Re: Re: intel8x0, ad198x, no sound

[Takashi Iwai <tiwai@suse.de>]

At Sun, 18 May 2003 14:17:25 -0700,
Jeffrey Baker wrote:
> 
> On Sun, May 18, 2003 at 01:27:06PM -0700, Jeffrey Baker wrote:
> > On Sun, May 18, 2003 at 12:46:17PM -0700, Jeffrey Baker wrote:
> > > This is related to the mail last week regarding the Asus P4PE with
> > > ad1980 sound chip.  I got an Asus P4P800 with the ad1985.  The
> > > intel8x0 driver works but there are problems with the sound routing.
> > > There is no sound at all coming from the main (line out) output.
> > > If I plug headphones into the line in jack, there is sound but it
> > > appears to be some kind of surround output, because it lacks all low
> > > frequencies entirely.  No sound comes from the microphone jack.
> > > 
> > > I think the problem may be related to the "intelligent" jack sensing
> > > built into this hardware.  It is supposed to be able to detect when
> > > you've got your headphones plugged into the wrong jack and reroute
> > > the audio.  Either that, or the chip is stuck in some surround mode
> > > that isn't quite working.
> > 
> > Hi.  It helps to read the datasheet.  Although this isn't stated in
> > the motherboard's documentation, the headphone amplifier is, in
> > fact, connected to the line-in jack.  There is a register (0x76)
> > which controls the input for the headphone amp.  It can be either 0
> > for the surround DAC or 1 for the mixer.  By default it is 0.  This
> > explains why I am hearing high-passed surround audio from the
> > headphones.
> > 
> > I glanced at intel8x0.c, but it wasn't immediately obvious how to
> > set registers on the codec.
> 
> Ah OK.  I added this patch to ac97_codec.c:
> 
> --- ac97_codec.c        2003-04-11 00:45:23.000000000 -0700
> +++ /root/ac97_codec.c  2003-05-18 14:12:27.000000000 -0700
> @@ -104,6 +104,7 @@
>  { 0x41445363, 0xffffffff, "AD1886A",           patch_ad1881,   NULL },
>  { 0x41445370, 0xffffffff, "AD1980",            patch_ad1980,   NULL },
>  { 0x41445372, 0xffffffff, "AD1981A",           patch_ad1881,   NULL },
> +{ 0x41445375, 0xffffffff, "AD1985",            patch_ad1980,   NULL },
>  { 0x414c4300, 0xfffffff0, "RL5306",            NULL,           NULL },
>  { 0x414c4310, 0xfffffff0, "RL5382",            NULL,           NULL },
>  { 0x414c4320, 0xfffffff0, "RL5383",            NULL,           NULL },
> 
 
thanks, the above patch is already in cvs.

> This makes proper headphone-power audio emerge from the line-out
> jack.  But, the main level is controlled by the "Surround" slider in
> the mixer, which isn't exposed to the OSS mixer interface.  
> 
> It seems like mixer routing for this codec is a little screwed up.
> If the SURR_OUT and LINE_OUT are going to be swapped once in
> software (reg 0x76 <- 0x420), and then the motherboard vendor is
> swapping them again in hardware (by wiring the SURR_OUT/HP_OUT pin
> to the line out and the LINE_OUT to the LINE_IN), it's all very
> confusing.
> 
> What needs to happen to make the mixer sliders correspond to their
> actual effects?

then we'll need to stop swapping them.
or, more simply, we can swap the controls surround and line-out.

anyway, the workaround would be device-basis.  please let me know the
sub vendor/device id (output of lspci -vv and lspci -nvv).


Takashi


-------------------------------------------------------
This SF.net email is sponsored by: If flattening out C++ or Java
code to make your application fit in a relational database is painful, 
don't do it! Check out ObjectStore. Now part of Progress Software.
http://www.objectstore.net/sourceforge