bs) {