Alsa 1.0.18 Installation Script

[soundcheck]

@Jaciss

Please change your alsa-base file (sudo gedit /etc/modprobe.d/alsa-base) :

snd-hda-intel model=hp

to

snd-hda-intel index=0 model=gateway

and reboot

If this won't work, please run and attach the log of alsa-info.sh


Cheers

[druellan]

Update: still can't make the internal mic work using mode=basic or mode=quanta. On mode=auto works really well, so I think that perhaps its a matter of found the right combination of controls, perhaps a bad labeled mic.

[Jaciss]

@soundcheck:
Well, alsa-info to the rescue. The file shows:

snd-hda-intel: index=0 model=3 stack position_fix=0 single_cmd=0

as well as the correct:
snd-hda-intel: index=0 model=3stack position_fix=0 single_cmd=0

under modprobe options, which would explain the 'stack' error. So it amounts to a typo. What is amusing is that said typo was in a backup file I'd made and stupidly left in the /etc directory. Thus, --purge and AlsaUpdater and such did nothing for the problem.

Sometimes I wonder how many of my hours have been spent seeking a missing ; or an added space. Thank you very much for your time and your script - it's great to have a quick method for updating alsa...and had I not run alsa-info I'd still be being thwarted by a single space. Doh.

[soundcheck]

@soundcheck:
Well, alsa-info to the rescue. The file shows:

snd-hda-intel: index=0 model=3 stack position_fix=0 single_cmd=0

as well as the correct:
snd-hda-intel: index=0 model=3stack position_fix=0 single_cmd=0

under modprobe options, which would explain the 'stack' error. So it amounts to a typo. What is amusing is that said typo was in a backup file I'd made and stupidly left in the /etc directory. Thus, --purge and AlsaUpdater and such did nothing for the problem.

Sometimes I wonder how many of my hours have been spent seeking a missing ; or an added space. Thank you very much for your time and your script - it's great to have a quick method for updating alsa...and had I not run alsa-info I'd still be being thwrted by a single space. Doh.

Tell me:

Did you get it working with model=3stack now & all controls up'n running?
I looked up model=hp from your alsa-base file in the ux-checker.log!?!?
And option "gateway" should do the trick for your codec!?!?

To look for double entries you can scan the entire directory:
sudo grep -R "snd-hda-intel" /etc/modprobe.d

let me know what it says.

[blackest_knight]

Hi.

I received a new git snapshot of the ALSA drivers. The 2.6.24 problem
should be solved with this. This is a very early revision- nothing official!!

I am not really sure how to handle this. I'd need somebody capable
to test this manually.

Volunteers?

What you would have to do:

Download the snapshot from here:

ftp://ftp.kernel.org/pub/linux/kerne...napshot.tar.gz

NOTE: Above snapshot can be also used for getting the very latest design status, that might resolve also other probelms. For people willing to
play around a bit, I'd recommend to try it.

Then execute following commands:

cd <your-downloaddir>
sudo tar xvvf alsa-driver-snapshot.tar.gz
cd alsa-driver
sudo ./configure --with-cards=all --with-card-options=all --with-sequencer=yes --with-oss=yes  --prefix=/usr
sudo make

Let me know if this works!

If yes - we got two options: Either you install this snapshot manually or I'll check
with the Alsa guys when we can expect an offical update.

If above compiles properly, the manual installation would look like this.

1. You copy the alsa-driver directory to /usr/src/Alsa-1.0.18a/alsa-driver-1.0.18a

   cd alsa-driver
   sudo cp -rf * /usr/src/Alsa-1.0.18a/alsa-driver-1.0.18a

2. Now you run the Upgrade script with -i option ( don't use the -d or -f options anymore !!!)

THX

Well the latest version of the installation script was failing with this kernel version on Hardy on an Aspire One.

The above Post solved the issue allowing the drivers to compile properly.
I can confirm a partly working sound system which remembers the mixer settings.
I havent tested very much so far, if I get time tonight I will test further

[blackest_knight]

few problems initially but editing /etc/modprobe.d/alsabase and adding/editing this line to

options snd-hda-intel model=acer-aspire and things started to work better on the aspire one.

HDA Intel (hw:intel,0) selected for input device on skype allowed me to successfully make a test call a bit hissy but it was using a 3g phone as a modem (three skype phone amoi wa01 i think).

things are looking up

[Jaciss]

@soundcheck

Did you get it working with model=3stack now & all controls up'n running?

Yes, everything seemed to be up and running, but...

I looked up model=hp from your alsa-base file in the ux-checker.log!?!?

I did change it back to model=hp since that's what the alsa docs (and the helpful post on page two of this topic) said was correct. The other options were what I had to use previously for things to work. The hp option is preferable, though, as 3stack seems to break the headphone jacks. Alsamixer and gnome-sound* things all seem normal.

And option "gateway" should do the trick for your codec!?!?

I have no idea.

To look for double entries you can scan the entire directory:
sudo grep -R "snd-hda-intel" /etc/modprobe.d

let me know what it says.

/etc/modprobe.d/sound:alias snd-card-0 snd-hda-intel
/etc/modprobe.d/sound:alias sound-slot-0 snd-hda-intel
/etc/modprobe.d/alsa-base:options snd-hda-intel index=0 model=hp

However, it had multiple entries for snd-hda-intel until I removed that backup file.

Thanks again. I'm happy to answer questions if you've any more - it's the least I can do.

[soundcheck]

Well the latest version of the installation script was failing with this kernel version on Hardy on an Aspire One.

The above Post solved the issue allowing the drivers to compile properly.
I can confirm a partly working sound system which remembers the mixer settings.
I havent tested very much so far, if I get time tonight I will test further

Did you run the script three times with options 1. -d 2. -snap 3. -i ???? In this case the manual installation of the snapshot shouldn't be needed any longer.
If you just ran -di or -i , no wonder it didn't work.

[blackest_knight]

with kernel 2.6.24.22 on hardy version 1.15 of the script failed to compile (used with option di initially then just i)

finding your post about the git snapshot i downloaded that, compiled it successfully (couple of warns but nothing related to my hardware) but didnt install it.
then I moved the driver folder to source as detailed above and ran a slightly modded version of your script ( i just commented out the exec lines so i could watch what was going on as it was compiling.
There was some audio working (log on sound) but initially the only mic sound was something distorted which might have had some kind of speech in it.

I found there was a new module option module=acer-aspire and then I got a working internal mic. The mixer settings are retained between boots (just checked no skype audio problems ).

thanks for that 1.0.18rc3 was working quite well, just the having to set the internal mike before using skype was annoying. your solution solves the problem.

btw didn't realise there was a -snap option for the script, that would have been easier still.

[druellan]

Update: still can't make the internal mic work using mode=basic or mode=quanta. On mode=auto works really well, so I think that perhaps its a matter of found the right combination of controls, perhaps a bad labeled mic.

Confirmed. mode=basic and mode=quanta seems unable to unmute the mic. So, there is still a regression from Alsa 1.0.16.