Subject: Re: [linux-audio-dev] Audio engine stuff
From: sbenno_AT_em.gardena.net
Date: to helmi 17 2000 - 03:40:26 EST
> >Anyway disk buffering schemes are as old as ENIACS :-)
>
> The simple disk buffering is different; some buffering schemes may not
> give the fast start at playing! Also, my audiofile virtual memory system
> is yet another scheme.
the fast start playing is really more trivial that the
"Amazon 1-click TM" thing.
:-)
just mlock() the first part of the mmap()ed sample
>
> You may see basic buffering in programs but not these advanced ones.
> Some Linux programs still loads the whole audiofiles/samples to memory.
>
> If you can't find a single program (earlier than 1997) with preload for
> fast start at playing, please let me know! (Or with special audiofile
> virtual memory...)
>
> >a lock-free ringbuffer
>
> I guess I have missed something in recent discussions...
Actually it was a private discussion between me, PBD, and David Olofson,
it took a few mails to convince Paul that my mmap() + lock-free ringbuffer
concept for multitrack is actually correct and efficient.
:-)
>
> Hmm.. hey, no comments about inserting plug-ins to the engine on-the-fly?
I see this not as a big deal:
just use one thread with non RT-privileges to load the plugin() on the
fly via dlopen() , mlock() the plugin-code , and then call the
instantiation code.
When done, notify the main engine (running with RT privileges) that
the plugin is ready for execution.
At this point you are now able to call any functions of the plugin
without nasty delays.
Juhana, enough for now ? :-)
I hope that I am not wrong. (folks, correct me please)
Benno.
This archive was generated by hypermail 2b28 : pe maalis 10 2000 - 07:23:27 EST