Re: [Jack-Devel] ALSA PCM multi plugin and xruns

PrevNext  Index
DateSat, 08 Dec 2012 15:48:35 -0800
From Devin Anderson <[hidden] at gmail dot com>
To[hidden] at chriscaudle dot org
CcJack devel <[hidden] at lists dot jackaudio dot org>
In-Reply-ToChris Caudle Re: [Jack-Devel] ALSA PCM multi plugin and xruns
Hi Chris,

On Sat, Dec 8, 2012 at 1:26 PM, Chris Caudle <[hidden]> wrote:
> On Sat, December 8, 2012 5:57 am, Devin Anderson wrote:
>> This leads me to believe that the poll descriptors
>> indicate that there is data available on the master device,
>> but there isn't any data available on one of the other slave devices.
>
> I don't remember hearing other people talk about this interface before.
>
> Does the 3G version still use an Envy24 chip?

IDK.  `envy24control` says, "No ICE1712 cards found", but I'm not sure
if that sufficiently answers your question.

> Are you absolutely sure that the synchronization is set up correctly so
> that the cards are sharing word clock?

I *believe* I've set up word clock sync correctly using `echomixer`,
though I don't know how to verify that word clock is *actually*
working.  I assume that word clock is working because everything runs
smoothly most of the time - that is, I think, until I do something
that causes an initial xrun, like loading a song in Renoise, or
starting an application that uses JACK.  Not all *real* xruns cause
the spew of xruns, but there's some set of circumstances where it
happens.  I'm going to try to figure out exactly what that set of
circumstances is.

I just tried replacing the cable with a cable that I verified is
working with a cable tester, and I'm able to duplicate the problem.

> What you described sounds like
> what might happen if the clock sync starts slipping between two cards and
> they are not filling their buffers at a locked rate any more.

If that happened, I would assume that the cards would each output the
same amount of frames at different times (please correct me if I'm
wrong).  I can't yet verify that this is or isn't happening.

I'm confused by why there are 32 frames available for reading *until*
256 more frames are made available by the card.

This thread seems relevant to the situation:

    http://linux-audio.4202.n7.nabble.com/ALSA-multi-plugin-and-JACK-xrun-callbacks-td74462.html

I'm open to ideas on how to proceed.  For now, I'm going to add some
error messages to the multi plugin in ALSA, and see if I can figure
out what's going on.

Thanks,

-- 
Devin Anderson
surfacepatterns (at) gmail (dot) com

blog - http://surfacepatterns.blogspot.com/
psinsights - http://psinsights.googlecode.com/
synthclone - http://synthclone.googlecode.com/
PrevNext  Index

1355010522.12496_0.ltw:2, <CAG7zqToqqMpzaMTf2cCtkn87JKcxROKhRh6kx8YikHY6UNtUGw at mail dot gmail dot com>