no sound Intel HDA (ICH6) and ALC260 (or ALC880) mixer; no errors

[Václav Šmilauer]

This bug is to track upstream ALSA bugs
https://bugtrack.alsa-project.org/alsa-bug/view.php?id=1618 and its duplicate
https://bugtrack.alsa-project.org/alsa-bug/view.php?id=1517.

This bug causes no sound at all at many newer centrino laptops (Acer, for
example), without any errors from ALSA or from apps. It seems to be caused by
incorrect sound-routing in the mixer and is model-specific (some configs with
these chips probably may work). The issue has been brought up in ubuntu forums
several times (notably http://www.ubuntuforums.org/showthread.php?t=76019, but
also http://ubuntuforums.org/showthread.php?t=75979, possibly others). At the
moment, upstream bug is assigned but still open.

[Sebastiaan Samyn (sebastiaan-samyn)]

Hi,

I had the same problem, but updating to kernel 2.6.14 did the trick. Since dapper uses 2.6.15 there shouldn't be a problem any more, I suppose.
The only problem left is that the headphones output doesn't work. I know this can be solved using the drivers from realtek (http://www.realtek.com.tw/downloads/dlhd-2.aspx?lineid=2004052&famid=2004052&series=2004061&Software=True).
These look like adjusted alsa drivers, so maybe they could help.

I truly hope this helps,
Sebastiaan Samyn

[Václav Šmilauer (eudoxos)]

Unfortunately, it does not work with 2.6.15 here (note that the ALSA bug is still open). This week I will try to change internal routing on the realtek chip, as described in the ALSA bugreport.

[Václav Šmilauer (eudoxos)]

This bug was fixed in ALSA CVS. I doubt kernel devs want to go with that version. The fix is AFAIK as simple as replacing linux-kernel/drivers/sound/pci/hda/patch_realtek.c with the new CVS version, without any further interference.

Is there interest from kernel maintainers that I test this file against kernel tree currently in dapper and post it? (Note that this bug affects many recent Acer models and definitely should be fixed before release).

[Daniel T Chen (crimsun)]

There is currently (low-priority) work for merging upstream HDA changes, so the bits needed for this chipset may be affected coincidentally.

[Paul Sladen (sladen)]

Hello Václav and others,

Could you download that Realtek version and attach a:

  diff -Nru old/ new/

to compare the two 'pci/hda' directories so we can easily see what they changed.

If it's more an simple then it should be easy to get applied.

[Daniel T Chen (crimsun)]

Please test with _current_ (i.e., today's) Dapper. The necessary patches have been merged into the default kernel (2.6.15-19.29) already.

[Václav Šmilauer (eudoxos)]

Grepping though current git source (http://www.kernel.org/git/?p=linux/kernel/git/bcollins/ubuntu-2.6.git;a=blob;h=d0e92e540d0548afca53c98b958935a0b876f038;hb=b0e24f9b4bbdb1dbfd3df4ce9637513f27fb7dbc;f=sound/pci/hda/patch_realtek.c), the model 'acer' exists. It was not added to ALSA CVS until the bug had been resolved (see alsa bug thread), so it should work. I will try to install 2.6.15-19 from dapper on the Acer notebook in a few days. Please keep it open for the moment, I will get back with (most likely positive) report. Or someone else will confirm that it works.

[Václav Šmilauer (eudoxos)]

I confirm that the driver works flawlessly AFAICT with dapper 2.6.15-19. The acer model is autodetected (does not need the model=acer parameter).

Many thanks for you work, you can close this bug now.

[Daniel T Chen (crimsun)]

Closing per submitter's update.

[C. Cooke (ccooke)]

I have the same problem - sound on this Sony Vaio A397xp has not worked since 2.6.15-18. I'm up to date as of this morning (2006-05-05 08:41 GMT) and there's still nothing.

I've seen (and commented on) at least one other variation of this bug (it seems to have been opened about a dozen times, all affecting the intel HDA driver)

[Václav Šmilauer (eudoxos)]

The Acer portion of the bug seems to be resolved, though. For the Vaio, try recompiling ALSA driver with debug enabled and use the 'test' model to find out routing on the ALC chip (see ALSA bugs referenced above for howto; the 'test' model is useful if Vaio doesn't use GPIO registers, as Acers do).

I recommend using the ALSA bugtrack, the devs there may help you.

Also, consider opening a separate bugreport specifically for Vaio, perhaps renaming this one to have Acer in the title.

[Daniel T Chen (crimsun)]

On Fri, May 05, 2006 at 08:43:14AM -0000, C. Cooke wrote:
> I have the same problem - sound on this Sony Vaio A397xp has not worked since 2.6.15-18. I'm up to date as of this morning (2006-05-05 08:41 GMT) and there's still nothing.

@C. Cooke: Have you tried passing model=vaio to modprobe? Are you using
2.6.15-22.33?

Thanks,
--
Daniel T. Chen <email address hidden>
GPG key: www.sh.nu/~crimsun/pubkey.gpg.asc

[C. Cooke (ccooke)]

> @C. Cooke: Have you tried passing model=vaio to modprobe? Are you using
> 2.6.15-22.33?

I've now got model=vaio as an option fro snd_hda_intel and as of this morning I'm on 2.6.15-22.34. No differences.

[Florian L (flad)]

I seem to have the same problem here with:

0000:00:1b.0 0403: Intel Corporation 82801G (ICH7 Family) High Definition Audio Controller (rev 01)

cat /proc/asound/card0/pcm0p/sub0/info | grep name
name: ALC260 Analog

...at least I think so, since it is a ALC260 as well.

Using dapper, daily updated-no sound, no errors

[Daniel T Chen (crimsun)]

The HDA ALC* updates (backported from 1.0.11) are in the latest linux-source-2.6.15-23.35 upload, which should be available next week.

[Florian L (flad)]

Unfortunately, 2.6.15-23.35 didn't help. Does this mean that I have another problem?
Some more information:

-----
0000:00:1b.0 0403: Intel Corporation 82801G (ICH7 Family) High Definition Audio Controller (rev 01)
        Subsystem: Hewlett-Packard Company: Unknown device 3010
        Flags: bus master, fast devsel, latency 0, IRQ 225
        Memory at e0540000 (64-bit, non-prefetchable) [size=16K]
        Capabilities: [50] Power Management version 2
        Capabilities: [60] Message Signalled Interrupts: 64bit+ Queue=0/0 Enable-
        Capabilities: [70] #10 [0091]
-----

-----
lsmod | grep snd
snd_hda_intel 20468 1
snd_hda_codec 150768 1 snd_hda_intel
snd_pcm_oss 56448 0
snd_mixer_oss 20544 2 snd_pcm_oss
snd_pcm 96676 3 snd_hda_intel,snd_hda_codec,snd_pcm_oss
snd_timer 26884 1 snd_pcm
snd 60004 6 snd_hda_intel,snd_hda_codec,snd_pcm_oss,snd_mixer_oss,snd_pcm,snd_timer
soundcore 10784 2 snd
snd_page_alloc 11304 2 snd_hda_intel,snd_pcm
-----

-----
amixer
Simple mixer control 'Headphone',0
  Capabilities: pvolume pswitch
  Playback channels: Front Left - Front Right
  Limits: Playback 0 - 64
  Mono:
  Front Left: Playback 64 [100%] [on]
  Front Right: Playback 64 [100%] [on]
Simple mixer control 'PCM',0
  Capabilities: pvolume
  Playback channels: Front Left - Front Right
  Limits: Playback 0 - 255
  Mono:
  Front Left: Playback 245 [96%]
  Front Right: Playback 245 [96%]
Simple mixer control 'Front',0
  Capabilities: pvolume pswitch
  Playback channels: Front Left - Front Right
  Limits: Playback 0 - 64
  Mono:
  Front Left: Playback 61 [95%] [on]
  Front Right: Playback 61 [95%] [on]
Simple mixer control 'Front Mic',0
  Capabilities: pvolume pswitch
  Playback channels: Front Left - Front Right
  Limits: Playback 0 - 65
  Mono:
  Front Left: Playback 64 [98%] [on]
  Front Right: Playback 64 [98%] [on]
Simple mixer control 'Line',0
  Capabilities: pvolume pswitch
  Playback channels: Front Left - Front Right
  Limits: Playback 0 - 65
  Mono:
  Front Left: Playback 64 [98%] [on]
  Front Right: Playback 64 [98%] [on]
Simple mixer control 'CD',0
  Capabilities: pvolume pswitch
  Playback channels: Front Left - Front Right
  Limits: Playback 0 - 65
  Mono:
  Front Left: Playback 56 [86%] [on]
  Front Right: Playback 56 [86%] [on]
Simple mixer control 'Mic',0
  Capabilities: pvolume pswitch
  Playback channels: Front Left - Front Right
  Limits: Playback 0 - 65
  Mono:
  Front Left: Playback 64 [98%] [on]
  Front Right: Playback 64 [98%] [on]
Simple mixer control 'Mono',0
  Capabilities: pvolume pvolume-joined pswitch pswitch-joined
  Playback channels: Mono
  Limits: Playback 0 - 65
  Mono: Playback 62 [95%] [on]
Simple mixer control 'Capture',0
  Capabilities: cvolume cswitch
  Capture channels: Front Left - Front Right
  Limits: Capture 0 - 35
  Front Left: Capture 34 [97%] [on]
  Front Right: Capture 34 [97%] [on]
Simple mixer control 'Input Source',0
  Capabilities: enum
  Items: 'Mic' 'Front Mic' 'Line' 'CD'
  Item0: 'Front Mic'
-----
...

[C. Cooke (ccooke)]

No improvement on the ICH6, either.

Why has this bug been set to Rejected?

[Martin Pitt (pitti)]

It was not rejected, the linux-source-2.6.15 task has been marked as 'Fix released'.

[Martin Pitt (pitti)]

reopening since not yet fixed according to reporter.

[Václav Šmilauer (eudoxos)]

IMHO you should open separate bugreport, since for some people the bug has been fixed (including me, the original reporter) -- perhaps there is different kind of problems for different models. I definitely did not say it was nor yet fixed; the ALSA bug that was being tracked by this one is closed, too.

This bug has grown too long to be useful other people, anyway. (not to speak of the horrible malone interface that does not even allow threads, collapsing them, quoting, has everything in one narrow column in the middle etc.)

[Pazkooda (pazkooda)]

Hi,

I just want to confirm that this bug still exists, but I have a few words:

My hardware is HP dc7600 USD with Intel HDA (ALC260).

With Breezy I've got sound on headphones (front) and only there.
When I've ran Dapper Beta2 Live CD (booted from CD) I've had sound everywhere (everything seems to be fine).
When I upgrade installed version of Breezy (final) -> Dapper (final) sound gone away totally (sound card is detected, mixers shows volume controls etc. - but no sound at all).
So there is a way to enable it and it should be fixed.

PS. I've also tried to recompile alsa-{driver|lib|util} to 1.0.11. I doesn't help.

[kmARC (korondi-mark)]

Hello All,

I've also experienced this bug, but on my laptop the card _do_ sotimes works. I don't know why, it sometimes works, other times doesn't. All of the modules are always loaded:

kmarc@gilgamesh:~$ lsmod | grep snd
snd_hda_intel 20468 2
snd_hda_codec 163088 1 snd_hda_intel
snd_pcm_oss 56448 0
snd_mixer_oss 20544 1 snd_pcm_oss
snd_pcm 96708 3 snd_hda_intel,snd_hda_codec,snd_pcm_oss
snd_timer 26884 1 snd_pcm
snd 60004 10 snd_hda_intel,snd_hda_codec,snd_pcm_oss,snd_mixer_oss,snd_pcm,snd_timer
soundcore 10784 1 snd
snd_page_alloc 11304 2 snd_hda_intel,snd_pcm

[Christian Kellner (gicmo)]

I seem to have the same/very similar problem. I am using edgy's kernel version 2.6.17-10 with alsa 1.0.12rc1. The hardware is a very new MacPro (new enough so hal-device-manager shows tons of "unkown" strings).

--snip --

00:1b.0 Audio device: Intel Corporation Enterprise Southbridge High Definition
Audio (rev 09)
        Flags: bus master, fast devsel, latency 0, IRQ 225
        Memory at 52f00000 (64-bit, non-prefetchable) [size=16K]
        Capabilities: [50] Power Management version 2
        Capabilities: [60] Message Signalled Interrupts: 64bit+ Queue=0/0 Enable-
        Capabilities: [70] Express Unknown type IRQ 0
----
gicmo@photon /proc/asound % cat cards
 0 [Intel ]: HDA-Intel - HDA Intel
                      HDA Intel at 0x52f00000 irq 225
---

gicmo@photon /proc/asound/Intel % cat codec\#0
Codec: Realtek ALC885
Address: 0
Vendor Id: 0x10ec0885
Subsystem Id: 0x106b0c00
Revision Id: 0x100103
Default PCM: rates 0x560, bits 0x0e, types 0x1
Default Amp-In caps: N/A
Default Amp-Out caps: N/A
[... (lots of output omitted, if you need it please ask) ..]

---

snd_hda_intel 20116 2
snd_hda_codec 164608 1 snd_hda_intel
snd_pcm_oss 47360 0
snd_mixer_oss 19584 1 snd_pcm_oss
snd_pcm 84612 3 snd_hda_intel,snd_hda_codec,snd_pcm_oss
snd_timer 25348 1 snd_pcm
snd 58372 9 snd_hda_intel,snd_hda_codec,snd_pcm_oss,snd_mixer_oss,snd_pcm,snd_timer
soundcore 11232 1 snd
snd_page_alloc 11400 2 snd_hda_intel,snd_pcm

------
I of course checked mixer settings (unmute and stuff) and I also played a bit around with the model=XXX module parameter, but no luck.

[DasJott (das-jott)]

I tried the new Ubuntu 6.10.
My Soundchip is Intel HDA (ICH6 Family) ALC880.
No sound at all, although it seems to have been recognized. A driver was loaded (seems so in the device manager), a volume icon exists in the upper bar and the mixer volumes are all on 100%. No sound.
Removed Edgy first, will wait for a decent update.

[LarsBjerregaard (lars-rubyglow)]

Sound worked for me in Breezy, but in a completely new install of Dapper - no sound at all. I see many people all over the place struggling with this. After _many_ hours of googling, I found a simple workaround which works for me. It was described on:
https://bugtrack.alsa-project.org/alsa-bug/view.php?id=2038

So basically what I did (as root):
1) Backup /etc/modprobe.d/alsa-base
2) Add to the bottom of /etc/modprobe.d/alsa-base:
options snd-hda-intel model=ref
3) Reboot

Voila! Sound works.

My system:

cat /proc/asound/version gives:
Advanced Linux Sound Architecture Driver Version 1.0.10rc3 (Mon Nov 07 13:30:21 2005 UTC).

lspci -vv | fgrep Audio gives:
0000:00:1b.0 0403: Intel Corporation 82801G (ICH7 Family) High Definition Audio Controller (rev 01)

cat /proc/asound/card0/codec#0 gives:
Codec: SigmaTel STAC9221 A1

The problem seems to be Alsa headaches with STAC9221, but seems like they are fixing it in source upstream.

[Daniel T Chen (crimsun)]

All models using Sigmatel HDA codecs currently have audible sound. The fixes were pushed to kernel-team@ in early October.

[Václav Šmilauer (eudoxos)]

The bug title states clearly that it is ICH6 that was concerned. As such (acer model), it was closed a long time ago in ALSA. I will (try to) close this bug and if someone has troubles with non-ICH6, please be so kind to open a new bugreport with precise specification of the problem. This is NOT a catch-all bug for "sound doesn't work".

[LarsBjerregaard (lars-rubyglow)]

That's great.
I saw some Edgy users having the same problem, but perhaps it's being
(has been?) fixed in updates?

Thanks
/Lars

Daniel T Chen wrote:
> All models using Sigmatel HDA codecs currently have audible sound. The
> fixes were pushed to kernel-team@ in early October.
>

[Andrew Ash (ash211)]

This bug has been reported fixed by the original reporter.

Most other reports seem to be different bugs. Please file new bugs instead of attaching data to this bug report.

Thanks.