4f1c730736
## Description This is an embarrassing bug... In #619, a change was made to avoid a race condition and wait for the video output surface to be ready by re-enqueuing commands. Except I missed cancelling executing the current command. This meant the command would both be executed and queued again. So, if attaching the video surface takes a while, such as needing to wait for a refresh rate change, _a lot_ of commands could have been added to the queue and trigger a constant start-stop playback loop. I think technically it would eventually resolve itself, but can take a while. ### Related issues Should fix #626 |
||
|---|---|---|
| .. | ||
| wholphin | ||