Subject: RE: [linux-audio-dev] Real time plug ins
From: Richard W.E. Furse (richard_AT_muse.demon.co.uk)
Date: Tue Mar 14 2000 - 14:20:51 EST
Nice to see a new potential LADSPA host! I'm a little uneasy on the
practicalities in this case though, for a couple of reasons. Hopefully we
can think of a way around them:
(1) Currently, LADSPA plugin memory allocation is handled implicitly inside
instantiate(). I suppose this could be changed, but is it really a problem?
Perhaps there's some hacky way any malloc() call could be re-routed through
SC without confusing the current interface?
(2) There is no numeric timing information provided by the LADSPA interface
- the host is assumed to time the plugin to check up on its speed (see the
'Hard RT' flag though). SC could simply time the instantiate() calls for
potential plugins offline and check against some tolerance. I've no
fundamental objection to adding an extra flag to indicate 'Fast Start'
plugins, however at the moment this is a very woolly concept and I cannot
help but think SC would be get much more accurate information by working
things out for itself.
-- Richard
-----Original Message-----
From: Benno Senoner [SMTP:sbenno_AT_gardena.net]
Sent: Tuesday, March 14, 2000 5:33 PM
To: James McCartney
Cc: linux-audio-dev_AT_ginette.musique.umontreal.ca
Subject: Re: [linux-audio-dev] Real time plug ins
[...]
I see a solution here:
we should introduce the following convention for plugin-instantiation:
there could be two categories of plugins:
1)
the ones whose instantiation functions are not hard-realtime:
these plugin should do instantiation in the low-priority thread
2)
the ones which can instantiate "in realtime", that means
no malloc() / disk I/O etc allowed.
James, If you said that SC has its own memory management, the problem is
basically solved.
Since you simply have to instantiate the plugins in the main DSP loop of
the
host.
But if a plugin breaks this rule , then audio dropouts during instantiation
will be unavoidable.
LAD folks, is my solution acceptable ?
Benno.
This archive was generated by hypermail 2b28 : Tue Mar 14 2000 - 22:32:17 EST