[Prev] Thread [Next]  |  [Prev] Date [Next]

Re: Does CoreAudio use the alarm signal in the IO proc thread? Herbie Robinson Wed Jan 24 00:01:40 2007

At 1:59 PM -0800 1/23/07, Corey O'Connor wrote:
I'm trying to figure out why I cannot use Haskell from a CoreAudio HAL IO proc. One thing Haskell does, which is questionable IMO, is use the alarm signal for it's own thread scheduler. The program is behaving as if the Haskell code is terminating early, but only in the HAL IO proc thread. I think this would be consistent with the Haskell code triggering a signal but instead of the Haskell scheduler recieving the signal another signal handler is triggered. Is it possible that CoreAudio registers it's own signal handler in the IO proc thread? Or are there other caveats to the CoreAudio realtime thread that would account for this behavior?

Signals aren't a native thing for MACH, right?

That would mean they are probably implemented with message passing and something context dependent to catch the messages...
**  http://www.curbside-recording.com/  **
Do not post admin requests to the list. They will be ignored.
Coreaudio-api mailing list      ([EMAIL PROTECTED])
Help/Unsubscribe/Update your Subscription:

This email sent to [EMAIL PROTECTED]