[alsa-devel] HDA intel, wired headset w/mic, jack mic detect

[Kevin Hilman]

Takashi Iwai <tiwai at suse.de> writes:

> At Tue, 6 Mar 2012 09:28:12 +0800,
> Raymond Yau wrote:
>> 
>> 2012/3/6, Kevin Hilman <khilman at ti.com>:
>> > Raymond Yau <superquad.vortex2 at gmail.com> writes:
>> >
>> >> 2012/3/3, Kevin Hilman <khilman at ti.com>:
>> >>> Raymond Yau <superquad.vortex2 at gmail.com> writes:
>> >>>
>> >>>> 2012/3/2, Kevin Hilman <khilman at ti.com>:
>> >>>>
>> >>>>
>> >>>>>> Does your headset/mic has a TRRS (Tip, ring, ring, sleeve) connector
>> >>>>>> instead of TRS (Tip, ring, sleeve) connector ?
>> >>>>
>> >>>>> Yes.  It's TRRS.
>> >>>>
>> >>>> Does it mean that the headset/mic is a mono mic ?
>> >>>>
>> >>>> Is the internal mic mono or stereo ?
>> >>>
>> >>> I believe it's mono.  Based on vumeter activity, and the fact that
>> >>> changing only one slider in hda_analyzer is needed to quiet/mute the
>> >>> mic.
>> >>>
>> >>
>> >> Not sure any difference between cs4206 and cs4207
>> >>
>> >> It seem that you can select ADC1 and ADC2 channel mode in cs4207
>> >
>> >> ‘00’b - ADC2 left channel is mapped to HDA left
>> >> channel and ADC2 right channel is mapped HDA
>> >> right channel (normal mode).
>> >> ‘01’b - ADC2 left channel is mapped to both HDA
>> >> left and right channels. ADC2 right channel is
>> >> discarded (mono mode).
>> >> ‘10’b - ADC2 right channel is mapped to both
>> >> HDA left and right channels. ADC2 left channel is
>> >> discarded (alternate mono mode).
>> >> ‘11’b - ADC2 left channel is mapped to HDA right
>> >> channel and ADC2 right channel is mapped to
>> >> HDA left channel (channel swap mode).
>> >
>> > Do any of the hda tools allow me to tweak this from userspace?
>> >
>> >
>> 
>> you should ask Takarshi as he added this patch, he may know whether
>> cs4206 support the above feature
>> 
>> http://git.kernel.org/?p=linux/kernel/git/tiwai/sound.git;a=commitdiff;h=40c20fa05a29766565f56ede17d0ffa539e1c9a9;hp=ea35929b886975a240660b3ba6c61826761731ad
>
> You can play with hda-emu to send vendor-specific COEF, as found in
> the commit above.  But it's at your own risk; the patch I made is just
> a translation from other information I've got from Cirrus at that
> time.

Unfortunately, I'm not really familiar with audio codecs, and am finding
that undersanding the verbs and and sending the right ones with the
right coef (and finding out what was actually written) seems to be a bit
out of my league.

If someone can suggest what to try with hda-emu, I'm willing, but
otherwise I guess I have to opt for an external USB audio adapater for
the headset. :(

Kevin