cloud-setup/trivial: fix gtkdoc comment for nmcs_utils_poll()
@probe_start_fcn is not called the first time synchronously. Fix the comment. While at it, reword a bit.
This commit is contained in:
@@ -304,14 +304,15 @@ _poll_cancelled_cb(GObject *object, gpointer user_data)
|
||||
* @poll_timeout_ms: if >= 0, then this is the overall timeout for how long we poll.
|
||||
* When this timeout expires, the request completes with failure (and error set).
|
||||
* @ratelimit_timeout_ms: if > 0, we ratelimit the starts from one prope_start_fcn
|
||||
* call to the next.
|
||||
* call to the next. We will wait at least this time between two consecutive polls.
|
||||
* @sleep_timeout_ms: if > 0, then we wait after a probe finished this timeout
|
||||
* before the next. Together with @ratelimit_timeout_ms this determines how
|
||||
* frequently we probe.
|
||||
* @probe_start_fcn: used to start a (asynchronous) probe. A probe must be completed
|
||||
* by calling the provided callback. While a probe is in progress, we will not
|
||||
* start another. This function is already invoked the first time synchronously,
|
||||
* during nmcs_utils_poll().
|
||||
* frequently we probe. We will wait at least this time between the end of the
|
||||
* previous poll and the next one.
|
||||
* @probe_start_fcn: used to start a (asynchronous) probe. A probe must be
|
||||
* completed by calling the provided callback. While a probe is in progress, we
|
||||
* will not start another. The function is called the first time on an idle
|
||||
* handler, afterwards it gets called again on each timeout for polling.
|
||||
* @probe_finish_fcn: will be called from the callback of @probe_start_fcn. If the
|
||||
* function returns %TRUE (polling done) or an error, polling stops. Otherwise,
|
||||
* another poll will be started.
|
||||
|
Reference in New Issue
Block a user