[gst-devel] autoplugging and 0.9

[David Schleef]

On Mon, Aug 16, 2004 at 09:09:50PM +0100, Andy Wingo wrote:
> On Mon, 16 Aug 2004, Benjamin Otte wrote:
> 
> > Which brings me to another thing we recently discussed on IRC:
> > If we make threading implicit and allow the schedulers to insert new
> > threads (which obviously need queues), how do we guarantee or at least
> > achieve low latency?
> 
> I could be off my rocker, but I don't think multiple threads and low
> latency are achievable together. Multiple threads within the flow of
> data of course; other threads that are truly independent don't affect
> the latency.
> 
> The reason is that locking a mutex, necessary for sharing data between
> threads, is not a bounded-time operation.

Yes it is, if you have a finite number of threads and the time the
mutex is held is bounded, and threads are awakened in a sane order.
The last, I assume, is fixed in Linux-2.6.

One of the goals of changing the scheduling model is to make threads
unnecessary, but they will never be uninteresting, since they allow
full utilization of multiple processors and are (for now) the only
way to paint an X[V]Image at an exact time.



dave...