Re: [Openvortex-dev] Re: [ALSA - driver 0001047]: module hangs at seemingly random times

[Raymond]

Although au8810, au8820 and au8830 have 16 Sample Rate Convertors.

For au8820 and au8830 driver, snd_pcm_hw_params() will return "no more 
mixer error" and "no more sample rate convertor error" before all 
configured substreams are opened.

It seem to me that au8810's problem is related to snd_pcm_open()
after all configured substreams are opened.

May be you can change NR_ADB from 0x10 to 0x20 (a dirty workaround) ?


Alien wrote:
> I'm not too sure, but i _can_ tell you that lastly i had two weird errors 
> about something adb, when snd-au8810 didn't play any sound anymore, i was 
> able to kill all apps who use sound and modprobe -r snd-au8810, and modprobe 
> again, and without reboot i was able to get sound working again.

In your case, if there are one or more stereo stream playing, 
snd_pcm_hw_params() will return "no more sample rate convertor error" 
before all configured substreams are opened.


> I must say that i haven't been on this PC much and haven't had much time to 
> test these things, allthough i'm planning too, (but i think i'll update CVS 
> first with these 2 patches who got lost accidentally before i test this. I 
> hope i'll get some time this week(end).

Igor V. Kovalenko already include the two lost patches in his patch to 
fix oopses on amd64

> thanks for all the help and info concerning snd-au8810 (in 64bit)
>
> AL13N
>
> Op zondag 31 juli 2005 09:59, schreef Raymond:
>
> > If there are 16 instances of aplay playing mono stream.
> >
> > What error will be returned by snd_pcm_open() for the 17th instance of
> > aplay ?
> >
> >
> >
> > alsa-kernel/pci/au88x0/au8810.h
> >
> > #define NR_ADB   0x10
> >
> > alsa-kernel/pci/au88x0/au8830.h
> >
> > #define NR_ADB   0x20
> >
> >
> > alsa-kernel/pci/au88x0/au88x0.c
> >
> >         if ((err = snd_vortex_new_pcm(chip, VORTEX_PCM_ADB, NR_ADB)) < 0) {
> >                 snd_card_free(card);
> >                 return err;
> >         }