Re: [PATCH] hda: STAC927x invalid association value

["Anders Rønningen" <andersronningen@gmail.com>]

HG:
http://pastebin.ca/916114

1.0.16:
http://pastebin.ca/916117

This is a regression from 1.0.16. In 1.0.16 everything works nice, 
except that the volume changes almost seem exponential. With PCM and 
Front set to max, I can increase the Master slider from 0% up to about 
50% almost without hearing any change (or sound). From ~70% and up, the 
volume increases _very_ much with each step. Don't know if it's supposed 
to work that way.

--
Anders

Takashi Iwai wrote:
> At Sat, 23 Feb 2008 23:29:56 +0100,
> Anders Rønningen wrote:
>   
>> I finally got around to compiling the latest source.
>>
>> [   14.160000] ALSA 
>> /mnt/storage/source/alsa/mercurial/alsa-driver/pci/hda/hda_codec.c:2994: 
>> autoconfig: line_outs=1 (0xd/0x0/0x0/0x0/0x0)
>> [   14.160000] ALSA 
>> /mnt/storage/source/alsa/mercurial/alsa-driver/pci/hda/hda_codec.c:2998:    
>> speaker_outs=0 (0x0/0x0/0x0/0x0/0x0)
>> [   14.160000] ALSA 
>> /mnt/storage/source/alsa/mercurial/alsa-driver/pci/hda/hda_codec.c:3002:    
>> hp_outs=2 (0xa/0xf/0x0/0x0/0x0)
>> [   14.160000] ALSA 
>> /mnt/storage/source/alsa/mercurial/alsa-driver/pci/hda/hda_codec.c:3003:    
>> mono: mono_out=0x0
>> [   14.160000] ALSA 
>> /mnt/storage/source/alsa/mercurial/alsa-driver/pci/hda/hda_codec.c:3011:    
>> inputs: mic=0xc, fmic=0xe, line=0x0, fline=0x0, cd=0x0, aux=0x0
>> [   14.160000] ALSA 
>> /mnt/storage/source/alsa/mercurial/alsa-driver/pci/hda/../../alsa-kernel/pci/hda/patch_sigmatel.c:2287: 
>> dac_nids=3 (0x2/0x5/0x3/0x0/0x0)
>>
>> Now I have sound on both HP and Internal speakers, but levelcontrol is 
>> not ok.
>>
>> Internal Speakers: PCM and Front adjust the level, Master does nothing
>> Left HP Jack: PCM and Front adjust the level, Master does nothing
>> Right HP Jack: PCM adjusts level, Master and Front does nothing.
>>     
>
> I guess this is also same on 1.0.16?  Or is it a regression after
> 1.0.16?
>
> Could you take alsa-info.sh snapshot for each mixer state?  Then we
> can check the inconsistent mapping between mixer elements and the
> actual I/Os.
>
>
> thanks,
>
> Takashi
>