Re: [linux-audio-dev] Re: Sync Issues (was Re: External MIDI Sync using

[Roger Larsson <roger.larsson@skelleftea.mail.telia.com>]

Benno Senoner wrote:
> 
> On Thu, 28 Oct 1999, Paul Barton-Davis wrote:

> > to an old ISA-based Diamond Speedstar, the performance of my audio
> > stuff has gone down the tubes. i can't do *anything* with X and hope
> > to not get dropouts. this never happened with the Matrox card, and i
> > am very confident that as soon as Matrox ships me a new one, the
> > problem will go away.
> >
> > right now, my system is essentially useless for audio work unless you
> > avoid using X. totally useless. i can't even play WAV files without
> > *huge* dropouts, and thats on a *dual* PII-450 !
> 
> Yes,
> 
> the idiotproof test is the following:
> install 2.2.10 + lowlatency patch  ( not SMP !),
> tune the IDE diskas as described in the README
> and run a testsession
> if the GFX stress graph is the only bad graph,
> then I'm sorry but you have to throw away your videocard if you want
> to do reliable low latency stuff.
> :-)
> 

Or you can try to find out what part of your video card driver that
destroys latency and correct that.

You can find out the problematic spot with my patch:
 latency-profiling-2.2.10-r6.patch [for Linux 2.2.10]
I will put it on some own web page at the moment you could get it from:

http://www.gardena.net/benno/linux/audio/patches/latency-profiling-2.2.10-r5.patch

And remove this part of the patch since it is not correct...
(you should not be tempted to improve things in an instrumentation
patch...)
 
-       if (tq_scheduler)
+       //RL latency improvement, correct?
+       //RL skip if forced reschedule (a normal will arrive soon...)
+       if (!current->need_resched && 
+           tq_scheduler)

The results are messages in 'dmesg', latency time and instruction
pointers
are printed. You then decode the instruction pointers to routine name.
 
/RogerL

-- 

The Internet interprets Windows as damage, 
             and routes around it.

Roger Larsson
Skellefteå
Sweden