[FFmpeg-devel] [Jack-Devel] [PATCH] libavdevice: JACK demuxer

Reimar Döffinger Reimar.Doeffinger
Sun Mar 8 01:28:09 CET 2009


On Sun, Mar 08, 2009 at 12:51:48AM +0100, Olivier Guilyardi wrote:
> > furthermore, you would be very foolish to imagine (especially based on 
> > your incredibly naive comments about uint8_t) that you understand the 
> > subtleties of these buffers. the JACK community (and a couple of other 
> > exclusively audio-focused development groups) have been working with 
> > this buffer design for many, many years, and we are absolutely confident 
> > that our buffers are (a) SMP/multi-core safe (b) as efficient as they 
> > can be without using assembler. you should also be aware that there are 
> > very good arguments for  the current structure of the ringbuffer code, 
> > which explicitly does NOT use any memory barriers. if you don't 
> > understand why it works without them, then you should probably refrain 
> > from commenting on the design of these buffers at all.

Just for anyone wondering, I think the code is here:
http://jackit.sourceforge.net/cgi-bin/lxr/http/source/libjack/ringbuffer.c

Unless I am missing something it is in the same state fifo.c was quite
some time ago already, current fifo.c has the advantage that it allows
using all space. Thus what makes it work without memory barriers is luck.




More information about the ffmpeg-devel mailing list