Intel Azalia problems

Intel Azalia problems

[Josh Green]

Hello, I recently purchased a new laptop which has an Intel Azalia sound
interface.  I noticed that this is a pretty recent addition to the ALSA
drivers.  Although it seems to be basically working, I'm having a couple
problems.

The first problem has to do with the headphone jack, which on this
laptop is an SPDIF/output combo jack. I can hear output on the built in
speakers (after un-muting and turning up the "Front" level), but when I
plug in headphones I don't hear anything (despite the fact that the
"Headphone" mixer is turned up and unmuted).

The other problem is with certain applications which use oss.  Quake3
for example prints the following in the terminal it is started from:

------- sound initialization -------
Could not mmap dma buffer PROT_WRITE|PROT_READ
trying mmap PROT_WRITE (with associated better compatibility / less
performance code)
/dev/dsp: Input/output error
Could not mmap /dev/dsp
------------------------------------

An strace reveals this, which may or may not be helpful.

[pid 15829] open("/dev/dsp", O_RDWR)    = 20
[pid 15829] getuid32()                  = 501
[pid 15829] setresuid32(-1, 501, -1)    = 0
[pid 15829] ioctl(20, SNDCTL_DSP_GETCAPS, 0xbffff574) = 0
[pid 15829] ioctl(20, SNDCTL_DSP_SPEED or SOUND_PCM_READ_RATE,
0x81776ac) = 0
[pid 15829] ioctl(20, SNDCTL_DSP_STEREO, 0xbffff56c) = 0
[pid 15829] ioctl(20, SNDCTL_DSP_SPEED or SOUND_PCM_READ_RATE,
0x88446b4) = 0
[pid 15829] ioctl(20, SNDCTL_DSP_SETFMT or SOUND_PCM_READ_BITS,
0xbffff568) = 0
[pid 15829] ioctl(20, SNDCTL_DSP_GETOSPACE, 0xbffff580) = 0
[pid 15829] mmap2(NULL, 32768, PROT_READ|PROT_WRITE, MAP_SHARED, 20, 0)
= -1 EIO (Input/output error)
[pid 15829] write(2, "Could not mmap dma buffer PROT_W"..., 47Could not
mmap dma buffer PROT_WRITE|PROT_READ
) = 47
[pid 15829] write(2, "trying mmap PROT_WRITE (with ass"..., 86trying
mmap PROT_WRITE (with associated better compatibility / less performance
code)

Other OSS programs (such as "play") work fine, so perhaps its just
programs that try and MMAP the device.

I copied the file /usr/share/alsa/cards/Azalia.conf to /etc/asound.conf,
but with the same results.  Despite the low volume output of the
internal speakers, I can hear clicks in the output with mplayer and
totem (perhaps a sampling rate issue?)

Here is some info about my system:
alsa-driver 1.0.9rc2
alsa-lib 1.0.9rc2
alsa-utils 1.0.8
kernel 2.6.7-gentoo-r11
NVidia GeForce Go 6600 (I'm using the commercial drivers)

ALSA modules loaded:

snd_hda_intel          16496  2
snd_hda_codec          59008  1 snd_hda_intel
snd_seq_oss            36480  0
snd_seq_midi_event      8064  1 snd_seq_oss
snd_seq                53392  4 snd_seq_oss,snd_seq_midi_event
snd_seq_device          8972  2 snd_seq_oss,snd_seq
snd_pcm_oss            52008  0
snd_pcm                92680  3 snd_hda_intel,snd_hda_codec,snd_pcm_oss
snd_timer              25604  2 snd_seq,snd_pcm
snd_page_alloc         10120  2 snd_hda_intel,snd_pcm
snd_mixer_oss          20096  2 snd_pcm_oss
snd                    54884  13
snd_hda_intel,snd_hda_codec,snd_seq_oss,snd_seq_midi_event,snd_seq,snd_seq_device,snd_pcm_oss,snd_pcm,snd_timer,snd_mixer_oss
soundcore               9952  2 snd


I would appreciate any info on using this card with ALSA.  Also, it
seems my modem is tied into this card as well (HDAUDIO soft modem), any
idea if this is supported or if there are plans to?  Cheers!
Josh Green




-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click

Re: Intel Azalia problems

[Takashi Iwai]

Hi Josh,

At Wed, 13 Apr 2005 05:07:07 -0700,
Josh Green wrote:
> 
> Hello, I recently purchased a new laptop which has an Intel Azalia sound
> interface.  I noticed that this is a pretty recent addition to the ALSA
> drivers.  Although it seems to be basically working, I'm having a couple
> problems.
> 
> The first problem has to do with the headphone jack, which on this
> laptop is an SPDIF/output combo jack. I can hear output on the built in
> speakers (after un-muting and turning up the "Front" level), but when I
> plug in headphones I don't hear anything (despite the fact that the
> "Headphone" mixer is turned up and unmuted).

Which HDA codec?  Please attach /proc/asound/card0/codec#*/* file.
(I guess it's CM9880, though..)
Anyway, try the latest CVS version.  Some improvement has been done
since rc2.

> The other problem is with certain applications which use oss.  Quake3
> for example prints the following in the terminal it is started from:
> 
> ------- sound initialization -------
> Could not mmap dma buffer PROT_WRITE|PROT_READ
> trying mmap PROT_WRITE (with associated better compatibility / less
> performance code)
> /dev/dsp: Input/output error
> Could not mmap /dev/dsp
> ------------------------------------

If your codec is CM9880, it supports only 48/96/192kHz while quake
prefers 44.1kHz.  This conflict can't be solved in the OSS mmap mode.

Try to tune via oss proc file as described in
Documentation/OSS-Emulation.txt.


Takashi


-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click

Re: Intel Azalia problems

[Josh Green]

[-- Attachment #1: Type: text/plain, Size: 2656 bytes --]

On Wed, 2005-04-13 at 16:33 +0200, Takashi Iwai wrote: 
> Hi Josh,
> 
> At Wed, 13 Apr 2005 05:07:07 -0700,
> Josh Green wrote:
> > 
> > Hello, I recently purchased a new laptop which has an Intel Azalia sound
> > interface.  I noticed that this is a pretty recent addition to the ALSA
> > drivers.  Although it seems to be basically working, I'm having a couple
> > problems.
> > 
> > The first problem has to do with the headphone jack, which on this
> > laptop is an SPDIF/output combo jack. I can hear output on the built in
> > speakers (after un-muting and turning up the "Front" level), but when I
> > plug in headphones I don't hear anything (despite the fact that the
> > "Headphone" mixer is turned up and unmuted).
> 
> Which HDA codec?  Please attach /proc/asound/card0/codec#*/* file.
> (I guess it's CM9880, though..)
> Anyway, try the latest CVS version.  Some improvement has been done
> since rc2.
> 

cat /proc/asound/card0/codec#0 >/tmp/hda_intel_codec.txt attached.  I
suppose "Codec: Realtek ALC880" is what you are asking about.  I just
checked out alsa-driver, alsa-kernel, alsa-lib cvs, same results though.
I think the clicking problem is due to the 48Khz fixed rate, since it
seems like it only affects programs that try to play at other rates
(such as 44.1Khz).  I'll do my homework and figure out how to setup
sample rate conversion, but if someone wants to offer me some tips for
adding this to the default device in asound.conf I'd appreciate it :)

The headphone jack is still an issue though, no output from there.  I'll
get familiar with the source code to get an idea of things to mess with
(trying other mixer targets, etc).

> > The other problem is with certain applications which use oss.  Quake3
> > for example prints the following in the terminal it is started from:
> > 
> > ------- sound initialization -------
> > Could not mmap dma buffer PROT_WRITE|PROT_READ
> > trying mmap PROT_WRITE (with associated better compatibility / less
> > performance code)
> > /dev/dsp: Input/output error
> > Could not mmap /dev/dsp
> > ------------------------------------
> 
> If your codec is CM9880, it supports only 48/96/192kHz while quake
> prefers 44.1kHz.  This conflict can't be solved in the OSS mmap mode.
> 
> Try to tune via oss proc file as described in
> Documentation/OSS-Emulation.txt.
> 

Gave this a try, Quake3 starts up and I hear sound for the initial "ID"
stamp intro, but the game locks up when you start a level (not a hard
lockup, can still change to virtual terminal and kill).

> 
> Takashi
> 

Thanks for the help.  Maybe I'll see you at the Linux Audio Developers
Conference?  Cheers!
	Josh Green


[-- Attachment #2: hda_intel_codec.txt --]
[-- Type: text/plain, Size: 6316 bytes --]

Codec: Realtek ALC880
Address: 0
Vendor Id: 0x10ec0880
Subsystem Id: 0x0
Revision Id: 0x90500
Default PCM: rates 0x560, bits 0x0e, types 0x1
Default Amp-In caps: N/A
Default Amp-Out caps: N/A
Node 0x02 [Audio Output] wcaps 0x411: Stereo
  PCM: rates 0x560, bits 0x0e, types 0x1
Node 0x03 [Audio Output] wcaps 0x411: Stereo
  PCM: rates 0x560, bits 0x0e, types 0x1
Node 0x04 [Audio Output] wcaps 0x411: Stereo
  PCM: rates 0x160, bits 0x0e, types 0x1
Node 0x05 [Audio Output] wcaps 0x411: Stereo
  PCM: rates 0x160, bits 0x0e, types 0x1
Node 0x06 [Audio Output] wcaps 0x211: Stereo Digital
  PCM: rates 0x160, bits 0x1e, types 0x1
Node 0x07 [Audio Input] wcaps 0x10051b: Stereo Amp-In
  Amp-In caps: ofs=0x00, nsteps=0x23, stepsize=0x03, mute=1
  Amp-In vals: 0x00 0x00
  PCM: rates 0x160, bits 0x06, types 0x1
  Connection: 7
     0x18 0x19 0x1a* 0x1b 0x1c 0x14 0x15
Node 0x08 [Audio Input] wcaps 0x10051b: Stereo Amp-In
  Amp-In caps: ofs=0x00, nsteps=0x23, stepsize=0x03, mute=1
  Amp-In vals: 0x00 0x00
  PCM: rates 0x160, bits 0x06, types 0x1
  Connection: 7
     0x18 0x19 0x1a* 0x1b 0x1c 0x14 0x15
Node 0x09 [Audio Input] wcaps 0x10051b: Stereo Amp-In
  Amp-In caps: ofs=0x00, nsteps=0x23, stepsize=0x03, mute=1
  Amp-In vals: 0x00 0x00
  PCM: rates 0x160, bits 0x06, types 0x1
  Connection: 10
     0x18 0x19 0x1a 0x1b 0x1c* 0x0b 0x14 0x15 0x16 0x17
Node 0x0a [Audio Input] wcaps 0x100391: Stereo Digital
  PCM: rates 0x160, bits 0x1e, types 0x1
  Connection: 1
     0x1f
Node 0x0b [Audio Mixer] wcaps 0x20010b: Stereo Amp-In
  Amp-In caps: ofs=0x23, nsteps=0x41, stepsize=0x03, mute=1
  Amp-In vals: 0x00 0x00
  Connection: 8
     0x18 0x19 0x1a 0x1b 0x1c 0x1d 0x14 0x15
Node 0x0c [Audio Mixer] wcaps 0x20010f: Stereo Amp-In Amp-Out
  Amp-In caps: ofs=0x00, nsteps=0x00, stepsize=0x00, mute=1
  Amp-In vals: 0x00 0x00
  Amp-Out caps: ofs=0x40, nsteps=0x40, stepsize=0x03, mute=0
  Amp-Out vals: 0x36 0x36
  Connection: 2
     0x02 0x0b
Node 0x0d [Audio Mixer] wcaps 0x20010f: Stereo Amp-In Amp-Out
  Amp-In caps: ofs=0x00, nsteps=0x00, stepsize=0x00, mute=1
  Amp-In vals: 0x00 0x00
  Amp-Out caps: ofs=0x40, nsteps=0x40, stepsize=0x03, mute=0
  Amp-Out vals: 0x35 0x35
  Connection: 2
     0x03 0x0b
Node 0x0e [Audio Mixer] wcaps 0x20010f: Stereo Amp-In Amp-Out
  Amp-In caps: ofs=0x00, nsteps=0x00, stepsize=0x00, mute=1
  Amp-In vals: 0x00 0x00
  Amp-Out caps: ofs=0x40, nsteps=0x40, stepsize=0x03, mute=0
  Amp-Out vals: 0x35 0x30
  Connection: 2
     0x04 0x0b
Node 0x0f [Audio Mixer] wcaps 0x20010f: Stereo Amp-In Amp-Out
  Amp-In caps: ofs=0x00, nsteps=0x00, stepsize=0x00, mute=1
  Amp-In vals: 0x00 0x00
  Amp-Out caps: ofs=0x40, nsteps=0x40, stepsize=0x03, mute=0
  Amp-Out vals: 0x2e 0x2e
  Connection: 2
     0x05 0x0b
Node 0x10 [Audio Selector] wcaps 0x300101: Stereo
  Connection: 4
     0x0c* 0x0d 0x0e 0x0f
Node 0x11 [Audio Selector] wcaps 0x300101: Stereo
  Connection: 4
     0x0c 0x0d* 0x0e 0x0f
Node 0x12 [Audio Selector] wcaps 0x300101: Stereo
  Connection: 4
     0x0c* 0x0d 0x0e 0x0f
Node 0x13 [Audio Selector] wcaps 0x300101: Stereo
  Connection: 4
     0x0c* 0x0d 0x0e 0x0f
Node 0x14 [Pin Complex] wcaps 0x40018d: Stereo Amp-Out
  Amp-Out caps: ofs=0x00, nsteps=0x00, stepsize=0x00, mute=1
  Amp-Out vals: 0x00 0x00
  Pincap 0x083f: IN OUT HP
  Pin Default 0x00000000: Line Out at Ext N/A
    Conn = Unknown, Color = Unknown
  Pin-ctls: 0x40: OUT
  Connection: 1
     0x0c
Node 0x15 [Pin Complex] wcaps 0x40018d: Stereo Amp-Out
  Amp-Out caps: ofs=0x00, nsteps=0x00, stepsize=0x00, mute=1
  Amp-Out vals: 0x80 0x80
  Pincap 0x083f: IN OUT HP
  Pin Default 0x00000000: Line Out at Ext N/A
    Conn = Unknown, Color = Unknown
  Pin-ctls: 0x40: OUT
  Connection: 1
     0x0d
Node 0x16 [Pin Complex] wcaps 0x40018d: Stereo Amp-Out
  Amp-Out caps: ofs=0x00, nsteps=0x00, stepsize=0x00, mute=1
  Amp-Out vals: 0x80 0x80
  Pincap 0x083f: IN OUT HP
  Pin Default 0x00000000: Line Out at Ext N/A
    Conn = Unknown, Color = Unknown
  Pin-ctls: 0x40: OUT
  Connection: 1
     0x0e
Node 0x17 [Pin Complex] wcaps 0x40018d: Stereo Amp-Out
  Amp-Out caps: ofs=0x00, nsteps=0x00, stepsize=0x00, mute=1
  Amp-Out vals: 0x80 0x80
  Pincap 0x083f: IN OUT HP
  Pin Default 0x00000000: Line Out at Ext N/A
    Conn = Unknown, Color = Unknown
  Pin-ctls: 0x40: OUT
  Connection: 1
     0x0f
Node 0x18 [Pin Complex] wcaps 0x40018d: Stereo Amp-Out
  Amp-Out caps: ofs=0x00, nsteps=0x00, stepsize=0x00, mute=1
  Amp-Out vals: 0x00 0x00
  Pincap 0x08133f: IN OUT HP
  Pin Default 0x00000000: Line Out at Ext N/A
    Conn = Unknown, Color = Unknown
  Pin-ctls: 0x24: IN
  Connection: 1
     0x10
Node 0x19 [Pin Complex] wcaps 0x40018d: Stereo Amp-Out
  Amp-Out caps: ofs=0x00, nsteps=0x00, stepsize=0x00, mute=1
  Amp-Out vals: 0x00 0x00
  Pincap 0x08133f: IN OUT HP
  Pin Default 0x00000000: Line Out at Ext N/A
    Conn = Unknown, Color = Unknown
  Pin-ctls: 0x40: OUT
  Connection: 1
     0x11
Node 0x1a [Pin Complex] wcaps 0x40018d: Stereo Amp-Out
  Amp-Out caps: ofs=0x00, nsteps=0x00, stepsize=0x00, mute=1
  Amp-Out vals: 0x00 0x00
  Pincap 0x08133f: IN OUT HP
  Pin Default 0x00000000: Line Out at Ext N/A
    Conn = Unknown, Color = Unknown
  Pin-ctls: 0x20: IN
  Connection: 1
     0x12
Node 0x1b [Pin Complex] wcaps 0x40018d: Stereo Amp-Out
  Amp-Out caps: ofs=0x00, nsteps=0x00, stepsize=0x00, mute=1
  Amp-Out vals: 0x80 0x80
  Pincap 0x08133f: IN OUT HP
  Pin Default 0x00000000: Line Out at Ext N/A
    Conn = Unknown, Color = Unknown
  Pin-ctls: 0x24: IN
  Connection: 1
     0x13
Node 0x1c [Pin Complex] wcaps 0x400001: Stereo
  Pincap 0x0820: IN
  Pin Default 0x00000000: Line Out at Ext N/A
    Conn = Unknown, Color = Unknown
  Pin-ctls: 0x00:
Node 0x1d [Pin Complex] wcaps 0x400000: Mono
  Pincap 0x0820: IN
  Pin Default 0x00000000: Line Out at Ext N/A
    Conn = Unknown, Color = Unknown
  Pin-ctls: 0x00:
Node 0x1e [Pin Complex] wcaps 0x400300: Mono Digital
  Pincap 0x0810: OUT
  Pin Default 0x00000000: Line Out at Ext N/A
    Conn = Unknown, Color = Unknown
  Pin-ctls: 0x00:
  Connection: 1
     0x06
Node 0x1f [Pin Complex] wcaps 0x400200: Mono Digital
  Pincap 0x0820: IN
  Pin Default 0x00000000: Line Out at Ext N/A
    Conn = Unknown, Color = Unknown
  Pin-ctls: 0x00:
Node 0x20 [Vendor Defined Widget] wcaps 0xf00040: Mono
Node 0x21 [Volume Knob Widget] wcaps 0x600080: Mono

Re: Intel Azalia problems (some progress)

[Josh Green]

I finally got headphone output with my Azalia sound interface.  I ended
up trying all the models listed in the patch_realtek.c file until
finally "model=w810" worked and I now get headphone output.

Useful info from lspci:

lspci
0000:00:1b.0 Class 0403: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6
Family) High Definition Audio Controller (rev 04)

lspci -n
0000:00:1b.0 Class 0403: 8086:2668 (rev 04)


I imagine this probably isn't exactly the right model.  The "Surround"
control affects the headphone volume but not the "Front" output (it
seems there is no volume level for the "Headphone" control for the w810
model).  The "Front" volume works as expected (internal speakers).  I
have not tested it any more than that.

Still clicking in output, seemingly even with 48KHz audio, still
investigating.  Also many programs lock up at the end of the audio
output (like Rhythmbox when a song is finished for instance).

I'd like to look into getting proper support for this configuration, any
tips on what kind of info to try and obtain?  Cheers!
	Josh Green




-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click

Re: Intel Azalia problems (some progress)

[Takashi Iwai]

At Fri, 15 Apr 2005 05:12:59 -0700,
Josh Green wrote:
> 
> I finally got headphone output with my Azalia sound interface.  I ended
> up trying all the models listed in the patch_realtek.c file until
> finally "model=w810" worked and I now get headphone output.

It's good to know :)

> Useful info from lspci:
> 
> lspci
> 0000:00:1b.0 Class 0403: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6
> Family) High Definition Audio Controller (rev 04)
> 
> lspci -n
> 0000:00:1b.0 Class 0403: 8086:2668 (rev 04)
> 
> 
> I imagine this probably isn't exactly the right model.  The "Surround"
> control affects the headphone volume but not the "Front" output (it
> seems there is no volume level for the "Headphone" control for the w810
> model).  The "Front" volume works as expected (internal speakers).  I
> have not tested it any more than that.

That means the pin 0x16 is connected to the headphoone output.
Note that ALC880 has no overall playback controls for all channels.
The volume of each stereo channels is controlled independently.
(That's why there is no "Master" control there.)

Does the device have surround and center-lfe outputs?  Are they shared
with line-in and mic-in?
Also, does "Front" volume affect the internal speaker volume?


> Still clicking in output, seemingly even with 48KHz audio, still
> investigating.  Also many programs lock up at the end of the audio
> output (like Rhythmbox when a song is finished for instance).

This is an unknown problem, so far.
We got a similar problem but it was solved by setting TCREG bits 0-2.
Possibly a similar mis-configuration by broken BIOS.


Takashi


-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click

Re: Intel Azalia problems (some progress)

[Josh Green]

On Fri, 2005-04-15 at 14:41 +0200, Takashi Iwai wrote:
> > 
> > I imagine this probably isn't exactly the right model.  The "Surround"
> > control affects the headphone volume but not the "Front" output (it
> > seems there is no volume level for the "Headphone" control for the w810
> > model).  The "Front" volume works as expected (internal speakers).  I
> > have not tested it any more than that.
> 
> That means the pin 0x16 is connected to the headphoone output.
> Note that ALC880 has no overall playback controls for all channels.
> The volume of each stereo channels is controlled independently.
> (That's why there is no "Master" control there.)
> 
> Does the device have surround and center-lfe outputs?  Are they shared
> with line-in and mic-in?
> Also, does "Front" volume affect the internal speaker volume?
> 

It basically has two jacks, one labeled S/PDIF which is S/PDIF and
headphone out combined, and the other seems to be mic input.  There is
also a built in mic at the top of the LCD screen.

I just figured out that this notebook is based on the Z71V ASUS platform
which should help a lot.  Yes, "Front" volume works as expected and
controls internal speaker volume.

> 
> > Still clicking in output, seemingly even with 48KHz audio, still
> > investigating.  Also many programs lock up at the end of the audio
> > output (like Rhythmbox when a song is finished for instance).
> 
> This is an unknown problem, so far.
> We got a similar problem but it was solved by setting TCREG bits 0-2.
> Possibly a similar mis-configuration by broken BIOS.
> 

Anything I can try?  I'll poke around in the source for the register you
mentioned.

> 
> Takashi
> 

Cheers!
	Josh Green




-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click

Re: Intel Azalia problems (some progress)

[Josh Green]

On Fri, 2005-04-15 at 14:41 +0200, Takashi Iwai wrote:
> > Still clicking in output, seemingly even with 48KHz audio, still
> > investigating.  Also many programs lock up at the end of the audio
> > output (like Rhythmbox when a song is finished for instance).
> 
> This is an unknown problem, so far.
> We got a similar problem but it was solved by setting TCREG bits 0-2.
> Possibly a similar mis-configuration by broken BIOS.
> 

An additional note about the clicking.  It seems that some programs work
fine and others don't.  XMMS using OSS API will play 48KHz or 44KHz
files without clicks, while sox play command will click with 44KHz but
works fine with 48KHz.

I think the difference may be that XMMS mmaps the device while sox does
not, here is some strace output from the two programs:

XMMS (works fine with both 44.1 or 48KHz)

[pid 30072] open("/dev/sound/dsp", O_WRONLY) = 9
[pid 30072] ioctl(9, SNDCTL_DSP_RESET, 0) = 0
[pid 30072] ioctl(9, SNDCTL_DSP_SETFRAGMENT, 0xbffff1dc) = 0
[pid 30072] ioctl(9, SNDCTL_DSP_SETFMT or SOUND_PCM_READ_BITS, 0xb7280428) = 0
[pid 30072] ioctl(9, SNDCTL_DSP_SETFMT or SOUND_PCM_READ_BITS, 0xb7280428) = 0
[pid 30072] ioctl(9, SNDCTL_DSP_STEREO, 0xbffff1d8) = 0
[pid 30072] ioctl(9, SNDCTL_DSP_SPEED or SOUND_PCM_READ_RATE, 0xb728042c) = 0
[pid 30072] ioctl(9, SNDCTL_DSP_GETBLKSIZE, 0xb7280400) = 0
[pid 30072] select(10, NULL, [9], NULL, {0, 50000}) = 1 (out [9], left {0, 50000})
[pid 30072] sched_getscheduler(0)       = 0
[pid 30072] mmap2(NULL, 667648, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0xb642c000
[pid 30072] mmap2(NULL, 8392704, PROT_READ|PROT_WRITE|PROT_EXEC, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0xb5c2b000

Sox (clicks with 44.1KHz, fine with 48KHz)

[pid 30040] open("/dev/dsp", O_WRONLY|O_CREAT|O_TRUNC, 0666) = 4
[pid 30040] fstat64(4, {st_mode=S_IFCHR|0660, st_rdev=makedev(14, 3), ...}) = 0
[pid 30040] ioctl(4, SNDCTL_TMR_TIMEBASE or TCGETS, 0xbffff49c) = -1 EINVAL (Invalid argument)
[pid 30040] mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0xb7caa000
[pid 30040] fstat64(4, {st_mode=S_IFCHR|0660, st_rdev=makedev(14, 3), ...}) = 0
[pid 30040] ioctl(4, SNDCTL_DSP_RESET, 0) = 0
[pid 30040] ioctl(4, SNDCTL_DSP_GETFMTS, 0xbffff5e4) = 0
[pid 30040] ioctl(4, SNDCTL_DSP_SETFMT or SOUND_PCM_READ_BITS, 0xbffff5e4) = 0
[pid 30040] ioctl(4, SNDCTL_DSP_STEREO, 0xbffff5e4) = 0
[pid 30040] ioctl(4, SNDCTL_DSP_SPEED or SOUND_PCM_READ_RATE, 0xbffff5e4) = 0
[pid 30040] ioctl(4, SNDCTL_DSP_GETBLKSIZE, 0x80a17d4) = 0
[pid 30040] ioctl(4, SNDCTL_DSP_SYNC, 0) = 0
[pid 30040] munmap(0xb7caa000, 4096)    = 0


aplay on the other hand is fine with 48KHz wav files, but 44KHz mono
files click badly (and lock up aplay at the end, requiring a kill -9),
44KHz STEREO on the other hand works fine.  Strange..

I sent ASUS a technical documentation query, maybe I'll get a response.

> 
> Takashi
> 

Cheers.
	Josh Green




-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click

Re: Intel Azalia problems (some progress)

[Takashi Iwai]

At Fri, 15 Apr 2005 15:15:15 -0700,
Josh Green wrote:
> 
> On Fri, 2005-04-15 at 14:41 +0200, Takashi Iwai wrote:
> > > Still clicking in output, seemingly even with 48KHz audio, still
> > > investigating.  Also many programs lock up at the end of the audio
> > > output (like Rhythmbox when a song is finished for instance).
> > 
> > This is an unknown problem, so far.
> > We got a similar problem but it was solved by setting TCREG bits 0-2.
> > Possibly a similar mis-configuration by broken BIOS.
> > 
> 
> An additional note about the clicking.  It seems that some programs work
> fine and others don't.  XMMS using OSS API will play 48KHz or 44KHz
> files without clicks, while sox play command will click with 44KHz but
> works fine with 48KHz.
> 
> I think the difference may be that XMMS mmaps the device while sox does
> not, here is some strace output from the two programs:

Or, it may depend on the period/buffer size.
Could you try alsa-lib CVS version, BTW?


> aplay on the other hand is fine with 48KHz wav files, but 44KHz mono
> files click badly (and lock up aplay at the end, requiring a kill -9),
> 44KHz STEREO on the other hand works fine.  Strange..

Hmm, HDA codecs (at least ALSA drivers) don't support mono, so it must
be converted in alsa-lib thus identical with stereo...


Takashi


-------------------------------------------------------
This SF.Net email is sponsored by: New Crystal Reports XI.
Version 11 adds new functionality designed to reduce time involved in
creating, integrating, and deploying reporting solutions. Free runtime info,
new features, or free trial, at: http://www.businessobjects.com/devxi/728

Re: Intel Azalia problems (some progress)

[Takashi Iwai]

At Tue, 19 Apr 2005 14:11:17 +0200,
I wrote:
> 
> At Fri, 15 Apr 2005 15:15:15 -0700,
> Josh Green wrote:
> > 
> > On Fri, 2005-04-15 at 14:41 +0200, Takashi Iwai wrote:
> > > > Still clicking in output, seemingly even with 48KHz audio, still
> > > > investigating.  Also many programs lock up at the end of the audio
> > > > output (like Rhythmbox when a song is finished for instance).
> > > 
> > > This is an unknown problem, so far.
> > > We got a similar problem but it was solved by setting TCREG bits 0-2.
> > > Possibly a similar mis-configuration by broken BIOS.
> > > 
> > 
> > An additional note about the clicking.  It seems that some programs work
> > fine and others don't.  XMMS using OSS API will play 48KHz or 44KHz
> > files without clicks, while sox play command will click with 44KHz but
> > works fine with 48KHz.
> > 
> > I think the difference may be that XMMS mmaps the device while sox does
> > not, here is some strace output from the two programs:
> 
> Or, it may depend on the period/buffer size.
> Could you try alsa-lib CVS version, BTW?
> 
> 
> > aplay on the other hand is fine with 48KHz wav files, but 44KHz mono
> > files click badly (and lock up aplay at the end, requiring a kill -9),
> > 44KHz STEREO on the other hand works fine.  Strange..
> 
> Hmm, HDA codecs (at least ALSA drivers) don't support mono, so it must
> be converted in alsa-lib thus identical with stereo...

Another try:  if "IEC958 Playback Switch" is on, turn it off.


Takashi


-------------------------------------------------------
This SF.Net email is sponsored by: New Crystal Reports XI.
Version 11 adds new functionality designed to reduce time involved in
creating, integrating, and deploying reporting solutions. Free runtime info,
new features, or free trial, at: http://www.businessobjects.com/devxi/728

Re: Intel Azalia problems (some progress)

[Takashi Iwai]

[-- Attachment #1: Type: text/plain, Size: 1050 bytes --]

At Fri, 15 Apr 2005 12:49:53 -0700,
Josh Green wrote:
> 
> It basically has two jacks, one labeled S/PDIF which is S/PDIF and
> headphone out combined, and the other seems to be mic input.  There is
> also a built in mic at the top of the LCD screen.
> 
> I just figured out that this notebook is based on the Z71V ASUS platform
> which should help a lot.  Yes, "Front" volume works as expected and
> controls internal speaker volume.

OK, the below is an experimental (untested) patch to add Z71V-specific
configuration.  Pass model=z71v.  To set this model default, define
the PCI subsystem vendor/device IDs in alc880_cfg_tbl list (you can
find the string XXX there).  Note that it's "subsystem" ID, which is
shown via lspci -nv.

I don't know how the headphone and SPDIF are shared.  At least, ALC880
codec itself has no such mechanism, so it must be implemented in the
device itself.  Possibly, turning on/off the SPDIF pin/amp control
(and the corresponding HP amp control) does the job, and this might be
the reason of click noises.


Takashi

[-- Attachment #2: Type: text/plain, Size: 6242 bytes --]

Index: alsa-kernel/pci/hda/patch_realtek.c
===================================================================
RCS file: /home/iwai/cvs/alsa/alsa-kernel/pci/hda/patch_realtek.c,v
retrieving revision 1.8
diff -u -r1.8 patch_realtek.c
--- alsa-kernel/pci/hda/patch_realtek.c	22 Mar 2005 20:51:50 -0000	1.8
+++ alsa-kernel/pci/hda/patch_realtek.c	19 Apr 2005 12:45:51 -0000
@@ -39,6 +39,7 @@
 	ALC880_5ST,
 	ALC880_5ST_DIG,
 	ALC880_W810,
+	ALC880_Z71V,
 };
 
 struct alc_spec {
@@ -90,6 +91,11 @@
 	0x02, 0x03, 0x04
 };
 
+static hda_nid_t alc880_z71v_dac_nids[1] = {
+	/* front only? */
+	0x02
+};
+
 static hda_nid_t alc880_adc_nids[3] = {
 	/* ADC0-2 */
 	0x07, 0x08, 0x09,
@@ -284,6 +290,10 @@
 	{ 6, NULL }
 };
 
+static struct alc_channel_mode alc880_z71v_modes[1] = {
+	{ 2, NULL }
+};
+
 /*
  */
 static int alc880_ch_mode_info(snd_kcontrol_t *kcontrol, snd_ctl_elem_info_t *uinfo)
@@ -475,6 +485,31 @@
 	{ } /* end */
 };
 
+static snd_kcontrol_new_t alc880_z71v_mixer[] = {
+	HDA_CODEC_VOLUME("Front Playback Volume", 0x0c, 0x0, HDA_OUTPUT),
+	HDA_CODEC_MUTE("Front Playback Switch", 0x14, 0x0, HDA_OUTPUT),
+	HDA_CODEC_VOLUME("Headphone Playback Volume", 0x0d, 0x0, HDA_OUTPUT),
+	HDA_CODEC_MUTE("Headphone Playback Switch", 0x15, 0x0, HDA_OUTPUT),
+	HDA_CODEC_VOLUME("Capture Volume", 0x07, 0x0, HDA_INPUT),
+	HDA_CODEC_MUTE("Capture Switch", 0x07, 0x0, HDA_INPUT),
+	HDA_CODEC_VOLUME_IDX("Capture Volume", 1, 0x08, 0x0, HDA_INPUT),
+	HDA_CODEC_MUTE_IDX("Capture Switch", 1, 0x08, 0x0, HDA_INPUT),
+	{
+		.iface = SNDRV_CTL_ELEM_IFACE_MIXER,
+		/* The multiple "Capture Source" controls confuse alsamixer
+		 * So call somewhat different..
+		 * FIXME: the controls appear in the "playback" view!
+		 */
+		/* .name = "Capture Source", */
+		.name = "Input Source",
+		.count = 2,
+		.info = alc_mux_enum_info,
+		.get = alc_mux_enum_get,
+		.put = alc_mux_enum_put,
+	},
+	{ } /* end */
+};
+
 /*
  */
 static int alc_build_controls(struct hda_codec *codec)
@@ -719,6 +754,58 @@
 	{ }
 };
 
+static struct hda_verb alc880_z71v_init_verbs[] = {
+	/* front channel selector/amp: input 0: DAC: unmuted, (no volume selection) */
+	{0x0c, AC_VERB_SET_AMP_GAIN_MUTE, 0x7000},
+	/* front channel selector/amp: input 1: capture mix: muted, (no volume selection) */
+	{0x0c, AC_VERB_SET_AMP_GAIN_MUTE, 0x7180},
+	/* front channel selector/amp: output 0: unmuted, max volume */
+	{0x0c, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000},
+	/* front out pin: muted, (no volume selection)  */
+	{0x14, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080},
+	/* front out pin: NOT headphone enable, out enable, vref disabled */
+	{0x14, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40},
+	/* headphone channel selector/amp: input 0: DAC: unmuted, (no volume selection) */
+	{0x0d, AC_VERB_SET_AMP_GAIN_MUTE, 0x7000},
+	/* headphone channel selector/amp: input 1: capture mix: muted, (no volume selection) */
+	{0x0d, AC_VERB_SET_AMP_GAIN_MUTE, 0x7180},
+	/* headphone channel selector/amp: output 0: unmuted, max volume */
+	{0x0d, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000},
+	/* headphone out pin: muted, (no volume selection)  */
+	{0x15, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080},
+	/* headpohne out pin: headphone enable, out enable, vref disabled */
+	{0x15, AC_VERB_SET_PIN_WIDGET_CONTROL, 0xc0},
+
+	/* Line In pin widget for input */
+	{0x1a, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x20},
+	/* CD pin widget for input */
+	{0x1c, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x20},
+	/* Mic1 (rear panel) pin widget for input and vref at 80% */
+	{0x18, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x24},
+	/* Mic2 (front panel) pin widget for input and vref at 80% */
+	{0x1b, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x24},
+	/* unmute amp left and right */
+	{0x07, AC_VERB_SET_AMP_GAIN_MUTE, 0x7000},
+	/* set connection select to line in (default select for this ADC) */
+	{0x07, AC_VERB_SET_CONNECT_SEL, 0x02},
+
+	/* Unmute input amps (CD, Line In, Mic 1 & Mic 2) for mixer
+	 * widget(nid=0x0B) to support the input path of analog loopback
+	 */
+	/* Note: PASD motherboards uses the Line In 2 as the input for front panel mic (mic 2) */
+	/* Amp Indexes: CD = 0x04, Line In 1 = 0x02, Mic 1 = 0x00 & Line In 2 = 0x03*/
+	/* unmute CD */
+	{0x0b, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x04 << 8))},
+	/* unmute Line In */
+	{0x0b, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x02 << 8))},
+	/* unmute Mic 1 */
+	{0x0b, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x00 << 8))},
+	/* unmute Line In 2 (for PASD boards Mic 2) */
+	{0x0b, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x03 << 8))},
+
+	{ }
+};
+
 static int alc_init(struct hda_codec *codec)
 {
 	struct alc_spec *spec = codec->spec;
@@ -993,6 +1080,9 @@
 	{ .modelname = "w810", .config = ALC880_W810 },
 	{ .pci_vendor = 0x161f, .pci_device = 0x203d, .config = ALC880_W810 },
 
+	{ .modelname = "z71v", .config = ALC880_Z71V },
+	// { .pci_vendor = XXXX, .pci_device = XXXX, .config = ALC880_Z71V },
+
 	{}
 };
 
@@ -1023,6 +1113,10 @@
 		spec->mixers[spec->num_mixers] = alc880_five_stack_mixer;
 		spec->num_mixers++;
 		break;
+	case ALC880_Z71V:
+		spec->mixers[spec->num_mixers] = alc880_z71v_mixer;
+		spec->num_mixers++;
+		break;
 	default:
 		spec->mixers[spec->num_mixers] = alc880_base_mixer;
 		spec->num_mixers++;
@@ -1033,6 +1127,7 @@
 	case ALC880_3ST_DIG:
 	case ALC880_5ST_DIG:
 	case ALC880_W810:
+	case ALC880_Z71V:
 		spec->multiout.dig_out_nid = ALC880_DIGOUT_NID;
 		break;
 	default:
@@ -1063,6 +1158,11 @@
 		spec->channel_mode = alc880_w810_modes;
 		spec->num_channel_mode = ARRAY_SIZE(alc880_w810_modes);
 		break;
+	case ALC880_Z71V:
+		spec->init_verbs = alc880_z71v_init_verbs;
+		spec->channel_mode = alc880_z71v_modes;
+		spec->num_channel_mode = ARRAY_SIZE(alc880_z71v_modes);
+		break;
 	default:
 		spec->init_verbs = alc880_init_verbs_three_stack;
 		spec->channel_mode = alc880_threestack_modes;
@@ -1086,6 +1186,11 @@
 		spec->multiout.dac_nids = alc880_w810_dac_nids;
 		// No dedicated headphone socket - it's shared with built-in speakers.
 		break;
+	case ALC880_Z71V:
+		spec->multiout.num_dacs = ARRAY_SIZE(alc880_z71v_dac_nids);
+		spec->multiout.dac_nids = alc880_z71v_dac_nids;
+		spec->multiout.hp_nid = 0x03;
+		break;
 	default:
 		spec->multiout.num_dacs = ARRAY_SIZE(alc880_dac_nids);
 		spec->multiout.dac_nids = alc880_dac_nids;