Audio thread safety, improve AudioCVT #312
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
The size of SDL_AudioCVT is known at compile-time, so no heap allocation is necessary. convert() can only fail if buf is NULL, which, in this case, it never can be. Additionally, the function can still succeed if no conversion is needed; just do nothing to the buffer! AudioCVT.len unfortunately uses an int and not a size_t, so an overflow check is added to make sure the buffer length is within c_int.
These methods can help the user improve performance. * is_conversion_needed() can be used to decide if convert() needs to be called. If not, the user can decide to omit the call. * get_capacity() can be used to avoid vector reallocations that are possible in the convert() method.
Because the audio callback data is used by an audio thread, we should declare that said data can be transferred to other threads. The trait does not implement Sync; the data is not accessed concurrently by many threads.
std::sync::MutexGuard does the same thing. Classically, mutexes cannot be unlocked from another thread. `!Send` is a negative trait, and thus requires us to ungate the `optin_builtin_traits` feature.
|
Great stuff! Cheers |
AngryLawyer
added a commit
that referenced
this pull request
Feb 2, 2015
Audio thread safety, improve AudioCVT
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
The
AudioCallbacktrait must implementSend.Senddeclares that the audio callback data is safe to be sent across other threads (i.e. the audio callback thread).As a result, types that can only be used on a single thread, such as
Rc<T>, raw pointers and even our ownRenderer, cannot be sent.As long as the struct implementing
AudioCallbackdoes not include any non-Sendable fields, user code should compile without explicitly having to declareSend. Notice the audio-whitenoise example didn't change at all.The only drawback is that
Sendimposes a'staticlifetime. That means no (non-static) references are allowed to be stored in the audio callback. However, the restriction should (hopefully) be lifted soon with rust-lang/rfcs#458.Changes to
AudioCVTare fairly simple. Because I'm getting lazy and don't want to write more things, you can just read the commit messages if you want more details. :D