Fix iOS AVAudioSession before engine init#22
Open
txbrown wants to merge 6 commits into
Open
Conversation
There was a problem hiding this comment.
Pull request overview
This PR updates the iOS native audio startup path so AVAudioSession is configured before AVAudioEngine is created, with the goal of preventing the startup-time buffer-size mismatch that currently produces corrupted audio on launch.
Changes:
- Configure
AVAudioSessionduringElementaryinitialization, including playback category, options, preferred I/O buffer duration, and activation. - Extend init logging to include the granted
IOBufferDuration. - Derive the Elementary runtime block size from the active session’s actual buffer duration instead of hard-coding
512.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
txbrown
force-pushed
the
fix/ios-audio-session-setup
branch
2 times, most recently
from
48913e8 to
417c7d3
Compare
…er mismatch iOS defaults to AVAudioSessionCategorySoloAmbient with a system-chosen buffer size (~4096 frames on device). Elementary's runtime is initialized with a fixed block size of 512, causing a mismatch that produces garbage audio (audible as a low-frequency square wave) immediately on app launch. Fix: configure AVAudioSession with Playback category before creating AVAudioEngine, then derive the actual block size from session.IOBufferDuration so the runtime always matches what the engine delivers per callback. - Initialize all NSError* to nil and reset between calls, since AVAudioSession APIs don't clear NSError** on success and stale pointers caused false error checks. - Extract audio session configuration into configureAudioSession which checks for an already-active session before overwriting settings, preventing breakage for apps needing Ambient/PlayAndRecord/mute-switch behavior. - Recreate runtime with current sample rate and block size after engine config changes (e.g. Bluetooth/AirPlay route switch) to avoid reintroducing buffer-size mismatch. - Extract computeBufferSizeFromSession helper to deduplicate block size calculation.
txbrown
force-pushed
the
fix/ios-audio-session-setup
branch
3 times, most recently
from
417c7d3 to
b8f8c12
Compare
Add configurable iOS audio session ownership so apps can choose playback settings or opt out when another audio stack owns AVAudioSession. Lazily initialize the native engine so examples can configure the session before first render.
- Fix stale NSError: use separate local errors and nil-assign before each call so AVAudioSession APIs cannot leave garbage in the pointer. - Early return in handleAudioInterruption when setActive fails so engine restart is not attempted on an inactive session. - Compute preferred IO buffer duration from actual session sample rate (after category is set) instead of hardcoding 48000 Hz. - Replace magic 0x4 with availability-checked constant for Bluetooth HFP. - Serialize audio session state changes on the main thread to prevent data races on audioSessionActive from notification handlers and JS thread methods. - JS: default configureAudioSession() argument to empty object so calling it with no args works.
AVAudioSessionCategoryOptionAllowBluetoothHFP is only available from Xcode 26/iOS SDK 26.2. The CI runs Xcode 16.4 where the constant doesn't exist. Fall back to the raw value with a comment explaining why.
This file contains hidden or 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
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.
iOS defaults to AVAudioSessionCategorySoloAmbient with a system-chosen buffer size (~4096 frames on device). Elementary's runtime is initialised with a fixed block size of 512, causing a mismatch that produces garbage audio (audible as a low-frequency square wave) immediately on app launch.
Fix: set AVAudioSessionCategoryPlayback with MixWithOthers and AllowBluetoothA2DP, request a 512-frame preferred buffer duration, activate the session before creating AVAudioEngine, then derive the actual block size from session.IOBufferDuration so the runtime always matches what the engine delivers per callback.
I didn't notice this issue on Midicircuit because it still has its own AVAudioEngine setup and some custom native code to bridge between it and react-native-elementary-audio.