Re: [Jack-Devel] ringbuffer problems

Hi,

On Thursday 28 April 2011 07:28:27 Timothy Godfrey wrote:
> The jack_ringbuffer_t documentation says
> 
> "The key attribute of a ringbuffer is that it can be safely accessed
> by two threads simultaneously -- one reading from the buffer and the
> other writing to it -- without using any synchronization or mutual
> exclusion primitives. For this to work correctly, there can only be a
> single reader and a single writer thread. Their identities cannot be
> interchanged. "
> 
> What are the specifics of these conditions? I am writing to the
> ringbuffer in the process() callback of my application, and I am
> reading from the ringbuffer in the main loop. Does this count as being
> in different threads?

As the process()-callback is executed within the audio-thread jack creates for 
you inside your application, this counts as two different threads. And its 
exactly what the ringbuffer is created for.

> For testing/education purposes, does it work to read and write to the
> ringbuffer from within a single thread (to check that data went in
> correctly, for example), without any other threads involved?

That is also possible. The things that don't work with a lock-free ringbuffer 
like the one in jack:
 - Multiple writers in several threads writing to the same ringbuffer.
 - Multiple readers in several threads reading from the same ringbuffer.

> Would
> constitute a violation of the rule, 'Their identities cannot be
> interchanged'?

This rule is there because when one thread is the reader and then another 
thread starts reading, there is a chance that for a short time both threads 
read. And that is not safe without locks on the ringbuffer or somewhere else. 
(Same for writers.)

> Will I run into trouble if I am mixing a c++ (jack2?) application with
> a jack1 server?

That shouldn't give you any trouble...

Have fun,

Arnold

1303970606.23378_0.ltw:2,a <201104280803.11739.arnold at arnoldarts dot de>