Hi to everyone,

I installed an Asus P7P55D PRO motherboard and now I can't mute Master channel with alsamixer, even not using KDE mixer. Probably ALSA doesn't support properly, despite I upgraded alsa libraries (alsa-lib-1.0.21a). The key to increase or decrease the volume through keyboard works fine, but the key for mute is not working.
These are the alsamixer values:

Card: HDA Intel
Chip: VIA ID 444

I also used alsaconf command and it detects a snd-hda-intel board. I have a G15 keyboard and I even tried to make a shortcut with KDE mixer, but it doesn't work. Is there any way to solve it or have I to wait for a new version of alsa?
I use Slackware64 13.0 with kernel 2.6.32.7.

Thanks,

Scorpion77

You could try appending acpi_osi=Linux to the kernel in lilo. That has been known to work on some machines.

If you're still having this problem, please try the bold command below, and paste in here the comparable 6 lines or so from the beginning of the output:

command:
Lines of output:
NOTE: The file *should* be the same on your system; but if you get a file not found, please look in /proc and follow the folders in /asound until you locate the codec# file, and provide the contents using the `cat` command as shown above.

Sasha

This is the output:

Scorpion77

This codec is a VT1828S. The driver doesn't yet know about this chip and therefore cannot display its name.

Try a newer kernel (currently 2.6.33-rc8).

Note: It is entirely possible that this codec does not actually have a master mute control.

Luckily I can control the volume but it's annoying that I can't use mute.
Anyway this audio chip is capable to set the mute, because with Windows 7 is working.

Thank you very much.

Scorpion77

Hi,

definitely try a newer kernel, and if you're feeling daring, grab the ALSA libs from git if you want to try a REALLY new version.
From what I can tell from browsing both the ALSA git tree, and the Linus kernel git tree, is that your audio device is supported in the next kernel and/or alsa releases, using (as cladisch mentioned) the vt1828S codec (which actually appears to point to the already-existing code for the vt1718S codec). It also has mute functionality.

One thing which I cannot figure out though, is whether you might benefit from specifying a particular MODEL to the snd-hda-intel driver when it loads. With my own sound chip, reading the kernel's realtek-patch.c file, I could easily figure out what model= argument would work for my realtek device (I subsequently sent a couple patches to the kernel for my motherboard). But in your case, with a VIA device, it appears to me that the via-patch.c file is constructed differently, making it hard for me to figure out what model= arguments are relevant (if any??).

You might like to have a look at the possibilities for models, here: http://www.kernel.org/doc/Documentat...dio-Models.txt
and maybe try a few yourself to see if any makes a difference. NOTE that this depends on whether the kernel you now run, is new enough (and the alsa libs too) to have this particular codec support in it; something tells me it is not, but I haven't checked the exact date on your kernel vs the git kernel vs your alsa libs vs the git alsa, nor have I checked your exact kernel version's tree for the necessary codec support. Of course, specifying a MODEL will do no good if the codec support is not yet there

ALso NOTE: The /Documentation folder inside your kernel source folder is the most reliable place to read specifics that apply to your exact kernel.


For specifying models when the driver loads, use a file in /etc/modprobe.d/ called alsa.conf or sound.conf or whatever you like; it would contain something like this (I highlighted the important stuff):
Best of luck,
Sasha

This is the content of /etc/modprobe.d/sound:

I think I will wait for a full support of future kernels. I can recompile a kernel, but adding patches is a bit more complicated.

Thank you very much,

Scorpion77

No need to add patches -- the "model=" line will, in theory, accomplish the same thing as the patch, provided the codec information for your sound device is existing already in the kernel and alsa software. I no longer need to use the "model=" line, because since I patched the kernel, my sound device is identified correctly by the kernel during bootup, automatically, and configured properly.

The "model=" line simply tells the sound driver that: even though it does not recognize your sound device completely right away, it is the same as <some other device> that it does know about, so treat it like that one.

So, if you wanted to "test" the support for your sound in upcoming kernel release, download a super-fresh kernel source, like a -33-rc kernel, or a kernel from Git, and just build it as you normally would.

Or.... Just wait till 2.6.33 is officially released and try it then.

Cheers
Sasha