Re: newbie question: activateAudioEngine failure consequences
Re: newbie question: activateAudioEngine failure consequences
- Subject: Re: newbie question: activateAudioEngine failure consequences
- From: Mark Cookson <email@hidden>
- Date: Wed, 23 Feb 2005 12:07:42 -0600
You've got a retain on an object, possibly totally unrelated to the engine,
that you don't need. It's the retain on this object that's preventing your
engine from going away.
If you have all of your creates/retain/release calls balanced, the engine
will go away when you release it when it returns failure for its init or
start (or most likely, initHardware) call.
They're a pain to find, but brute force is about the fastest way to actually
find them (if you get desperate you can use gdb to set breakpoints, but
that's not the first thing I would try). The Search feature of XCode to
look for "new" (or any static object creation method calls) and "retain" is
what I use...
Good luck,
Mark
--
Mark Cookson
M-Audio, a division of Avid
225 Locust St.
Hudson, WI 54016
On 2/23/05 10:47 AM, "Philip Lukidis" <email@hidden> wrote:
>
Hello. I'm a complete newbie to coreaudio, and I have a simple question,
>
whose answer I cannot find in the documentation. Perhaps it's very basic.
>
When a call of activateAudioEngine fails, what action beyond a call to
>
release() is necessary? I'm finding that I cannot unload my driver after
>
this occurrence, because myAudioEngine->free() is never called (although my
>
audio device is freed). I'm not sure that I have a right to call a double
>
release(), or the system may panic. The simple PCI audio driver sample does
>
not check the return value of activateAudioEngine.
>
>
Does anyone have an answer to this?
>
>
thanks,
>
>
Philip Lukidis
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Coreaudio-api mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden