Use IPC::Semaphore instead of sending an IPC message for every captured audio sample
Created attachment 427873 [details] Patch
I reused the web audio model but this might not be an optimal approach given we use chunks of 128 samples, while the AudioUnit may use larger chunks. Another approach would be to keep using fixed sample size, say 5 or 10 ms, something like std::max(128, sampleRate / 100). Thoughts?
Created attachment 427879 [details] Patch
Created attachment 427880 [details] Patch
Created attachment 427887 [details] Patch
Created attachment 428000 [details] Patch
Created attachment 428004 [details] Patch
Committed r277256 (237525@main): <https://commits.webkit.org/237525@main> All reviewed patches have been landed. Closing bug and clearing flags on attachment 428004 [details].
<rdar://problem/77731343>