HID and threads WAS: [PD-dev] audio interruptions from outside events (only with Pd)

Hans-Christoph Steiner hans at eds.org
Mon Dec 4 23:17:27 CET 2006


On Dec 4, 2006, at 4:55 PM, Tim Blechmann wrote:

> hi hans,
>
>> [comport] and [hid] both use the standard calls and both do not use
>> threads at all, yet neither has problems with blocking I/O (or at
>> least no one has been able to make a patch that causes it to
>> happen).  I am pretty sure they both use high priority queues in the
>> kernel so that return very quickly.
>
> after having a brief look at the hidio sources, you seem to use a  
> mutex
> for synchronization with a child thread ... this is basically not save
> and may lead to audio dropouts, when running with low latencies. i  
> know,
> miller's portaudio implementation is not able to run under low
> latencies, but the callback scheduler of devel is able to do so...

We are currently in the discussion of how to use threads in externals/ 
io/hidio, so we are testing various configurations, so this is  
probably going to change drastically.  (one hidio author that shall  
remain unnamed prefers to avoid this list for understandable but  
unfortunate reasons).

How would you handle this particular situation?

.hc



------------------------------------------------------------------------

The arc of history bends towards justice.     - Dr. Martin Luther  
King, Jr.






More information about the Pd-dev mailing list