Re: [PATCH v5 0/12] MR1948: Draft: windows.media.speech: Recognition session work the second.

[Jinoh Kang (＠iamahuman)]

Jinoh Kang (@iamahuman) commented about dlls/windows.media.speech/recognizer.c:
> }
>
> static HRESULT WINAPI session_StartAsync( ISpeechContinuousRecognitionSession *iface, IAsyncAction **action )
>  {
> - FIXME("iface %p, action %p stub!\n", iface, action);
> - return async_action_create(NULL, start_callback, action);
> - struct session *impl = impl_from_ISpeechContinuousRecognitionSession(iface);
> - BOOLEAN invalid_state = FALSE;
> - HRESULT hr = S_OK;
> -
> - TRACE("iface %p, action %p.\n", iface, action);
> -
> - *action = NULL;
> -
> - EnterCriticalSection(&impl->cs);
> - if (!impl->worker_running && !impl->worker_thread && SUCCEEDED(hr = async_action_create(NULL, session_start_async, action)))

`async_action_create` should be taken out of the critical section. In general, minimize the number of potentially failing or blocking operations inside a critical section (ideally to zero).