[karmic]snd_hda_intel: audio moves to polling mode and starts repeating (Toshiba A100-259/ATI IXP SB4x0 HD/ALC861)

[Nigel Pugh]

Bug observed on karmic alpha-2: related to snd_hda_intel module

uname -a:
Linux nigel-laptop 2.6.30-8-generic #9-Ubuntu SMP Wed Jun 3 15:23:55 UTC 2009 i686 GNU/Linux

Laptop is a Toshiba A100-259 with Realtek ALC861 HD Audio on an ATI chipset (Radeon Xpress 200M)

lspci shows:
00:14.2 Audio device: ATI Technologies Inc IXP SB4x0 High Definition Audio Controller (rev 01

lshw shows:
 *-multimedia
             description: Audio device
             product: IXP SB4x0 High Definition Audio Controller
             vendor: ATI Technologies Inc
             physical id: 14.2
             bus info: pci@0000:00:14.2
             version: 01
             width: 64 bits
             clock: 33MHz
             capabilities: bus_master cap_list
             configuration: driver=HDA Intel latency=64 module=snd_hda_intel

aplay -l shows:
**** List of PLAYBACK Hardware Devices ****
card 0: SB [HDA ATI SB], device 0: ALC861 Analog [ALC861 Analog]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
card 0: SB [HDA ATI SB], device 6: Si3054 Modem [Si3054 Modem]
  Subdevices: 1/1
  Subdevice #0: subdevice #0

dmesg extracts:

At boot time it shows:

[ 11.379095] HDA Intel 0000:00:14.2: PCI INT A -> GSI 16 (level, low) -> IRQ 16
[ 11.499983] hda_codec: Unknown model for ALC861, trying auto-probe from BIOS...
[ 11.500116] input: HDA Digital PCBeep as /devices/pci0000:00/0000:00:14.2/input/input9
...snip...
[ 137.069706] Too big adjustment 32
[ 137.086909] Too big adjustment 32
[ 137.134515] Too big adjustment 32

About half the time this line is also in the dmesg on boot:

hda-intel: IRQ timing workaround is activated for card #0. Suggest a bigger bdl_pos_adj.

Sound works ok for a while then you get:

hda_intel: azx_get_response timeout, switching to polling mode: last cmd=0x001f0500

After that all sound just loops/repeats continually. A reboot is required to bring it back.

ProblemType: Bug
Architecture: i386
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC0: nigel 3951 F.... mixer_applet2
Card0.Amixer.info:
 Card hw:0 'SB'/'HDA ATI SB at 0xc0000000 irq 16'
   Mixer name : 'Realtek ALC861'
   Components : 'HDA:11c13026,11790001,00100700 HDA:10ec0861,1179ff10,00100300'
   Controls : 14
   Simple ctrls : 10
Date: Tue Jun 23 14:00:25 2009
DistroRelease: Ubuntu 9.10
Package: alsa-base 1.0.20+dfsg-1ubuntu4
PackageArchitecture: all
ProcEnviron:
 LANG=en_GB.UTF-8
 SHELL=/bin/bash
ProcVersionSignature: Ubuntu 2.6.30-8.9-generic
SourcePackage: alsa-driver
Uname: Linux 2.6.30-8-generic i686

[Daniel T Chen (crimsun)]

On Tue, Jun 23, 2009 at 9:20 AM, Nigel Pugh<email address hidden> wrote:
> [  137.069706] Too big adjustment 32

Red herring.

> hda-intel: IRQ timing workaround is activated for card #0. Suggest a
> bigger bdl_pos_adj.

Won't help in this case.

Instead, try enable_msi=1.

[Nigel Pugh (nigel-pugh)]

Thanks for the suggestion!

Ok, tried that option and the sound looping/repeating starts immediately at login.

These lines now in the dmesg at boot:
[ 46.152017] hda_intel: No response from codec, disabling MSI: last cmd=0x001f0500
[ 46.152086] hda-intel: IRQ timing workaround is activated for card #0. Suggest a bigger bdl_pos_adj.

...everything else the same.

[Nigel Pugh (nigel-pugh)]

Apologies, there was one more (possibly relevant) dmesg line I missed from my previous comment:
[ 11.165206] HDA Intel 0000:00:14.2: irq 24 for MSI/MSI-X

A random thought occurred; I know very little about this sort of thing and so this is possibly a complete red herring:

[ 1.832519] ata2: SATA max UDMA/100 mmio m512@0xc0004000 tf 0xc00040c0 irq 22
[ 1.833418] alloc irq_desc for 16 on cpu 0 node 0
..snip...
[ 1.833431] pata_atiixp 0000:00:14.1: PCI INT A -> GSI 16 (level, low) -> IRQ 16

Could these lines in the dmesg be indicating that there is some unwelcome interrupt sharing going on? It looks (to the untrained eye) like some of the IDE or SATA is being mapped to IRQ16 along with the audio.

The reason I mention it is that I also have a bug open for intermittent timeout issues on the SATA controller:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/382812

As I say, probably not relevant, but mentioned in the interests of completeness.

[Nigel Pugh (nigel-pugh)]

I may have a fix... it appears this laptop needs this in /etc/modprobe.d/alsa-base.conf

# Power down HDA controllers after 10 idle seconds
options snd-hda-intel power_save=10 probe_mask=8 enable_msi=1 model=auto

lets see how it goes!

[Nigel Pugh (nigel-pugh)]

ok, I was too optimistic... the looping/repeating still happens.

Every few boots, or after using sound for about 5 minutes the looping returns. reboot required to recover.

[arekkusu (arekkusu-r)]

Running Ubuntu 9.04 with all updates. Laptop is a Toshiba A100-742.
Same Chipset and audio card (attached lspci) and exact same problem.

Audio work for some time and then stop working. Sometimes audio stop working after just a few second from the startup sound, sometimes I can watch something for 1-2 or more hour before it goes down.
The audio loop

Audio did work properly on ubuntu 8.04, didn't work well on ubuntu 8.10 either.

A bit off topic but this laptop is a real nightmare on Linux... wired network fail after heavy transfer, USB is slow and fail (the chipset is notorious for it's slow USB even on WIndows, so it's partially an hardware problem, still USB that stop completely working on Linux is bad)

[non_j (tchambalaia)]

Hi.

Im running the 9.10 on an Asus a6rp and I have the almost the same problem.
I say almost because my audio simply doesn't work.

00:14.2 Audio device: ATI Technologies Inc IXP SB4x0 High Definition Audio Controller (rev 01)

Linux rteixeira-laptop 2.6.31-8-generic #28-Ubuntu SMP Thu Aug 27 14:43:30 UTC 2009 i686 GNU/Linux

**** List of PLAYBACK Hardware Devices ****
card 0: SB [HDA ATI SB], device 0: ALC660 Analog [ALC660 Analog]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
card 0: SB [HDA ATI SB], device 1: ALC660 Digital [ALC660 Digital]
  Subdevices: 0/1
  Subdevice #0: subdevice #0
card 0: SB [HDA ATI SB], device 6: Si3054 Modem [Si3054 Modem]
  Subdevices: 1/1
  Subdevice #0: subdevice #0

my dmesg also shows the same messages as from the original post of this bug

[waspbr (notaduck)]

bug also present in the hp tx1320us with the latest karmic (alpha 6)

here is the output of aplay -l

**** List of PLAYBACK Hardware Devices ****
card 0: NVidia [HDA NVidia], device 0: ALC861-VD Analog [ALC861-VD Analog]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
card 0: NVidia [HDA NVidia], device 1: ALC861-VD Digital [ALC861-VD Digital]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
card 0: NVidia [HDA NVidia], device 6: Si3054 Modem [Si3054 Modem]
  Subdevices: 0/1
  Subdevice #0: subdevice #0

and relevant line of lspci

00:10.1 Audio device: nVidia Corporation MCP51 High Definition Audio (rev a2)

the device is not listed/detected by pulseaudio

[JK (m0d)]

I've filed a bug report today with a similar problem:

https://bugs.launchpad.net/ubuntu/+source/alsa-driver/+bug/434511

Although playback works for me most of the time (sometimes I had only background buzz and no sound at all), the mic only works when I'm recording from within a live session (using the 9.04 and 9.10 CDs). Maybe you could try that too and see if it makes a difference...

[JK (m0d)]

The "azx_get_response timeout" message has been seen on other hardware too, but with different symptoms:
https://answers.launchpad.net/ubuntu/+source/alsa-driver/+question/81561

[JK (m0d)]

There's one more thing (sorry for the triple-posting): I can confirm that playback worked flawlessly with Ubuntu 8.04 (though the mic did not), so this is a regression...

[Daniel T Chen (crimsun)]

On Tue, Sep 22, 2009 at 9:18 AM, JK <email address hidden> wrote:
> The "azx_get_response timeout" message has been seen on other hardware too, but with different symptoms:

That message (switching to polling) is not necessarily a critical
error. However, switching to single_cmd generally is. Essentially, we
need to figure out which codec slots and which interrupt mode to use.

[JK (m0d)]

Since about an hour I have a new problem: no sound at all! It happened before I had the chance to change any options. Now I've appended "model=toshiba" (as suggested elsewhere) but it didn't help. However, I noticed that alsamixer doesn't show a control for the "Master" device (only PCM, Mic and Beep). And it also doesn't show any "Capture" devices (this is independent of the "model" option) although they seem to exist:

andi@andicompi:~$ cat /proc/asound/devices
  2: : timer
  3: : sequencer
  4: [ 0- 0]: digital audio playback
  5: [ 0- 0]: digital audio capture
  6: [ 0- 6]: digital audio playback
  7: [ 0- 6]: digital audio capture
  8: [ 0] : control

andi@andicompi:~$ cat /proc/asound/cards
 0 [SB ]: HDA-Intel - HDA ATI SB
                      HDA ATI SB at 0xd0000000 irq 16

And I also noticed two devices called "Caller1" and "Off-hook" that remind me of a modem. There is indeed a modem built into this laptop and "cat /proc/asound/card0/codec#0" shows this:

Codec: LSI Si3054
Address: 0
Function Id: 0x2
Vendor Id: 0x11c13026
Subsystem Id: 0x11790001
Revision Id: 0x100700
Modem Function Group: 0x1

The ALC861 is in "codec#3".

@Daniel : I also have that "switching to single_cmd" error in my dmesg now. Is there an option to choose the codec slot or interrupt? I searched the ALSA documentation but found only the "index" and "probe_mask" options that seem to be related to this problem. If you'd tell me which options I should try I would post my results! Currently, my alsa-base.conf looks like this:

# Prevent abnormal drivers from grabbing index 0
options bt87x index=-2
options cx88_alsa index=-2
options saa7134-alsa index=-2
options snd-atiixp-modem index=-2
options snd-intel8x0m index=-2
options snd-via82xx-modem index=-2
options snd-usb-audio index=-2
options snd-usb-us122l index=-2
options snd-usb-usx2y index=-2
options snd-usb-caiaq index=-2
# Ubuntu #62691, enable MPU for snd-cmipci
options snd-cmipci mpu_port=0x330 fm_port=0x388
# Keep snd-pcsp from being loaded as first soundcard
options snd-pcsp index=-2
# Power down HDA controllers after 10 idle seconds
options snd-hda-intel power_save=10
options snd-hda-intel model=toshiba

[JK (m0d)]

I found a great guide on the HDA codec:
http://www.kernel.org/pub/linux/kernel/people/tiwai/docs/HD-Audio.html#_codec_probing_problem

After reading it I realised that I experienced a "Codec-Probing Problem": the modem was asigned to the first codec slot while the "real" codec was assigned to the fourth slot. Therefore, I added "probe_mask=8" as an option and now I have sound! Furthermore, the devices in alsa-mixer are correct now: I have a capture device and can select Mic or CD as Input-Source.

However, one problem remains: recording doesn't work! When recording with audacity, the time-slider doesn't move forward but is constantly flickering. And when using GNOME's sound recorder, recording stops immediately after starting it and the app crashes on exit each time. PA's "volume meter" for recording doesn't show anything either. I'm running out of ideas right now...

One note: I had to remove the "model=toshiba" option. When it's enabled, the devices are also messed up within alsa-mixer (although in a different way than without probe_mask=8) and sound doesn't work!

I hope that the "probe_mask" option will help you too! To check if you have a codec-probing problem, look at "/proc/asound/card0/codec* and see if the modem has a lower slot than the soundcard.

Regards

[JK (m0d)]

Forgot to mention sth : the "position_fix" option may help with the sound loop problem (see the guide).

[JK (m0d)]

I managed to reactivate my mic by using the 'position_fix=1' option (in addition to the 'probe_mask=8' option). These are the only two options I've added! I've tried some more (like 'model=toshiba') but that made it worse again! Maybe this info is useful for you...

[rizlo (davide-molteni)]

https://bugs.launchpad.net/ubuntu/+source/alsa-driver/+bug/318795
https://bugs.launchpad.net/ubuntu/+source/alsa-driver/+bug/434511
https://bugs.launchpad.net/ubuntu/+source/alsa-driver/+bug/458563
https://bugs.launchpad.net/ubuntu/+source/alsa-driver/+bug/471687

and 391114 are the same bug

-The "no sound at all" problem is caused by enabling the restricted driver for modem, probe_mask=8 solves it, anyway if modem driver is not enabled sound works without this option.

-The audio makes loops and no mic. position_fix=1 makes the mic work (434511) but audio still loops

I'm affected on a Toshiba A100-259

00:14.2 Audio device: ATI Technologies Inc IXP SB4x0 High Definition Audio Controller (rev 01)

I think we should mark other bugs as duplicated.
I think also that it needs a certain relevance since loops makes any audio application useless.

[rizlo (davide-molteni)]

sl-modem-daemon is the modem driver that conflicts with audio

[Nigel Pugh (nigel-pugh)]

under karmic final this line in alsa-base.conf now fixes it for me:

options snd-hda-intel probe_mask=8 model=auto enable_msi=1

The login sound is "overdriven" for a couple of seconds, but quickly corrects itself.

[JK (m0d)]

>The "no sound at all" problem is caused by enabling the restricted driver for modem, probe_mask=8 solves it, anyway if modem driver is not enabled sound works without this option.

This is only partly true. While the probe_mask option reactivates sound (and disables the modem) when karmic is installed, both sound and modem work within a live session! At least that was the case for me (on a Toshiba A100) and for another guy in a related bug report. It seems that something gets broken during installation! Disabling the sl-modem driver did not help in my case either. And it's not only the sound, but also recording that works perfectly within a live session and stops working after installation. Furthermore, using position_fix=1 to (re-)enable the mic works but the quality is not nearly as good as within the live session...

@rizlo : could you please check if sound and recording works for you within a karmic live session? Since I'm not the owner of the laptop, I've got currently no access to it, but I've already posted information (collected with the also-info script) about the live and installed system in my own bug report at:

https://bugs.launchpad.net/ubuntu/+source/alsa-driver/+bug/434511

Regards

JK

[rizlo (davide-molteni)]

@Nigel Pugh Are you sure?

options snd-hda-intel probe_mask=8 model=auto enable_msi=1

don't fixes it for me! sound still loops... also no mic even if I add position_fix=1

[JK (m0d)]

It seems like the alsa options have very different effects, even on identical systems. In my case, appending "model=toshiba" (which should be the best, imho) caused only silence and appending "enable_msi=1" broke the sound completely (i.e. there was only infinite noise/buzz, very annoying). I hope I will still be alive on that day when sound on linux "just works"...

[Daniel T Chen (crimsun)]

On Sat, Nov 7, 2009 at 4:30 AM, JK <email address hidden> wrote:
> It seems like the alsa options have very different effects, even on
> identical systems. In my case, appending "model=toshiba" (which should

You need a pretty thorough understanding of bios and linux at least.
Different hardware are often rebadged, so you can *never* assume
identical model numbers (marketing, anyone?) correspond to identical
hardware.

[JK (m0d)]

> You need a pretty thorough understanding of bios and linux at least.

Maybe even more than that. Sometimes it seems that you actually have to be either an alsa, kernel and bios developer to really understand what's going wrong with your sound. Actually, I have a pretty good understanding of linux imho (I'm a full-time developer with some experience on writing device drivers, though rarely doing kernel hacking), but I must admit that I don't know much about sound drivers and obviously wasn't able to fix the sound myself. And that's exactly the problem, because this is even harder for the famous "average joe"...

[Daniel T Chen (crimsun)]

@Nigel Do you have linux-backports-modules-alsa-karmic-generic
installed? Can you verify that you still need your probe_mask and
other parameter additions to snd-hda-intel in
/etc/modprobe.d/alsa-base.conf?

[Nigel Pugh (nigel-pugh)]

@Daniel,
That package wasn't installed.

Am now tracking lucid(2.6.32-2-generic and libasound2 1.0.21a-1ubuntu1)
If I remove the line in alsa-base.conf the looping behaviour reliably appears after a short while.
The dmesg shows:
[84644.363779] HDA Intel 0000:00:14.2: PCI INT A disabled
[84644.720039] HDA Intel 0000:00:14.2: PCI INT A -> GSI 16 (level, low) -> IRQ 16
[84644.857625] hda_codec: ALC861: BIOS auto-probing.
[84644.858887] input: HDA Digital PCBeep as /devices/pci0000:00/0000:00:14.2/input/input9
[85114.760019] hda-intel: azx_get_response timeout, switching to polling mode: last cmd=0x001f0500

I booted back to the karmic kernel(2.6.31-14-generic) and installed the package you mentioned:
[ 190.323592] HDA Intel 0000:00:14.2: PCI INT A disabled
[ 190.538290] HDA Intel 0000:00:14.2: PCI INT A -> GSI 16 (level, low) -> IRQ 16
[ 190.538385] HDA Intel 0000:00:14.2: irq 24 for MSI/MSI-X
[ 190.664379] hda_codec: ALC861: BIOS auto-probing.
[ 190.665210] input: HDA Digital PCBeep as /devices/pci0000:00/0000:00:14.2/input/input9

So it appears to enable MSI automatically without the line in alsa-base.conf
Sound works just fine. Nice job!

@rizlo
Sorry to hear it doesn't work for you.

My bios version is 2.30. This is the latest version I found on the toshiba website.

[Daniel Bruckhaus (th3-b0b)]

Hi!

I'm running Kubuntu Karmic on an ASUS X51R:
uname -a:
 Linux daniel-laptop 2.6.31-16-generic #53-Ubuntu SMP Tue Dec 8 04:01:29 UTC 2009 i686 GNU/Linux

lshw -class multimedia:
   *-multimedia
       description: Audio device
       product: SBx00 Azalia (Intel HDA)
       vendor: ATI Technologies Inc
       physical id: 14.2
       bus info: pci@0000:00:14.2
       version: 00
       width: 64 bits
       clock: 33MHz
       capabilities: pm bus_master cap_list
       configuration: driver=HDA Intel latency=64
       resources: irq:16 memory:febf4000-febf7fff

dmesg (Excerpt):
...
[ 42.602126] HDA Intel 0000:00:14.2: PCI INT A -> GSI 16 (level, low) -> IRQ 16
[ 42.695146] input: HDA Digital PCBeep as /devices/pci0000:00/0000:00:14.2/input/input10
...
[33172.369014] hda-intel: azx_get_response timeout, switching to polling mode: last cmd=0x10170500
[33173.373016] hda_intel: azx_get_response timeout, switching to single_cmd mode: last cmd=0x10170500
[33173.373108] hda-intel: spurious response 0x0:0x0, last cmd=0x970600
[33173.373112] hda-intel: spurious response 0x0:0x0, last cmd=0x970600
[33173.373115] hda-intel: spurious response 0x0:0x0, last cmd=0x970600
...
[35645.095202] hda-intel: Too big adjustment 32
[35645.116719] hda-intel: Too big adjustment 32
[35645.140989] hda-intel: Too big adjustment 32
[35645.167113] hda-intel: Too big adjustment 32
...

Audio Output is working perfectly fine, only Input is broken.
The "..spurious response..." and ".... Too big adjustment..." messages begin when trying to record from within audcity.
They Don't start popping up when running skype.

I can, however, hear myself in the speaker if I turn up the Output level of the Mic in KMix, so the Mic is woring, and sound is actually passing through the system somehow... ;-)

Any hints? Or any further info you need?

[Brad Figg (brad-figg)]

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 391114

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.

[William (manbill)]

Architecture: i386
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC0: santiago 1913 F.... pulseaudio
 /dev/snd/pcmC0D0p: santiago 1913 F...m pulseaudio
Card0.Amixer.info:
 Card hw:0 'Intel'/'HDA Intel at 0xdc400000 irq 22'
   Mixer name : 'Realtek ALC861'
   Components : 'HDA:10ec0861,11791205,00100300 HDA:11c13026,11790001,00100700'
   Controls : 16
   Simple ctrls : 11
DistroRelease: Ubuntu 9.10
InstallationMedia: Ubuntu 9.10 "Karmic Koala" - Release i386 (20091028.5)
Package: alsa-base 1.0.20+dfsg-1ubuntu5
PackageArchitecture: all
ProcEnviron:
 SHELL=/bin/bash
 LANG=es_ES.UTF-8
ProcVersionSignature: Ubuntu 2.6.31-21.59-generic
Uname: Linux 2.6.31-21-generic i686
UserGroups: adm admin cdrom dialout lpadmin plugdev sambashare

[Brad Figg (brad-figg)]

Hi Nigel,

If you could also please 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.

    [This is an automated message. Apologies if it has reached you inappropriately; please just reply to this message indicating so.]

[rizlo (davide-molteni)]

The audio repeating is solved on lucid beta1 but the mic input is still not working. As other users reported mic works when system is running live from cd. I think there is another bug opened for this issue.

[rizlo (davide-molteni)]

mic works on lucid adding "position_fix=1" in alsa-base.conf

[Brad Figg (brad-figg)]

This bug report was marked as Incomplete and has not had any updated comments for quite some time. As a result this bug is being closed. Please reopen if this is still an issue in the current Ubuntu release http://www.ubuntu.com/getubuntu/download . Also, please be sure to provide any requested information that may have been missing. To reopen the bug, click on the current status under the Status column and change the status back to "New". Thanks.

[This is an automated message. Apologies if it has reached you inappropriately; please just reply to this message indicating so.]