[Regression] 0x1025007f needs model=will quirk

In fact I'm not sure this bug came from pulseaudio, I simply ran "ubuntu-bug pulseaudio" in the hope that it can brought some relevant information.

Is there something missing ?

This is truly an annoying bug you now.

I did a fresh install of Karmic alpha 5 and this bug is still alive.

What can I do for trying to figure out this issue ?

For me the problem was arecord holding open the snd device, this was because of the randomsound init script kicking it off to generate entropy. try lsof | grep snd, and see anything is opening those devices directly. Confirm, first of all with pulseaudio -vv.

Thank you for your answer.

Here are your log.
For pulseaudio -vv, I had to kill the pulseaudio daemon and run the command quickly before it automatically restart.
(otherwise, the command end simply with "Daemon already running")

As you can see, the problem seem to be related to snd, indeed.

I can't figure out what all these errors mean.

Here is another "lsof | grep snd" log but while a sound is being played in Totem.

You can notice that pulseaudio touch the snd device 5 times, is it normal ?

Still the same problem.

I tried, once more, to re-install with a daily-live but no progress.

What can I do ?

Today I found an old daily-live (06/25 → alpha 2.5 if you prefer) and I tested it.

Sound works properly on this, so this bug came after.

software versions for this .iso:
pulseaudio 1:0.9.15-4ubuntu1
linux 2.6.30-9-generic

Could this be kernel issue ?
How can I figure it out ?

Hi Malaria,

Thanks for testing and confirming the previous versions of the kernel and pulseaudio where this wasn't an issue. That'll help narrow down the regression. The easiest way to determine if this is a kernel issue is to try installing and booting back into the known working 2.6.30-9 kernel. If you still have no sound then it's likely a pulseaudio regression. If sound comes back using the 2.6.30-9 kernel vs the 2.6.31-11 kernel, then it's likely a kernel regression. You can find the old builds of 2.6.30-9 at the following. Please let us know your results. Thanks!

https://edge.launchpad.net/ubuntu/+source/linux/2.6.30-9.10/+build/1074090

ie. https://edge.launchpad.net/ubuntu/+source/linux/2.6.30-9.10/+build/1074090/+files/linux-image-2.6.30-9-generic_2.6.30-9.10_i386.deb

Hi Leann,

I already installed my older daily-live but, after all updates, the old kernel prevented X to start (and I didn't manage to resolve this).

Thank you for your link, I'm currently installing the beta and I will try with some older kernels.

OK, so, unsurprisingly, the beta has the same issue.
I tried jaunty's kernel, 2.6.30-9 (the one that used to work on the 6/25 daily) and 2.6.30-10 and I got absolutely no sound.
(whereas with an an up-to-date kernel there is "at least" very low and distort sounds on my headphones.)

I also tried with kernel 2.6.31-13 and 2.6.31-14 but this time the original bug came back (see bug description for symptoms)

I don't know if these tests are relevant, pre-2.6.31 incompatibility might be caused by some user-space change or other.

What else can I do ?

This bug is a show stopper for me, I really want this to be fix before the release !

I (once again) reinstalled with 10/08 daily-live with no results. I tried to remove pulseaudio for a test purpose, but it wanted to remove gnome entirely so I canceled.

I though about trying another distro for seeing if this bug is reproducible, but in fact I'm not sure this would be very helpful.
What do you think ?

If you have any idea or test to perform (even a very annoying one) which might help, I will do the necessary.

I can also confirm this bug. HDA Intel audio in a dell laptop (If more info is needed I can provide it). This is an extremely annoying bug as without sound the system loses lots of functionality. I am willing to help provide anything I can if I can help fix it!!

Jon, please file a separate bug using "ubuntu-bug pulseaudio".

On Oct 8, 2009 8:05 PM, "Jon Neal" <email address hidden> wrote:

I can also confirm this bug. HDA Intel audio in a dell laptop (If more
info is needed I can provide it). This is an extremely annoying bug as
without sound the system loses lots of functionality. I am willing to
help provide anything I can if I can help fix it!!

--
[Regression] No sound in Karmic (HDA Intel ICH6 & Realtek ALC260)

https://bugs.launchpad.net/bugs/418627 You received this bug notification
because you are a member o...

Audio Team, which is a bug assignee. Status in “linux” package in Ubuntu:
Incomplete Bug descripti...

Binary package hint: pulseaudio First of all, notice that sound used to work
well on Jaunty (previo...

I did some more kernel testing with karmic/jaunty, so here is a summary of all my tests in this area:

Current jaunty's kernel:
Work great with jaunty, no sound at all on karmic → (so, pulseaudio issue ?)

2.6.30-9
used to work with karmic prior alpha 3 but currently it brings no sound at all.

2.6.31 series (I tested many of them with the same result)
No sound at all on Jaunty.
On Karmic, no sound on internal speakers of my laptop but, with headphones, I can hear the sound but it's low and distort.

2.6.32 (tested with rc3 from here: http://kernel.ubuntu.com/~kernel-ppa/mainline/v2.6.32-rc3/ )
The same as with 2.6.31 (nothing on Jaunty, something strange on Karmic)

All of this seem strange to me, and this time I really don't know what else I can do to help.
Is there a way to downgrade easily pulseaudio for testing ?

@Jon Neal: Do you have the same behaviour with headphones ?

Confirmed also un Kubuntu karmic kernel 2.6.31-13. Kde don't use pulseaudio here so I think it is a kernel related issue.

I also tried with kubuntu beta and, indeed, there is exactly the same bug.

So, we know now that it is a kernel issue, and it seems to have been introduced since 2.6.31 (according to my former tests).

What's next ?

Beta is too old at this point; you need to try a daily-live:
http://cdimage.ubuntu.com/kubuntu/daily-live/20091011/

Confirm: same problem with 10/11 daily-live.

Please use "ubuntu-bug alsa-base" from the 20091011 daily-live.

On Oct 11, 2009 6:00 PM, "Malaria" <email address hidden> wrote:

Confirm: same problem with 10/11 daily-live.

--
[Regression] No sound in Karmic (HDA Intel ICH6 & Realtek ALC260)

https://bugs.launchpad.net/bugs/418627 You received this bug notification
because you are a member o...

Audio Team, which is a bug assignee. Status in “linux” package in Ubuntu:
Incomplete Bug descripti...

Binary package hint: pulseaudio First of all, notice that sound used to work
well on Jaunty (previo...

ubuntu-bug alsa-base is in bug #449715.

@Malaria

We have created a new package which contains the latest alsa. The latest alsa seems to have fixed a few problems that are similar to this. Could you please install linux-backports-modules-alsa and retest then report back to the bug.

Thanks

I have installed "linux-backports-modules-alsa-karmic-generic" and, unfortunately, after a reboot the problem still remain.

On Thu, Oct 15, 2009 at 3:52 PM, Malaria <email address hidden> wrote:
> I have installed "linux-backports-modules-alsa-karmic-generic" and,
> unfortunately, after a reboot the problem still remain.

Please use apport-collect -p alsa-base 418627

Architecture: i386
ArecordDevices:
 **** List of CAPTURE Hardware Devices ****
 card 0: Intel [HDA Intel], device 0: ALC260 Analog [ALC260 Analog]
   Subdevices: 1/1
   Subdevice #0: subdevice #0
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC0: nicolas 1936 F.... pulseaudio
Card0.Amixer.info:
 Card hw:0 'Intel'/'HDA Intel at 0xd0000000 irq 16'
   Mixer name : 'Realtek ALC260'
   Components : 'HDA:10ec0260,1025007f,00100400 HDA:14f12bfa,1025007f,00090000'
   Controls : 23
   Simple ctrls : 12
DistroRelease: Ubuntu 9.10
InstallationMedia: Ubuntu 9.10 "Karmic Koala" - Alpha i386 (20091012)
Package: alsa-base 1.0.20+dfsg-1ubuntu5
PackageArchitecture: all
ProcEnviron:
 SHELL=/bin/bash
 LANG=fr_FR.UTF-8
ProcVersionSignature: Ubuntu 2.6.31-14.47-generic
Uname: Linux 2.6.31-14-generic i686
UserGroups: adm admin cdrom dialout lpadmin plugdev sambashare
XsessionErrors:
 (gnome-settings-daemon:1952): GLib-CRITICAL **: g_propagate_error: assertion `src != NULL' failed
 (gnome-settings-daemon:1952): GLib-CRITICAL **: g_propagate_error: assertion `src != NULL' failed
 (nautilus:2052): Eel-CRITICAL **: eel_preferences_get_boolean: assertion `preferences_is_initialized ()' failed
 (polkit-gnome-authentication-agent-1:2082): GLib-CRITICAL **: g_once_init_leave: assertion `initialization_value != 0' failed

Ok, here it is. Now I know how to use apport-collect :)

Architecture: amd64
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC0: Slmodemd 1408 F.... slmodemd
 /dev/snd/pcmC0D6c: Slmodemd 1408 F.... slmodemd
 /dev/snd/pcmC0D6p: Slmodemd 1408 F.... slmodemd
Card0.Amixer.info:
 Card hw:0 'Intel'/'HDA Intel at 0xd8240000 irq 22'
   Mixer name : 'Analog Devices AD1986A'
   Components : 'HDA:11d41986,144dc504,00100500 HDA:11c13026,2117144d,00100700'
   Controls : 39
   Simple ctrls : 23
DistroRelease: Ubuntu 9.10
Package: alsa-driver (not installed)
ProcEnviron:
 PATH=(custom, no user)
 LANG=fr_FR.UTF-8
 SHELL=/bin/bash
ProcVersionSignature: Ubuntu 2.6.31-14.47-generic
Uname: Linux 2.6.31-14-generic x86_64
UserGroups:

XsessionErrors:
 (gnome-settings-daemon:1850): GLib-CRITICAL **: g_propagate_error: assertion `src != NULL' failed
 (gnome-settings-daemon:1850): GLib-CRITICAL **: g_propagate_error: assertion `src != NULL' failed
 (polkit-gnome-authentication-agent-1:1960): GLib-CRITICAL **: g_once_init_leave: assertion `initialization_value != 0' failed
 (nautilus:1945): Eel-CRITICAL **: eel_preferences_get_boolean: assertion `preferences_is_initialized ()' failed
 (gnome-panel:1944): Gdk-CRITICAL **: gdk_x11_colormap_foreign_new: assertion `GDK_IS_VISUAL (visual)' failed

I've got this problem too but with an :
00:1b.0 Audio device: Intel Corporation 82801G (ICH7 Family) High Definition Audio Controller (rev 02)

I post the data needed upper in the bug repport.

The issues is present for me in 2.6.31-14-generic but NOT in 2.6.31-12-generic. Attached is my 'lspci' in 2.6.31-12-generic.

Here are some linked images of what my Sound Preferences MUST LOOK LIKE to have sound. In the 2.6.31-14-generic kernel there is no option for the "Internal Audio" as THERE is when I use 2.6.31-12.

http://i37.tinypic.com/2i8exyw.png - Sound Preferences, Hardware Tab

http://i33.tinypic.com/ej7xu9.png - Sound Preferences, Output Tab

Bug now gone for me with CURRENT updates. Sound is back on 2.6.31.14-generic.

In my case 2.6.31.14 do nothing more than 2.6.31-12 → issue as explain in the description.
I do have an option for « Internal Audio » (this is the only one in fact) but, well, it's useless.

What more can I investigate to figure out this bug ?

If only I could have a small clue or a way to investigate... I wan't to hope !

I had a dream last night, I saw a dumb koala and, trust me, it was really sad :(

Try this:

rm -r ~/.pulse*

Then run gnome-alsamixer to re-adjust the volume settings (more specifically, the Master and PCM settings). This fixed the problem for me. Then again, I have an SB Live 5.1 card.

@Joshua Solomon: worked for me too.

I've installed gnome-alsamixer -without removing old pulse conf files- and found out PCM channel was disabled, the Mute checkbox was selected.

Effectively, some channels were off in gnome-alsa-mixer but, unfortunately, it didn't solve my problem.

I tried with many different levels combo and I figure out that the "Front" level has absolutely no effect, from 0% to 100% (but strangely mute is effectively mute !)

"Master", "Line" and "Speaker" seem to control the same level (logical, if there is no working "Front", isn't it ?)

Concerning "Beep", if I un-mute it, sound become even more cracking (on headphones, still nothing on speakers).
And for the two boxes "IEC958" and "IEC958 Default PCM" they do nothing (even if no one is selected).

So, the "Front" level might be the problem (or the symptom ?).

What's next ? I'm counting on you !

I downgraded alsa-base, alsa-utils and similar too jaunty's versions but it did not change anything. Even with jaunty's kernel.

And if it was not related too alsa nor the kernel ?

Yes, I'm desperate.

Guess what ? It still doesn't work with the release-candidate...
I also tried with fedora 12 beta and there is exactly the same bug.

Once more, what can I do ?

@Malaria Please use apport-collect -p alsa-base 418627 with current
Karmic and linux-backports-modules-alsa-karmic installed.

Architecture: i386
ArecordDevices:
 **** List of CAPTURE Hardware Devices ****
 card 0: Intel [HDA Intel], device 0: ALC260 Analog [ALC260 Analog]
   Subdevices: 1/1
   Subdevice #0: subdevice #0
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC0: nicolas 1640 F.... pulseaudio
Card0.Amixer.info:
 Card hw:0 'Intel'/'HDA Intel at 0xd0000000 irq 16'
   Mixer name : 'Realtek ALC260'
   Components : 'HDA:10ec0260,1025007f,00100400 HDA:14f12bfa,1025007f,00090000'
   Controls : 23
   Simple ctrls : 12
DistroRelease: Ubuntu 9.10
InstallationMedia: Ubuntu 9.10 "Karmic Koala" - Release Candidate i386 (20091020.3)
Package: alsa-base 1.0.20+dfsg-1ubuntu5
PackageArchitecture: all
ProcEnviron:
 SHELL=/bin/bash
 LANG=fr_FR.UTF-8
ProcVersionSignature: Ubuntu 2.6.31-14.48-generic
Uname: Linux 2.6.31-14-generic i686
UserGroups: adm admin cdrom dialout lpadmin plugdev sambashare
XsessionErrors:
 (gnome-settings-daemon:1655): GLib-CRITICAL **: g_propagate_error: assertion `src != NULL' failed
 (gnome-settings-daemon:1655): GLib-CRITICAL **: g_propagate_error: assertion `src != NULL' failed
 (nautilus:1758): Eel-CRITICAL **: eel_preferences_get_boolean: assertion `preferences_is_initialized ()' failed
 (polkit-gnome-authentication-agent-1:1793): GLib-CRITICAL **: g_once_init_leave: assertion `initialization_value != 0' failed

I re-did a fresh rc install and I just did the following:
- all upgrades
- install linux-backports-modules-alsa-karmic-generic
- put all level to 100% in alsamixer and sound-preference dialogue
- reboot
- apport-collect -p alsa-base 418627

Of course sound did not work better than before,but I just wanted to note my exact process in case I'm doing something wrong.

(And thank you for your help)

I have the same issue in a HP compaq nx6310 laptop, attached is my "lsof", Alsa reports that I have a HDA Intel Chip: Analog Devices AD1981.

Making some more checks I came to realize that no audio device is being shown on the sound preferences panel...

I just found that bug #394500 is exactly my problem and not this one, I'm really sorry for any inconvenience.

I tried with the real-time kernel but it did not work better.

(I didn't have much hope about it anyway)

Here are two screen-shots of alsamixer: one on Jaunty and one on Karmic.

As you can see, channels are not the same (more output on Jaunty, beep only on karmic). Also notice that sound works on Jaunty, and that I rose manually all levels on karmic (with no results).

So, I though about an alsa detection bug and I tried to add "options snd-hda-intel model=auto" or "options snd-hda-intel model=acer-aspire" or "options snd-hda-intel model=Realtek_ALC260 " to /etc/modprobe.d/alsa-base.conf but it did not work.

I also add myself to the audio group and I removed .pulse: absolutely no effect.

What can I do ? Is it hopeless ? :(

Sorry, I forgot to say that I also tried to disable the audio-card powersaving but, once more, with no result.

I may be wrong, but this may be a device detection issue: In
/etc/pulse/default.pa
I changed the following block

### Automatically load driver modules depending on the hardware available
.ifexists module-udev-detect.so
load-module module-udev-detect
.else
### Alternatively use the static hardware detection module (for systems that
### lack udev support)
load-module module-detect
.endif

to

### Automatically load driver modules depending on the hardware available
#.ifexists module-udev-detect.so
#load-module module-udev-detect
#.else
### Alternatively use the static hardware detection module (for systems that
### lack udev support)
load-module module-detect
#.endif

Then rebooted, and it worked. This was suggested by someone in another bug report: In my case, it worked.

Unfortunately, it didn't work for me (even combined with the removal of .pulse)

Anyway, thank you for your help.

I looked at some similar bugs, and I can add that this is not related to a conflict with slmodemd.

sudo fuser -v /dev/dsp* /dev/snd/* return

/dev/snd/controlC0: nicolas 1629 F.... pulseaudio

I tried to upgrade my jaunty to karmic and the same bug appears.

Someone can help me ?

Same thing with a vanilla install.

So, no one has an idea or just a start of a theory to solve or to figure out this bug ?

We saw (in fact, I understood) that this is likely to be an alsa issue. Now, is this auto-detection or the driver itself which doesn't want to do his job ?

i have lost sound on upgrading to karmic. this is a kingsize problem for me.

hopefully relevant info:

luke@orion:~$ lsmod | grep snd
snd_hda_codec_si3054 5856 1
snd_hda_codec_realtek 277860 1
snd_hda_intel 31880 3
snd_hda_codec 87584 3 snd_hda_codec_si3054,snd_hda_codec_realtek,snd_hda_intel
snd_hwdep 9352 1 snd_hda_codec
snd_pcm_oss 44704 0
snd_mixer_oss 18976 1 snd_pcm_oss
snd_pcm 93160 6 snd_hda_codec_si3054,snd_hda_intel,snd_hda_codec,snd_pcm_oss
snd_seq_dummy 3460 0
snd_seq_oss 33440 0
snd_seq_midi 8192 0
snd_rawmidi 27360 1 snd_seq_midi
snd_seq_midi_event 8448 2 snd_seq_oss,snd_seq_midi
snd_seq 60608 6 snd_seq_dummy,snd_seq_oss,snd_seq_midi,snd_seq_midi_event
snd_timer 26992 2 snd_pcm,snd_seq
snd_seq_device 8308 5 snd_seq_dummy,snd_seq_oss,snd_seq_midi,snd_rawmidi,snd_seq
snd 77096 17 snd_hda_codec_si3054,snd_hda_codec_realtek,snd_hda_intel,snd_hda_codec,snd_hwdep,snd_pcm_oss,snd_mixer_oss,snd_pcm,snd_seq_oss,snd_rawmidi,snd_seq,snd_timer,snd_seq_device
soundcore 9088 1 snd
snd_page_alloc 10928 2 snd_hda_intel,snd_pcm

relevant part of verbose lspci:

00:1b.0 Audio device: Intel Corporation 82801H (ICH8 Family) HD Audio Controller (rev 03)
 Subsystem: Hewlett-Packard Company Device 30cc
 Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR+ FastB2B- DisINTx-
 Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
 Latency: 0, Cache Line Size: 64 bytes
 Interrupt: pin A routed to IRQ 22
 Region 0: Memory at f8400000 (64-bit, non-prefetchable) [size=16K]
 Capabilities: <access denied>
 Kernel driver in use: HDA Intel
 Kernel modules: snd-hda-intel

also, sound preferences shows just a dummy device, and alsamixer shows the sound chip to be "Realtek ALC268".

mute led is is illuminated orange (should be blue). this has never happened before.

basically it looks to me like the kernel module has been broken bringing all audio down with it...

hope somebody can help. i will now try booting using the old kernel module from jaunty to see if im right...

i will be glad to provide further info please just tell me what to post.

thanks to everyone who helps people like me and makes ubuntu the best OS there is (albeit temporarily having a problem).

ex-oficio

the old kernel + modules didnt help.

dont know what has happen to the sound setup in ubuntu, but it doesnt wanty to play nicely with the driver and hardwiare. to expand on my previous post, the mute light is blue for most of the boot-up process (which is a total horror-show compared to the jaunty boot by the way) but then turns orange near the end. something is disabling the sound card!

all that dmesg has is:

[ 24.873949] HDA Intel 0000:00:1b.0: PCI INT A -> GSI 22 (level, low) -> IRQ 22
[ 24.873984] HDA Intel 0000:00:1b.0: setting latency timer to 64
[ 25.591505] input: HDA Digital PCBeep as /devices/pci0000:00/0000:00:1b.0/input/

?

the hack to "/etc/pulse/default.pa" posted by Nusret on 2009-10-28 (post#77) worked for me.

to be accurate, i made the change, rebooted opened sound preferences and saw that the "dummy device" had been replaced by "internal audio analog device". however mute led was orange, but sotware reported the device to be unmuted. after cycling the mute checkbox on and off, the led was blue and behaviour was correct. this behaviour has continued after a reboot.

audio jack behaviour is correct as well.

this is obviously not an acceptable situation for the less tech savvy, and really shouldnt be acceptable to any of us because it means that something somewhere isnt doing what it should be.

it also probably means that when the issue is solved those of us that implement this hack will end up with incorrect "/etc/pulse/default.pa" files, but hey at least i have sound now.

i will still gladly provide any pertinent info asked for in pursuit of fixing this bug as it isnt really fixed.

cheers,
ex-oficio.

@ex-oficio That's a separate PulseAudio bug.

I don't know if this is the "right" way, but I was without sound and I fixed it like this:

- Edit /etc/pulse/default.pa
- Uncomment the line:

#load-module module-alsa-sink

- "killall pulseaudio" in a terminal and wait a bit until it respawns (you'll know it because the mixer icon in the panel appears again)
- Use alsamixer to unmute all the "unmuteable"; in my case, my headphones are connected to the "front" plug, so I unmuted the master control and the front control

It worked for me, I can't guarantee it works for everyone.

@retrinqueiro: it didn't work for me.

No progress since the release (with all updates, even from proposed).

I built alsa from sources and also from the custom realtek version (found on their website) but I always fall in the same issue.

Of course, I tried several options in /etc/modprobe.d/alsa-base.conf but, no-results.

If you have ANY GUESS or ANY IDEA for this bug, just post ! I installed karmic on an external hdd so I can experiment safely.

I will try with Lucid I soon as a live .iso will be available, but I still want to find a solution for Karmic.

On Mon, Nov 9, 2009 at 1:48 PM, Malaria <email address hidden> wrote:
> If you have ANY GUESS or ANY IDEA for this bug, just post ! I installed
> karmic on an external hdd so I can experiment safely.

Many of the Realtek changes just landed in Linus's 2.6.32 tree, so
your symptoms should be resolved in current 2.6.32-git. Alternately,
you can try a snapshot of alsa-driver stable:
http://kernel.org/pub/linux/kernel/people/tiwai/snapshot/alsa-driver-20091110.tar.bz2

I tried with the alsa-driver snapshot but it did not work better :(

I simply did ./configure, make, sudo make install-modules and after a reboot this bug is still alive.
(the same with the common sudo make install)

I tried it too on Jaunty and then their is no sound at all (not even an ugly one on headphones)
→ it comes back to normal (only on jaunty) when I reinstall every things that synaptic find with a "sound" search.

Am I doing something wrong with this ?

What are the contents of /proc/asound/version ?

On karmic it was 1.0.21and the build date was not coherent (as I remember).

I will retry this with a newer snapshot (or it's perhaps better to wait for the coming-soon release of 2.6.32 rc7 ?)

Upstream kernels always lag alsa-driver snapshots.

On Nov 12, 2009 2:36 PM, "Malaria" <email address hidden> wrote:

On karmic it was 1.0.21and the build date was not coherent (as I
remember).

I will retry this with a newer snapshot (or it's perhaps better to wait
for the coming-soon release of 2.6.32 rc7 ?)

--
[Regression] No sound in Karmic (HDA Intel ICH6 & Realtek ALC260)

https://bugs.launchpad.net/bugs/418627 You received this bug notification
because you are a member o...

Audio Team, which is a bug assignee. Status in “alsa-driver” package in
Ubuntu: New Bug descriptio...

Binary package hint: pulseaudio First of all, notice that sound used to work
well on Jaunty (previo...

Ok, so I retried with today's snapshot.

On Jaunty, I noticed that the problem is indeed the same as in Karmic : I hear the low and disrupted sound when I set the speaker channel (in alsamixer for example) closely to 100 (see comment #55).
cat /proc/asound/version worked as expected : 1.0.21 with today as a build date.

On karmic, it's even more strange : after a reboot I hear the drum of gdm (only with headphones) but after login there is no sound at all.

For both Jaunty and Karmic, I did ./configure, make, make install-modules then reboot then test then make install then reboot and test.

Oh, one more thing, I tried to build alsa-driver 1.0.18rc3 or 1.0.19 on karmic, but it always failed with several " error: ‘struct proc_dir_entry’ has no member named ‘owner’ "...

I tried with 11/20 daily-live of Lucid and this bug is still there.

Once more, what can I do/try ?

Is there at least an upstream bug for this ?

I must admit that alsa's bugzilla seems odd to me.
Anyway I don't know how to explain this bug in an accurate way (which tests I already performed are relevant...)

Hello there ?

As I said, I won't give up on this.
I repeat that this bug is still alive on Lucid and the fact that it doesn't seem to be any start of a movement that can bring us a fix is very disappointing (and especially because this is a regression).

So I re-ask some questions :

- Why pre-1.0.20 alsa-driver don't build on Karmic ? If I could build it, it should be possible to see if alsa-driver really own the key of this bug.
Google doesn't help here, all I found is to manually patch some files but then the build just fail later with the same error. And this again and again...

- Is it relevant to fill an upstream bug against alsa-driver ? If yes, which part of this report would help upstream ?

- What else can I do ?

Hi !
I'm also affected by this bug (same laptop as Malaria's). But I didn't see any improvement !
Is there something else I can do, I mean excepted by waiting ?

after the latest kernel upgrade last week my sound has been behaving erratically again.

retrinqueiro posted this on 2009-11-02:

*********************

- Edit /etc/pulse/default.pa
- Uncomment the line:

#load-module module-alsa-sink

- "killall pulseaudio" in a terminal and wait a bit until it respawns (you'll know it because the mixer icon in the panel appears again)

*********************

i now have sound, but with odd symptoms regarding the little "mute" indicator light (which is also a touch sensitive button) on my keyboard. normally its orange when the sound is muted (or the sound is broken) and blue otherwise. now, it is orange when no sound is playing (when there are no applications listed as playing sound in the pulse preferences dialog), and blue when sounds are being played! the button functionality still works so its not so bad, but i keep seeing the orange light and thinking the sound has failed again!

what was wrong with the sound setup in intrepid? can somebody explain why we need pulseaudio? it has caused terrible sound issues with wine and dosbox for me.

Hi there !
I finally find the solution. (I only reboot one time. hope it will continue then).
I add this line "options snd-hda-intel model=will probe_mask=1" at the end of the file /etc/modprobe.d/alsa-base.conf.
Then reboot and it's a success.

All of the honours must go to this page : http://ubuntuforums.org/showthread.php?p=5397124

Indeed, I confirm that this workaround this bug. Many thanks.

So, this is a great news (now I can upgrade to Karmic), but this is only a workaround : this bug still needs a patch...

Malaria, do you actually need both the model quirk *and* the
position_fix=1 quirk?

On Mon, Dec 7, 2009 at 7:18 PM, Daniel Chen <email address hidden> wrote:
> Malaria, do you actually need both the model quirk *and* the
> position_fix=1 quirk?

And, in fact, it looks like model=will is the wrong quirk; if you need
one at all, it should be model=acer.

So, please first try omitting the model quirk completely and using
only position_fix=1. Then try omitting the position_fix=1 quirk and
using only model=acer. Finally, use both quirks. You'll need to reboot
between each of the three tests.

Thank you Daniel, here are the results :

options snd-hda-intel probe_mask=1 → doesn't work

options snd-hda-intel model=acer probe_mask=1 → it works ! (wtf, I was absolutely sure that I have tested this before !)
But, this time, sound is TOO high : 5% on the gnome applet seems already quite high (equivalent to 60-70% on Jaunty)

options snd-hda-intel model=acer → same as above (I nearly awoke all my family when I tested without headphones !)

options snd-hda-intel model=will probe_mask=1 → it works, and it provide more reasonable sound (let's say 50% with this equal 10% with model=acer)

And, one more test :

options snd-hda-intel model=will → same as above, so probe_mask=1 is indeed useless.

That's all for now, thank you for your help !

On Mon, Dec 7, 2009 at 7:57 PM, Malaria <email address hidden> wrote:
> Thank you Daniel, here are the results :

Please test both model quirks (acer and will) using microphone recording.

After testing, both quirks work for microphone recording.

Notice that, with "will", I have two microphone entry in sound-preferences and that only "microphone 1" works. Whereas there is only one entry with "acer" (and it works).

In a nutshell, it seems that acer is indeed the right setting but "will" is simply more usable in this case.

Malaria, can you verify that model=auto also does not result in audible playback or capture?

I confirm that model=auto result in this bug.

But I also confirm that model=will is, in this case, way more usable than model=acer.
(I think that jezeos could confirm this too)

Is this another bug ?
If not, we have to change the title to "[Regression] 0x1025007f needs model=will quirk"

Ok, one last check: please uninstall linux-backports-modules-alsa-$(uname -r), compile the latest stable snapshot (http://kernel.org/pub/linux/kernel/people/tiwai/snapshot/alsa-driver-20091213.tar.bz2), and test playback and capture without any model quirk and with model= quirks.

I am on a fresh install so there were no linux-backports-alsa on my system.

So, with the latest snapshot, there is no difference :

- Bug without quirck
- Work as expected with model=will (both microphone, speakers and headphones
- Same with model=acer, but sound is too high (after some extra tests, I may say that it is 2x more on headphones and perhaps even more on speakers. Considering that sound become very ugly around 40%, thanks too integrated speakers, it's more useful and flexible to use model=will)

I tried the option model=acer instead of model=will. I don't find that the sound is too high. (I didn't try eadphones).
The sound is very ugly, but by changing the subwoofer's volume (with alsamixer command) which is too high, the sound becomes good.
with model=will, I did not have the opportunity to change the subwoofer's volume. And the sound seems the same after change. So, according to me model=acer is a better choice.
(I'm pretty sure that earlier I tried model=acer and get no sound.)

Ok, so I retried with model=acer.
In fact the volume control in the top board, controls the subwoofer's volume. So it's necessary to use the alsamixer to change the volume. Not pratical, so !

@jezeos
Please use: apport-collect -p alsa-base 418627

AplayDevices:
 **** List of PLAYBACK Hardware Devices ****
 card 0: Intel [HDA Intel], device 0: ALC260 Analog [ALC260 Analog]
   Subdevices: 1/1
   Subdevice #0: subdevice #0
Architecture: i386
ArecordDevices:
 **** List of CAPTURE Hardware Devices ****
 card 0: Intel [HDA Intel], device 0: ALC260 Analog [ALC260 Analog]
   Subdevices: 1/1
   Subdevice #0: subdevice #0
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC0: jez 1516 F.... pulseaudio
Card0.Amixer.info:
 Card hw:0 'Intel'/'HDA Intel at 0xd0000000 irq 16'
   Mixer name : 'Realtek ALC260'
   Components : 'HDA:10ec0260,1025007f,00100400 HDA:14f12bfa,1025007f,00090000'
   Controls : 22
   Simple ctrls : 14
CheckboxSubmission: c94102629a78a92484df7519f3dc8001
CheckboxSystem: 6234f1c3ecba396013fc87a03e2fada9
DistroRelease: Ubuntu 9.10
Package: alsa-base 1.0.20+dfsg-1ubuntu5
PackageArchitecture: all
ProcEnviron:
 SHELL=/bin/bash
 LANG=fr_FR.UTF-8
 LANGUAGE=fr_FR.UTF-8
ProcVersionSignature: Ubuntu 2.6.31-17.54-generic
Uname: Linux 2.6.31-17-generic i686
UserGroups: adm admin audio cdrom dialout dip floppy fuse lpadmin netdev plugdev video

Done :)

Same bug and same workaround on Lucid (daily fresh install).

What is missing for a fix now that we know the problem ?

Architecture: amd64
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC1: cristian 2009 F.... kmix
 /dev/snd/controlC0: cristian 1978 F.... knotify4
                      cristian 2005 F.... pulseaudio
                      cristian 2009 F.... kmix
Card0.Amixer.info:
 Card hw:0 'SB'/'HDA ATI SB at 0xfdcf4000 irq 16'
   Mixer name : 'Realtek ALC888'
   Components : 'HDA:10ec0888,14622fbd,00100001 HDA:11c11040,11c10001,00100200'
   Controls : 31
   Simple ctrls : 19
Card1.Amixer.info:
 Card hw:1 'HDMI'/'HDA ATI HDMI at 0xfdee8000 irq 19'
   Mixer name : 'ATI RS690/780 HDMI'
   Components : 'HDA:1002791a,00791a00,00100000'
   Controls : 4
   Simple ctrls : 1
Card1.Amixer.values:
 Simple mixer control 'IEC958',0
   Capabilities: pswitch pswitch-joined
   Playback channels: Mono
   Mono: Playback [off]
DistroRelease: Ubuntu 9.10
Package: alsa-base 1.0.20+dfsg-1ubuntu5
PackageArchitecture: all
ProcEnviron:
 SHELL=/bin/bash
 LANG=es_UY.UTF-8
 LANGUAGE=
Uname: Linux 2.6.32-02063202-generic x86_64
UserGroups: adm admin cdrom dialout lpadmin plugdev sambashare vboxusers www-data

Still the same bug (and same workaround) on an up to date Lucid.

What is the meaning of "Fix Commited" ?

This bug was reported a while ago and there hasn't been any activity in it recently. We were wondering if this is still an issue? Can you try with the latest development release of Ubuntu? ISO CD images are available from http://cdimage.ubuntu.com/releases/ .

If it remains an issue, could you run the following command from a Terminal (Applications->Accessories->Terminal). It will automatically gather and attach updated debug information to this report.

apport-collect -p alsa-base %BUG_ID%

Also, if you could test the latest upstream kernel available that would be great. It will allow additional upstream developers to examine the issue. Refer to https://wiki.ubuntu.com/KernelMainlineBuilds . Once you've tested the upstream kernel, please remove the 'needs-upstream-testing' tag. This can be done by clicking on the yellow pencil icon next to the tag located at the bottom of the bug description and deleting the 'needs-upstream-testing' text. Please let us know your results.

Thanks in advance.

please ignore the last comment, script gone wild

Last time I tried Lucid (a week ago?) this bug was still here.

I still don't understand what "fix committed" means in this context.

If necessary, I'm can do whatever testing you want...

RE my post on the 31st of october ( #84 ) and my problems with sound. i have found the solution and it was the sl-modem proprietary driver:

https://bugs.launchpad.net/ubuntu/+source/sl-modem/+bug/449762

disabling it (in the hardware drivers tool in the administration menu) and restoring my pulesaudio configuration files to their original state overcame the problem.

its interesting that nusret's post from the 28th of october also worked, although requiring manual edits to files.

just posting here in case anyone else has experienced this problem.

The following commit in sound-2.6 needs to be sent to stable@kernel:

commit 950200e2ff11daae1c5d9426703bdd494603f38b
Author: Daniel T Chen <email address hidden>
Date: Sun Dec 13 14:11:02 2009 -0500

    ALSA: hda: Use ALC260_WILL quirk for another Acer model (0x1025007f)

    BugLink: https://bugs.launchpad.net/bugs/418627

    The original reporter states that this quirk is necessary to obtain
    reasonable gain for playback. Without it, sound is inaudible. Tested
    with playback (spkr and hp) and capture.

    Signed-off-by: Daniel T Chen <email address hidden>
    Signed-off-by: Takashi Iwai <email address hidden>

Nice to ear from you Daniel :)

Is there anything I can do to help with this?

Cherry-picked into my lp418627 branch (http://kernel.ubuntu.com/git?p=dtchen/ubuntu-lucid.git;a=shortlog;h=refs/heads/lp418627)

Sent to stable@kernel

Do we have to wait for an upstream inclusion or will the patch be in the default lucid kernel ?

It's already in 2.6.32.11, which is being merged for Lucid post-Beta 2.

I confirm this is fixed.

Many thanks ! :-)

This bug is awaiting verification that the kernel in -proposed solves the problem. Please test the kernel and update this bug with the results. If the problem is solved, change the tag 'verification-needed-lucid' to 'verification-done-lucid'.

If verification is not done by 5 working days from today, this fix will be dropped from the source code, and this bug will be closed.

See https://wiki.ubuntu.com/Testing/EnableProposed for documentation how to enable and use -proposed. Thank you!