[gst-devel] 0.9 proposals
Ronald S. Bultje
rbultje at ronald.bitfreak.net
Fri Dec 3 00:48:01 CET 2004
On Thu, 2004-12-02 at 19:44, David Schleef wrote:
> Who cares? Making the gstreamer core faster or slower by a factor
> of two is mostly irrelevant in the grand scheme of things. I'd
> *much* prefer to have a system that everyone can understand and
> bug fix with an overhead of 6% than a complicated system at 3%.
> The time that I save from fixing gstreamer bugs I will spend on
> liboil, which can have a direct and massive affect on applications.
> We can always optimize later, *once the bottlenecks are known*.
I'm fine with this point. Now, for 0.10, you're proposing to remove
threads and re-introduce them for 0.12. I say that's a very bad idea.
It's like introducing security features afterwards: if it wasn't
designed for it, it'll fail. Horribly. You're a designer, Dave, you know
this. Wim will help here, maybe even I can help a hand. Making something
threadsafe, especially if it's documented well, is not hard (though not
easy either). When multiple people know how it works and are able to
review each others work, it's as easy as that. A threadsafe core does
not automatically imply a non-simple core.
The good thing of using threads is that elements will stay simple. I can
block randomly if I wish, meaning demuxers and parsers stay simple.
Ronald S. Bultje <rbultje at ronald.bitfreak.net>
More information about the gstreamer-devel