Support for ALC662 on FUJITSU SIEMENS AMILO Li3710

Support for ALC662 on FUJITSU SIEMENS AMILO Li3710

[Mads Kiilerich]

On a friends FUJITSU SIEMENS AMILO Li3710 10601011427 sound didn't work 
with kernel-PAE-2.6.29.6-213.fc11.i686

Out of the box we get no sound and
HDA Intel 0000:00:1b.0: PCI INT A -> GSI 22 (level, low) -> IRQ 22
HDA Intel 0000:00:1b.0: setting latency timer to 64
hda_codec: Unknown model for ALC662, trying auto-probe from BIOS...
ALSA sound/pci/hda/hda_codec.c:3507: autoconfig: line_outs=1 
(0x14/0x0/0x0/0x0/0x0)
ALSA sound/pci/hda/hda_codec.c:3511: speaker_outs=0 (0x0/0x0/0x0/0x0/0x0)
ALSA sound/pci/hda/hda_codec.c:3515: hp_outs=1 (0x1b/0x0/0x0/0x0/0x0)
ALSA sound/pci/hda/hda_codec.c:3516: mono: mono_out=0x0
ALSA sound/pci/hda/hda_codec.c:3524: inputs: mic=0x18, fmic=0x19, 
line=0x1a, fline=0x0, cd=0x0, aux=0x0

But with /etc/modprobe.d/local.conf (from 
http://ubuntuforums.org/archive/index.php/t-940689.html)
options snd-hda-intel model=3stack-dig
options snd-hda-intel single_cmd=1
options snd-hda-intel enable_msi=1

we get sound and
HDA Intel 0000:00:1b.0: PCI INT A -> GSI 22 (level, low) -> IRQ 22
HDA Intel 0000:00:1b.0: irq 32 for MSI/MSI-X
HDA Intel 0000:00:1b.0: setting latency timer to 64
ALSA sound/pci/hda/hda_codec.c:2698: hda_codec: model '3stack-dig' is 
selected


I would like to have this fixed so Linux sound works on my friends 
machine without manual configuration.

After having read some documentation I assume that single_cmd and 
enable_msi perhaps shouldn't be needed. I will test that next time I get 
access to that machine.

HAL reports the sound card as 
/org/freedesktop/Hal/devices/pci_8086_293e, so should this patch be 
sufficient?

diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c
index e1563f4..0385353 100644
--- a/sound/pci/hda/patch_realtek.c
+++ b/sound/pci/hda/patch_realtek.c
@@ -16699,6 +16699,8 @@ static struct snd_pci_quirk alc662_cfg_tbl[] = {
                                         ALC662_3ST_6ch_DIG),
         SND_PCI_QUIRK_MASK(0x1854, 0xf000, 0x2000, "ASUS H13-200x",
                            ALC663_ASUS_H13),
+       SND_PCI_QUIRK(0x8086, 0x293e, "FUJITSU SIEMENS AMILO Li3710",
+                      ALC662_3ST_2ch_DIG),
         {}
  };

Untested and hard to test because I don't have direct access to the machine.

This issue was also reported on 
https://bugzilla.redhat.com/show_bug.cgi?id=487889

Please CC me.

/Mads

Re: Support for ALC662 on FUJITSU SIEMENS AMILO Li3710

[Takashi Iwai]

At Sun, 26 Jul 2009 00:54:24 +0200,
Mads Kiilerich wrote:
> 
> On a friends FUJITSU SIEMENS AMILO Li3710 10601011427 sound didn't work 
> with kernel-PAE-2.6.29.6-213.fc11.i686

How about with the latest alsa-driver snapshot?
    ftp://ftp.kernel.org/pub/linux/kernel/people/tiwai/snapshot/alsa-driver-snapshot.tar.gz

Also, please give alsa-info.sh output.  Run with --no-upload and attach
the generated file.


thanks,

Takashi



> 
> Out of the box we get no sound and
> HDA Intel 0000:00:1b.0: PCI INT A -> GSI 22 (level, low) -> IRQ 22
> HDA Intel 0000:00:1b.0: setting latency timer to 64
> hda_codec: Unknown model for ALC662, trying auto-probe from BIOS...
> ALSA sound/pci/hda/hda_codec.c:3507: autoconfig: line_outs=1 
> (0x14/0x0/0x0/0x0/0x0)
> ALSA sound/pci/hda/hda_codec.c:3511: speaker_outs=0 (0x0/0x0/0x0/0x0/0x0)
> ALSA sound/pci/hda/hda_codec.c:3515: hp_outs=1 (0x1b/0x0/0x0/0x0/0x0)
> ALSA sound/pci/hda/hda_codec.c:3516: mono: mono_out=0x0
> ALSA sound/pci/hda/hda_codec.c:3524: inputs: mic=0x18, fmic=0x19, 
> line=0x1a, fline=0x0, cd=0x0, aux=0x0
> 
> But with /etc/modprobe.d/local.conf (from 
> http://ubuntuforums.org/archive/index.php/t-940689.html)
> options snd-hda-intel model=3stack-dig
> options snd-hda-intel single_cmd=1
> options snd-hda-intel enable_msi=1
> 
> we get sound and
> HDA Intel 0000:00:1b.0: PCI INT A -> GSI 22 (level, low) -> IRQ 22
> HDA Intel 0000:00:1b.0: irq 32 for MSI/MSI-X
> HDA Intel 0000:00:1b.0: setting latency timer to 64
> ALSA sound/pci/hda/hda_codec.c:2698: hda_codec: model '3stack-dig' is 
> selected
> 
> 
> I would like to have this fixed so Linux sound works on my friends 
> machine without manual configuration.
> 
> After having read some documentation I assume that single_cmd and 
> enable_msi perhaps shouldn't be needed. I will test that next time I get 
> access to that machine.
> 
> HAL reports the sound card as 
> /org/freedesktop/Hal/devices/pci_8086_293e, so should this patch be 
> sufficient?
> 
> diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c
> index e1563f4..0385353 100644
> --- a/sound/pci/hda/patch_realtek.c
> +++ b/sound/pci/hda/patch_realtek.c
> @@ -16699,6 +16699,8 @@ static struct snd_pci_quirk alc662_cfg_tbl[] = {
>                                          ALC662_3ST_6ch_DIG),
>          SND_PCI_QUIRK_MASK(0x1854, 0xf000, 0x2000, "ASUS H13-200x",
>                             ALC663_ASUS_H13),
> +       SND_PCI_QUIRK(0x8086, 0x293e, "FUJITSU SIEMENS AMILO Li3710",
> +                      ALC662_3ST_2ch_DIG),
>          {}
>   };
> 
> Untested and hard to test because I don't have direct access to the machine.
> 
> This issue was also reported on 
> https://bugzilla.redhat.com/show_bug.cgi?id=487889
> 
> Please CC me.
> 
> /Mads
> _______________________________________________
> Alsa-devel mailing list
> Alsa-devel@alsa-project.org
> http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
> 

Re: Support for ALC662 on FUJITSU SIEMENS AMILO Li3710

[Mads Kiilerich]

Takashi Iwai wrote, On 07/26/2009 10:50 AM:
> At Sun, 26 Jul 2009 00:54:24 +0200,
> Mads Kiilerich wrote:
>    
>> On a friends FUJITSU SIEMENS AMILO Li3710 10601011427 sound didn't work
>> with kernel-PAE-2.6.29.6-213.fc11.i686
>>      
> How about with the latest alsa-driver snapshot?
>      ftp://ftp.kernel.org/pub/linux/kernel/people/tiwai/snapshot/alsa-driver-snapshot.tar.gz
>    

Thanks.

I looked at your sound-2.6.git tree and found no quirk for this card 
(AFAICS) and assumed that it hadn't been reported and fixed yet. I 
assume that that tree is pretty close to alsa-driver-snapshot?

I am a developer, but for alsa I am a user, no expert, and honestly I 
don't plan to become one. Can you give a pointer to a description of how 
can I test the snapshot? On Fedora 11?

(http://www.alsa-project.org/main/index.php/Main_Page has a "How do I 
install the ALSA driver from source" link, but it just points to the 
download page without any explanation.)

> Also, please give alsa-info.sh output.  Run with --no-upload and attach
> the generated file.
>    

I became aware of that command after having visited my friend. I have 
output of dmesg and  lshal and lspci - is that sufficient?

/Mads

Re: Support for ALC662 on FUJITSU SIEMENS AMILO Li3710

[Takashi Iwai]

At Sun, 26 Jul 2009 19:39:28 +0200,
Mads Kiilerich wrote:
> 
> Takashi Iwai wrote, On 07/26/2009 10:50 AM:
> > At Sun, 26 Jul 2009 00:54:24 +0200,
> > Mads Kiilerich wrote:
> >    
> >> On a friends FUJITSU SIEMENS AMILO Li3710 10601011427 sound didn't work
> >> with kernel-PAE-2.6.29.6-213.fc11.i686
> >>      
> > How about with the latest alsa-driver snapshot?
> >      ftp://ftp.kernel.org/pub/linux/kernel/people/tiwai/snapshot/alsa-driver-snapshot.tar.gz
> >    
> 
> Thanks.
> 
> I looked at your sound-2.6.git tree and found no quirk for this card 
> (AFAICS) and assumed that it hadn't been reported and fixed yet.

When no quirk is set, the automatic mode is used as default.
And, it's possible that the automatic mode has been fixed in the
upstream.  That's why I asked to test with the latest one.
We should avoid any device-specific quirk as much as possible, but
rather fix the automatic parser.

> I 
> assume that that tree is pretty close to alsa-driver-snapshot?

Yes.  The alsa-driver-snapshot tarball is created from GIT tree.

> I am a developer, but for alsa I am a user, no expert, and honestly I 
> don't plan to become one. Can you give a pointer to a description of how 
> can I test the snapshot? On Fedora 11?
> 
> (http://www.alsa-project.org/main/index.php/Main_Page has a "How do I 
> install the ALSA driver from source" link, but it just points to the 
> download page without any explanation.)

Did you take a look at INSTALL file in alsa-driver?
You can try sound git tree instead of that tarball, too, if you are
building your kernel from a GIT tree by yourself.

> > Also, please give alsa-info.sh output.  Run with --no-upload and attach
> > the generated file.
> >    
> 
> I became aware of that command after having visited my friend. I have 
> output of dmesg and  lshal and lspci - is that sufficient?

No.


thanks,

Takashi

Re: Support for ALC662 on FUJITSU SIEMENS AMILO Li3710

[Mads Kiilerich]

Takashi Iwai wrote, On 07/27/2009 08:14 AM:
> At Sun, 26 Jul 2009 19:39:28 +0200,
> Mads Kiilerich wrote:
>    
>> Takashi Iwai wrote, On 07/26/2009 10:50 AM:
>>      
>>> At Sun, 26 Jul 2009 00:54:24 +0200,
>>> Mads Kiilerich wrote:
>>>
>>>        
>>>> On a friends FUJITSU SIEMENS AMILO Li3710 10601011427 sound didn't work
>>>> with kernel-PAE-2.6.29.6-213.fc11.i686
>>>>
>>>>          
>>> How about with the latest alsa-driver snapshot?
>>>       ftp://ftp.kernel.org/pub/linux/kernel/people/tiwai/snapshot/alsa-driver-snapshot.tar.gz
>>>
>>>        
>> Thanks.
>>
>> I looked at your sound-2.6.git tree and found no quirk for this card
>> (AFAICS) and assumed that it hadn't been reported and fixed yet.
>>      
> When no quirk is set, the automatic mode is used as default.
> And, it's possible that the automatic mode has been fixed in the
> upstream.  That's why I asked to test with the latest one.
> We should avoid any device-specific quirk as much as possible, but
> rather fix the automatic parser.
>    

Ok, if you think that would be possible. I assumed that it was broken 
beond repair.

>> I am a developer, but for alsa I am a user, no expert, and honestly I
>> don't plan to become one. Can you give a pointer to a description of how
>> can I test the snapshot? On Fedora 11?
>>
>> (http://www.alsa-project.org/main/index.php/Main_Page  has a "How do I
>> install the ALSA driver from source" link, but it just points to the
>> download page without any explanation.)
>>      
> Did you take a look at INSTALL file in alsa-driver?
>    

alsa-driver-snapshot.tar.gz README says "integrated to 2.6+ kernels", so 
I assumed that this somehow only applied to 2.4 - which obviously would 
be very strange.

I missed the big overview. It seems like it would have been helpful to 
me (and the next guy) if something like the following blurb could be 
found somewhere - in README or INSTALL or from the "How do I install the 
ALSA driver from source" link or on 
http://www.alsa-project.org/main/index.php/Driver_Compilation :

"ALSA project develops both Linux kernel drivers as well as support 
libraries and tools for user-space. The kernel drivers are also included 
in vanilla Linux kernels, and usually the stock drivers are fine and 
should be used. It is however also possible to build the alsa-driver 
kernel modules for an existing compiled kernel and replace its drivers, 
for example from alsa-driver-snapshot.tar.gz to get the latest ALSA 
development and fixes. Note that the modules will have to be rebuild 
when a new kernel is installed."


INSTALL is very demotivating for use with distribution kernels. It would 
be helpful if you could add something like:

"On Fedora 11 (with PAE kernels) it works fine, but you need the package 
kernel-PAE-devel of the same version as the running kernel-PAE."

http://www.linlap.com/wiki/configuring+the+audio+and+updating+alsa+for+fedora+10 
might also be helpful.

I assume something similar applies to Suse.

>>> Also, please give alsa-info.sh output.  Run with --no-upload and attach
>>> the generated file.
>>>
>>>        
>> I became aware of that command after having visited my friend. I have
>> output of dmesg and  lshal and lspci - is that sufficient?
>>      
> No

Ok, thanks for the hints to help me understand.

I will try alsa-drivers-snapshot and alsa-info.sh next time I get access 
to that computer - that will probably take more than a month.

/Mads

Re: Support for ALC662 on FUJITSU SIEMENS AMILO Li3710

[Takashi Iwai]

At Mon, 27 Jul 2009 12:04:22 +0200,
Mads Kiilerich wrote:
> 
> Takashi Iwai wrote, On 07/27/2009 08:14 AM:
> > At Sun, 26 Jul 2009 19:39:28 +0200,
> > Mads Kiilerich wrote:
> >    
> >> Takashi Iwai wrote, On 07/26/2009 10:50 AM:
> >>      
> >>> At Sun, 26 Jul 2009 00:54:24 +0200,
> >>> Mads Kiilerich wrote:
> >>>
> >>>        
> >>>> On a friends FUJITSU SIEMENS AMILO Li3710 10601011427 sound didn't work
> >>>> with kernel-PAE-2.6.29.6-213.fc11.i686
> >>>>
> >>>>          
> >>> How about with the latest alsa-driver snapshot?
> >>>       ftp://ftp.kernel.org/pub/linux/kernel/people/tiwai/snapshot/alsa-driver-snapshot.tar.gz
> >>>
> >>>        
> >> Thanks.
> >>
> >> I looked at your sound-2.6.git tree and found no quirk for this card
> >> (AFAICS) and assumed that it hadn't been reported and fixed yet.
> >>      
> > When no quirk is set, the automatic mode is used as default.
> > And, it's possible that the automatic mode has been fixed in the
> > upstream.  That's why I asked to test with the latest one.
> > We should avoid any device-specific quirk as much as possible, but
> > rather fix the automatic parser.
> >    
> 
> Ok, if you think that would be possible. I assumed that it was broken 
> beond repair.
> 
> >> I am a developer, but for alsa I am a user, no expert, and honestly I
> >> don't plan to become one. Can you give a pointer to a description of how
> >> can I test the snapshot? On Fedora 11?
> >>
> >> (http://www.alsa-project.org/main/index.php/Main_Page  has a "How do I
> >> install the ALSA driver from source" link, but it just points to the
> >> download page without any explanation.)
> >>      
> > Did you take a look at INSTALL file in alsa-driver?
> >    
> 
> alsa-driver-snapshot.tar.gz README says "integrated to 2.6+ kernels", so 
> I assumed that this somehow only applied to 2.4 - which obviously would 
> be very strange.

A word "integration" doesn't mean to exclude the update possibility at
all.  Also, when the driver is built as a module, it can be replaced
even dynamically.


> I missed the big overview. It seems like it would have been helpful to 
> me (and the next guy) if something like the following blurb could be 
> found somewhere - in README or INSTALL or from the "How do I install the 
> ALSA driver from source" link or on 
> http://www.alsa-project.org/main/index.php/Driver_Compilation :
> 
> "ALSA project develops both Linux kernel drivers as well as support 
> libraries and tools for user-space. The kernel drivers are also included 
> in vanilla Linux kernels, and usually the stock drivers are fine and 
> should be used. It is however also possible to build the alsa-driver 
> kernel modules for an existing compiled kernel and replace its drivers, 
> for example from alsa-driver-snapshot.tar.gz to get the latest ALSA 
> development and fixes. Note that the modules will have to be rebuild 
> when a new kernel is installed."
> 
> 
> INSTALL is very demotivating for use with distribution kernels. It would 
> be helpful if you could add something like:
> 
> "On Fedora 11 (with PAE kernels) it works fine, but you need the package 
> kernel-PAE-devel of the same version as the running kernel-PAE."
> 
> http://www.linlap.com/wiki/configuring+the+audio+and+updating+alsa+for+fedora+10 
> might also be helpful.
> 
> I assume something similar applies to Suse.

Patches are welcome :)


> >>> Also, please give alsa-info.sh output.  Run with --no-upload and attach
> >>> the generated file.
> >>>
> >>>        
> >> I became aware of that command after having visited my friend. I have
> >> output of dmesg and  lshal and lspci - is that sufficient?
> >>      
> > No
> 
> Ok, thanks for the hints to help me understand.
> 
> I will try alsa-drivers-snapshot and alsa-info.sh next time I get access 
> to that computer - that will probably take more than a month.

OK, let me know any results.


thanks,

Takashi

Re: Support for ALC662 on FUJITSU SIEMENS AMILO Li3710

[Mads Kiilerich]

Takashi Iwai wrote, On 07/27/2009 12:10 PM:
> At Mon, 27 Jul 2009 12:04:22 +0200,
> Mads Kiilerich wrote:
>    
>> Takashi Iwai wrote, On 07/27/2009 08:14 AM:
>>      
>>> At Sun, 26 Jul 2009 19:39:28 +0200,
>>> Mads Kiilerich wrote:
>>>
>>>        
>>>> Takashi Iwai wrote, On 07/26/2009 10:50 AM:
>>>>
>>>>          
>>>>> At Sun, 26 Jul 2009 00:54:24 +0200,
>>>>> Mads Kiilerich wrote:
>>>>>
>>>>>
>>>>>            
>>>>>> On a friends FUJITSU SIEMENS AMILO Li3710 10601011427 sound didn't work
>>>>>> with kernel-PAE-2.6.29.6-213.fc11.i686
>>>>>>
>>>>>>
>>>>>>              
>>>>> How about with the latest alsa-driver snapshot?
>>>>>        ftp://ftp.kernel.org/pub/linux/kernel/people/tiwai/snapshot/alsa-driver-snapshot.tar.gz
>>>>>
>>>>>
>>>>>            
>>>> Thanks.
>>>>
>>>> I looked at your sound-2.6.git tree and found no quirk for this card
>>>> (AFAICS) and assumed that it hadn't been reported and fixed yet.
>>>>
>>>>          
>>> When no quirk is set, the automatic mode is used as default.
>>> And, it's possible that the automatic mode has been fixed in the
>>> upstream.  That's why I asked to test with the latest one.
>>> We should avoid any device-specific quirk as much as possible, but
>>> rather fix the automatic parser.
>>>
>>>        
>> Ok, if you think that would be possible. I assumed that it was broken
>> beond repair.
>>      
...
>>>>> Also, please give alsa-info.sh output.  Run with --no-upload and attach
>>>>> the generated file.
>>>>>
>>>>>
>>>>>            
>>>> I became aware of that command after having visited my friend. I have
>>>> output of dmesg and  lshal and lspci - is that sufficient?
>>>>
>>>>          
>>> No
>>>        
>> Ok, thanks for the hints to help me understand.
>>
>> I will try alsa-drivers-snapshot and alsa-info.sh next time I get access
>> to that computer - that will probably take more than a month.
>>      
> OK, let me know any results.
>    

Ok, I got access to the machine and tested successfully. The ALC662 on 
this machine works with the latest snapshot, so I assume that it 
eventually will start working with vanilla upstream kernels - somewhere 
between 2.6.30 and 2.6.32.

http://www.alsa-project.org/db/?f=c72a5f49d7f21b99173358a421ea9e0f18898810

Thanks!

/Mads

Re: Support for ALC662 on FUJITSU SIEMENS AMILO Li3710

[Takashi Iwai]

At Thu, 20 Aug 2009 00:45:39 +0200,
Mads Kiilerich wrote:
> 
> Takashi Iwai wrote, On 07/27/2009 12:10 PM:
> > At Mon, 27 Jul 2009 12:04:22 +0200,
> > Mads Kiilerich wrote:
> >    
> >> Takashi Iwai wrote, On 07/27/2009 08:14 AM:
> >>      
> >>> At Sun, 26 Jul 2009 19:39:28 +0200,
> >>> Mads Kiilerich wrote:
> >>>
> >>>        
> >>>> Takashi Iwai wrote, On 07/26/2009 10:50 AM:
> >>>>
> >>>>          
> >>>>> At Sun, 26 Jul 2009 00:54:24 +0200,
> >>>>> Mads Kiilerich wrote:
> >>>>>
> >>>>>
> >>>>>            
> >>>>>> On a friends FUJITSU SIEMENS AMILO Li3710 10601011427 sound didn't work
> >>>>>> with kernel-PAE-2.6.29.6-213.fc11.i686
> >>>>>>
> >>>>>>
> >>>>>>              
> >>>>> How about with the latest alsa-driver snapshot?
> >>>>>        ftp://ftp.kernel.org/pub/linux/kernel/people/tiwai/snapshot/alsa-driver-snapshot.tar.gz
> >>>>>
> >>>>>
> >>>>>            
> >>>> Thanks.
> >>>>
> >>>> I looked at your sound-2.6.git tree and found no quirk for this card
> >>>> (AFAICS) and assumed that it hadn't been reported and fixed yet.
> >>>>
> >>>>          
> >>> When no quirk is set, the automatic mode is used as default.
> >>> And, it's possible that the automatic mode has been fixed in the
> >>> upstream.  That's why I asked to test with the latest one.
> >>> We should avoid any device-specific quirk as much as possible, but
> >>> rather fix the automatic parser.
> >>>
> >>>        
> >> Ok, if you think that would be possible. I assumed that it was broken
> >> beond repair.
> >>      
> ...
> >>>>> Also, please give alsa-info.sh output.  Run with --no-upload and attach
> >>>>> the generated file.
> >>>>>
> >>>>>
> >>>>>            
> >>>> I became aware of that command after having visited my friend. I have
> >>>> output of dmesg and  lshal and lspci - is that sufficient?
> >>>>
> >>>>          
> >>> No
> >>>        
> >> Ok, thanks for the hints to help me understand.
> >>
> >> I will try alsa-drivers-snapshot and alsa-info.sh next time I get access
> >> to that computer - that will probably take more than a month.
> >>      
> > OK, let me know any results.
> >    
> 
> Ok, I got access to the machine and tested successfully. The ALC662 on 
> this machine works with the latest snapshot, so I assume that it 
> eventually will start working with vanilla upstream kernels - somewhere 
> between 2.6.30 and 2.6.32.

OK, good to hear!


thanks,

Takashi