On Wed, Jul 23, 2014 at 03:16:06PM -0700, J. Scott Heppler wrote:
> The recent change acpi(4) change regarding bogus interrupts allowed me
> to install Current on an old Everex Stepnote VA1500V notebook.
> Everything works with two small nits regarding the sound.
> The more important one is that plugging in headphone to the green
> line-out jack mutes the speaker but does not provide and sound.
>
> The mic input jack also does not record anything using aucat - I
> rarely record though. The *wav file is generated for the appropriate
> amount of time but has not discernible output on playback.
does:
mixerctl record.adc-0:1_source=mic
fix the recording problem?
> **********
> $ mixerctl -av
> record.adc-0:1_mute=off [ off on ]
> record.adc-0:1=120,120
> inputs.mix_source=dac-0:1,mic,dac-4:5 { dac-0:1 mic dac-4:5 }
> inputs.mix_dac-0:1=120,120
> inputs.mix_mic=120,120
> inputs.mix_dac-4:5=120,120
> record.adc-0:1_source=mix [ mix mic ]
> inputs.sel2_source= [ ]
> outputs.sel2_mute=off [ off on ]
> outputs.sel2=126,126
> inputs.dac-2:3_mute=off [ off on ]
> inputs.dac-2:3=126,126
> inputs.dac-4:5_mute=off [ off on ]
> inputs.dac-4:5=126,126
> outputs.mic_source=dac-2:3 [ dac-2:3 ]
> outputs.mic_dir=input-vr50 [ none output input input-vr50 ]
> outputs.spkr_source=mix [ mix ]
> outputs.spkr_mute=off [ off on ]
> outputs.spkr=144,144
> outputs.spkr_boost=off [ off on ]
> outputs.master=135,135
> outputs.master.mute=off [ off on ]
> outputs.master.slaves=spkr { sel2 dac-2:3 dac-4:5 spkr }
> record.volume=120,120
> record.volume.mute=off [ off on ]
> record.volume.slaves=adc-0:1 { adc-0:1 } ***********
> I do not see any hp or line-out outputs and I am unsure what to make of
> the empty sel2_source.
>
> The notebook chipset is Via based
> ***********
> dmesg | grep azalia
> azalia0 at pci4 dev 1 function 0 "VIA HD Audio" rev 0x10: apic 1 int 17
> azalia0: codecs: VIA/0x1708
> audio0 at azalia0 ***********
>
> The Windows drivers for the same notebook use the ALC 655 azalia codec.
> The ALC 655 codec appears to be in OpenBSD only under auvia driver
> instead of the azalia driver
> From:
> http://openbsd.7691.n7.nabble.com/VIA-chipsets-support-of-CN700-VT8237R-VT1708-A-td92231.html
> ***********
> auvia0 at pci0 dev 17 function 5 "VIA VT8233 AC97" rev 0x60: apic 2 int 22
> (irq 5)
> ac97: codec id 0x414c4760 (Avance Logic ALC655 rev 0)
> audio0 at auvia0 ***********
>
> I found the codec specs for both the VT1708A and Realtek ALC 655 sound
> drivers and can provide those (browser crashed with both 60+ page *pdf's
> open)
>
> Is there some way to have either have azalia call up the ALC 655 codec
> or change the device code to call up auvia? In linux/alsa one can tweak
> the codec in /etc/modprobe.d.
>
> I debated sending this during the 5.6-beta freeze. On one hand, it
> would be good to fix, but on the other, I can't imagine this is a common
> problem.
The codec has 6 independent channels, corresponding to the
speakers, headphones, and whatever is the third one. You could try
running sndiod with the "-c 0:5" option, to make it use all
channels and see if you get some signal on the headphones.
>
> Any additional info, debug outputs and trials of patches I should be
> able to provide.
>
There are chipsets that can expose both intel hda and ac'97
interfaces; if so there's probably a BIOS setting to choose between
the two interfaces, do you have such setting? If so, you could try
them and see which one is works better.
[...]
Anyway, all this is supposed to work by default, so I'd be
interested in your dmesg of a kernel built with AZALIA_DEBUG
option.
