[alsa-devel] Port C Out selector , Port E Out selector of ad1988 and ad1988_3stack_dac_nids

[Raymond Yau]

2011/9/17 Raymond Yau <superquad.vortex2 at gmail.com>:
> 2011/9/16 Takashi Iwai <tiwai at suse.de>:
>> At Fri, 16 Sep 2011 11:59:54 +0800,
>> Raymond Yau wrote:
>>>
>>> 2011/9/12 Takashi Iwai <tiwai at suse.de>:
>>> > At Wed, 31 Aug 2011 11:21:00 +0800,
>>> > Raymond Yau wrote:
>>> >>
>>> >> HDA - Add Headphone Playback Volume control for ad1988/ad1989
>>> >>
>>> >> - use DAC0 instead of DAC1 for Port-A Headphone
>>> >> - assign 0x03 to spec->multiout.hp_nid except model="6stack-dig-fp"
>>> >
>>> > Thanks, applied now.
>>> >
>>> >
>>> > Takashi
>>> >
>>>
>>> while testing the headphone volume control ,
>>>
>>> I discover that the "Port C Out selector" and "Port E Out Selector"
>>> (i.e. node 0x31 and 0x32 ) by commit in patch_anlog.c
>>>
>>> fd66e0d0591dd12eb0bea1e9f3aa194bb93cebbd
>>>
>>>
>>>
>>>  * AD1988
>>>  *
>>>  * Output pins and routes
>>>  *
>>>  *        Pin               Mix     Sel     DAC
>>>  * port-A 0x11 (mute/hp) <- 0x22 <- 0x37 <- 03/04/06
>>>  * port-B 0x14 (mute/hp) <- 0x2b <- 0x30 <- 03/04/06
>>>  * port-C 0x15 (mute)    <- 0x2c <- 0x31 <- 05/0a
>>>  * port-D 0x12 (mute/hp) <- 0x29         <- 04
>>>  * port-E 0x17 (mute/hp) <- 0x26 <- 0x32 <- 05/0a
>>>
>>>
>>>
>>> The connection list of node 0x31 and 0x32 are different from datasheet
>>>
>>>
>>> Table 10. Audio Widget
>>>
>>> 31                           Audio selector Selects DAC_2 and DAC_4 for PORT-C
>>>      PORT-E Out Selector
>>> 32                           Audio selector Selects DAC_2 and DAC_4 for PORT-E
>>>      PORT-C In Selector
>>>
>>>
>>>
>>> Codec: Analog Devices AD1988
>>> Address: 0
>>> AFG Function Id: 0x1 (unsol 0)
>>> Vendor Id: 0x11d41988
>>> Subsystem Id: 0x104381e1
>>> Revision Id: 0x100400
>>>
>>>
>>> Node 0x31 [Audio Selector] wcaps 0x300101: Stereo
>>>   Connection: 2
>>>      0x04* 0x0a
>>> Node 0x32 [Audio Selector] wcaps 0x300101: Stereo
>>>   Connection: 2
>>>      0x05* 0x04
>>>
>>>
>>>
>>> is the following patch based on real hardware ?
>>>
>>>
>>> commit 8c7c7b8dd2828b42c1230c6b0235e7d1dcf57e5
>>>
>>> hda-codec - Fix surrounds on 3stack mode of AD1988
>>
>> I don't see this commit id in my tree...
>>
>>
>> Takashi
>>
>
> http://git.alsa-project.org/?p=alsa-kernel.git;a=commit;h=fd66e0d0591dd12eb0bea1e9f3aa194bb93cebbd
>
> Thu, 17 Nov 2005 14:31:34 +0000 (15:31 +0100)
> Add AD1988 codec support to hda-codec driver.
>
>
> http://git.alsa-project.org/?p=alsa-kernel.git;a=commitdiff;h=f8c7c7b8dd2828b42c1230c6b0235e7d1dcf57e5;hp=d32410b1095cf93e8e31f8919de46f496d7b3ce0
>
> Thu, 24 Nov 2005 15:17:20 +0000 (16:17 +0100)]
> Fixed the swapped surround/CLFE on 3stack mode of AD1988.
>
>
> hda-emu codecs/ad1988a -m "3stack"
> route -a 0x15
>
> Out[0a] -x Sel[31] -- Mix[2c] -> Pin[15]
> Out[04] -> Sel[31] -- Mix[2c] -> Pin[15]
>
> route -a 0x17
>
> Out[04] -> Sel[32] -- Mix[26] -> Pin[17]
> Out[05] -x Sel[32] -- Mix[26] -> Pin[17]
>

In simple word,

surround51 does not work with 3stack model since the above route in
hda-emu indicate that blue jack and pink jack are connected to DAC1
(node 0x4)