Bug 48381 - [Regression][Bisected]Sound gets permanently muted a few minutes after booting

Bug 48381 - [Regression][Bisected]Sound gets permanently muted a few minutes after booting

[Da Fox]

Hi all,

Since the official bug-tracker at
https://bugtrack.alsa-project.org/alsa-bug/ seems to be down (and
moreover this appears to be a kernel bug), I have filed a bug at the
kernel's bug-tracker last week (
https://bugzilla.kernel.org/show_bug.cgi?id=48381 ). However, it seems
to have gone unnoticed thus-far.

Quick summary of the issue:
After a while of inactivity the sound-card seems to go into a sort of
power-saving mode (you can hear a faint 'pop' from the speakers).
Before the sound would automatically turn back on again as soon as a
program attempted to play sound. However now it seems that after
resuming from the power-saving mode the driver thinks that there is
always a headphone connected, even when it is not. I think this is
what happens because toggling the 'Auto-Mute Mode' (setting it to
'Disabled') in alsamixer allows the speakers to produce sound again.
This is all on a laptop (please see the linked bug report for more
detailed hardware description).

Would a developer please take a look at it? I have bisected the issue
down to a commit made by David Henningsson and/or Takashi Iwai:
---8<---------
80c8bfbe76869bfd6bdf3d260d316e7a32f318c3 is the first bad commit
commit 80c8bfbe76869bfd6bdf3d260d316e7a32f318c3
Author: David Henningsson <david.henningsson@canonical.com>
Date:   Mon Jun 4 09:33:51 2012 +0200

    ALSA: HDA: Create phantom jacks for fixed inputs and outputs

    PulseAudio sometimes have difficulties knowing that there is a
    "Speaker" or "Internal Mic", if they have no individual volume
    controls or selectors. As a result, only e g "Headphone" might
    be created for a laptop, but no "Speaker".
    To help out, create phantom jacks (that are always present,
    at least for now) for "Speaker", "Internal Mic" etc, in case we
    detect them.
    The naming convention is e g "Speaker Phantom Jack".

    In order not to pollute the /dev/input namespace with even more
    devices, these are added to the kcontrols only, not the input devices.

    Signed-off-by: David Henningsson <david.henningsson@canonical.com>
    Signed-off-by: Takashi Iwai <tiwai@suse.de>

:040000 040000 7a954b731bef80cee731763521f64a424a7058a0
7d176b02ebb4553b0b0f97d200f7527cb69c122a M    sound
--->8---------

If any additional information is required please let me know.

Best regards,
  Da Fox

Re: Bug 48381 - [Regression][Bisected]Sound gets permanently muted a few minutes after booting

[David Henningsson]

On 10/10/2012 02:59 PM, Da Fox wrote:
> Hi all,
>
> Since the official bug-tracker at
> https://bugtrack.alsa-project.org/alsa-bug/ seems to be down (and
> moreover this appears to be a kernel bug), I have filed a bug at the
> kernel's bug-tracker last week (
> https://bugzilla.kernel.org/show_bug.cgi?id=48381 ). However, it seems
> to have gone unnoticed thus-far.
>
> Quick summary of the issue:
> After a while of inactivity the sound-card seems to go into a sort of
> power-saving mode (you can hear a faint 'pop' from the speakers).
> Before the sound would automatically turn back on again as soon as a
> program attempted to play sound. However now it seems that after
> resuming from the power-saving mode the driver thinks that there is
> always a headphone connected, even when it is not. I think this is
> what happens because toggling the 'Auto-Mute Mode' (setting it to
> 'Disabled') in alsamixer allows the speakers to produce sound again.
> This is all on a laptop (please see the linked bug report for more
> detailed hardware description).
>
> Would a developer please take a look at it? I have bisected the issue
> down to a commit made by David Henningsson and/or Takashi Iwai:
> ---8<---------
> 80c8bfbe76869bfd6bdf3d260d316e7a32f318c3 is the first bad commit
> commit 80c8bfbe76869bfd6bdf3d260d316e7a32f318c3
> Author: David Henningsson <david.henningsson@canonical.com>
> Date:   Mon Jun 4 09:33:51 2012 +0200
>
>      ALSA: HDA: Create phantom jacks for fixed inputs and outputs
>
>      PulseAudio sometimes have difficulties knowing that there is a
>      "Speaker" or "Internal Mic", if they have no individual volume
>      controls or selectors. As a result, only e g "Headphone" might
>      be created for a laptop, but no "Speaker".
>      To help out, create phantom jacks (that are always present,
>      at least for now) for "Speaker", "Internal Mic" etc, in case we
>      detect them.
>      The naming convention is e g "Speaker Phantom Jack".
>
>      In order not to pollute the /dev/input namespace with even more
>      devices, these are added to the kcontrols only, not the input devices.
>
>      Signed-off-by: David Henningsson <david.henningsson@canonical.com>
>      Signed-off-by: Takashi Iwai <tiwai@suse.de>
>
> :040000 040000 7a954b731bef80cee731763521f64a424a7058a0
> 7d176b02ebb4553b0b0f97d200f7527cb69c122a M    sound
> --->8---------
>
> If any additional information is required please let me know.

Thanks for the heads up. I would need the following additional information:

1) One alsa-info taken during playback through internal speakers, when 
the internal speakers are working, and
2) One alsa-info taken during playback through internal speakers, when 
the internal speakers are not working.

(See https://wiki.ubuntu.com/Audio/AlsaInfo if you don't know what 
alsa-info is.)

If you're using PulseAudio for playback, please also include the output 
of "pacmd list" in the above two scenarios.


-- 
David Henningsson, Canonical Ltd.
https://launchpad.net/~diwic

Re: Bug 48381 - [Regression][Bisected]Sound gets permanently muted a few minutes after booting

[Da Fox]

Dear Ben,

I mostly use mplayer to play audio and video files, using the 'alsa'
output driver.
I have attached the requested information to the bugtracker in full,
the following is a quick diff of the two files:

---8<---------
--- alsa-info_fresh-boot-speakers-working.txt	2012-10-10
15:31:13.787442105 +0200
+++ alsa-info_after-a-minute-speakers-NOT-working.txt	2012-10-10
15:32:29.395605762 +0200
@@ -3,7 +3,7 @@
 !!ALSA Information Script v 0.4.61
 !!################################

-!!Script ran on: Wed Oct 10 13:31:13 UTC 2012
+!!Script ran on: Wed Oct 10 13:32:28 UTC 2012


 !!Linux Distribution
@@ -275,7 +275,7 @@
     Conn = ATAPI, Color = Unknown
     DefAssociation = 0x1, Sequence = 0x0
     Misc = NO_PRESENCE
-  Pin-ctls: 0x40: OUT
+  Pin-ctls: 0x00:
   Unsolicited: tag=00, enabled=0
   Power states:  D0 D1 D2 D3 EPSS
   Power: setting=D0, actual=D0
@@ -530,7 +530,7 @@

 crw-rw---- 1 root audio 116,  6 Oct 10  2012 /dev/snd/controlC0
 crw-rw---- 1 root audio 116,  5 Oct 10  2012 /dev/snd/pcmC0D0c
-crw-rw---- 1 root audio 116,  4 Oct 10 15:30 /dev/snd/pcmC0D0p
+crw-rw---- 1 root audio 116,  4 Oct 10 15:32 /dev/snd/pcmC0D0p
 crw-rw---- 1 root audio 116,  3 Oct 10  2012 /dev/snd/pcmC0D1p
 crw-rw---- 1 root audio 116,  2 Oct 10  2012 /dev/snd/pcmC0D3p
 crw-rw---- 1 root audio 116,  1 Oct 10  2012 /dev/snd/seq
@@ -550,7 +550,7 @@

 **** List of PLAYBACK Hardware Devices ****
 card 0: PCH [HDA Intel PCH], device 0: ALC665 Analog [ALC665 Analog]
-  Subdevices: 0/1
+  Subdevices: 1/1
   Subdevice #0: subdevice #0
 card 0: PCH [HDA Intel PCH], device 1: ALC665 Digital [ALC665 Digital]
   Subdevices: 1/1
--->8---------

Best regards,
  Da Fox

On Wed, Oct 10, 2012 at 3:20 PM, David Henningsson
<david.henningsson@canonical.com> wrote:
> On 10/10/2012 02:59 PM, Da Fox wrote:
>>
>> Hi all,
>>
>> Since the official bug-tracker at
>> https://bugtrack.alsa-project.org/alsa-bug/ seems to be down (and
>> moreover this appears to be a kernel bug), I have filed a bug at the
>> kernel's bug-tracker last week (
>> https://bugzilla.kernel.org/show_bug.cgi?id=48381 ). However, it seems
>> to have gone unnoticed thus-far.
>>
>> Quick summary of the issue:
>> After a while of inactivity the sound-card seems to go into a sort of
>> power-saving mode (you can hear a faint 'pop' from the speakers).
>> Before the sound would automatically turn back on again as soon as a
>> program attempted to play sound. However now it seems that after
>> resuming from the power-saving mode the driver thinks that there is
>> always a headphone connected, even when it is not. I think this is
>> what happens because toggling the 'Auto-Mute Mode' (setting it to
>> 'Disabled') in alsamixer allows the speakers to produce sound again.
>> This is all on a laptop (please see the linked bug report for more
>> detailed hardware description).
>>
>> Would a developer please take a look at it? I have bisected the issue
>> down to a commit made by David Henningsson and/or Takashi Iwai:
>> ---8<---------
>> 80c8bfbe76869bfd6bdf3d260d316e7a32f318c3 is the first bad commit
>> commit 80c8bfbe76869bfd6bdf3d260d316e7a32f318c3
>> Author: David Henningsson <david.henningsson@canonical.com>
>> Date:   Mon Jun 4 09:33:51 2012 +0200
>>
>>      ALSA: HDA: Create phantom jacks for fixed inputs and outputs
>>
>>      PulseAudio sometimes have difficulties knowing that there is a
>>      "Speaker" or "Internal Mic", if they have no individual volume
>>      controls or selectors. As a result, only e g "Headphone" might
>>      be created for a laptop, but no "Speaker".
>>      To help out, create phantom jacks (that are always present,
>>      at least for now) for "Speaker", "Internal Mic" etc, in case we
>>      detect them.
>>      The naming convention is e g "Speaker Phantom Jack".
>>
>>      In order not to pollute the /dev/input namespace with even more
>>      devices, these are added to the kcontrols only, not the input
>> devices.
>>
>>      Signed-off-by: David Henningsson <david.henningsson@canonical.com>
>>      Signed-off-by: Takashi Iwai <tiwai@suse.de>
>>
>> :040000 040000 7a954b731bef80cee731763521f64a424a7058a0
>> 7d176b02ebb4553b0b0f97d200f7527cb69c122a M    sound
>> --->8---------
>>
>> If any additional information is required please let me know.
>
>
> Thanks for the heads up. I would need the following additional information:
>
> 1) One alsa-info taken during playback through internal speakers, when the
> internal speakers are working, and
> 2) One alsa-info taken during playback through internal speakers, when the
> internal speakers are not working.
>
> (See https://wiki.ubuntu.com/Audio/AlsaInfo if you don't know what alsa-info
> is.)
>
> If you're using PulseAudio for playback, please also include the output of
> "pacmd list" in the above two scenarios.
>
>
> --
> David Henningsson, Canonical Ltd.
> https://launchpad.net/~diwic

Re: Bug 48381 - [Regression][Bisected]Sound gets permanently muted a few minutes after booting

[Takashi Iwai]

At Wed, 10 Oct 2012 14:59:08 +0200,
Da Fox wrote:
> 
> Hi all,
> 
> Since the official bug-tracker at
> https://bugtrack.alsa-project.org/alsa-bug/ seems to be down (and
> moreover this appears to be a kernel bug), I have filed a bug at the
> kernel's bug-tracker last week (
> https://bugzilla.kernel.org/show_bug.cgi?id=48381 ). However, it seems
> to have gone unnoticed thus-far.
> 
> Quick summary of the issue:
> After a while of inactivity the sound-card seems to go into a sort of
> power-saving mode (you can hear a faint 'pop' from the speakers).
> Before the sound would automatically turn back on again as soon as a
> program attempted to play sound. However now it seems that after
> resuming from the power-saving mode the driver thinks that there is
> always a headphone connected, even when it is not. I think this is
> what happens because toggling the 'Auto-Mute Mode' (setting it to
> 'Disabled') in alsamixer allows the speakers to produce sound again.
> This is all on a laptop (please see the linked bug report for more
> detailed hardware description).
> 
> Would a developer please take a look at it? I have bisected the issue
> down to a commit made by David Henningsson and/or Takashi Iwai:
> ---8<---------
> 80c8bfbe76869bfd6bdf3d260d316e7a32f318c3 is the first bad commit
> commit 80c8bfbe76869bfd6bdf3d260d316e7a32f318c3
> Author: David Henningsson <david.henningsson@canonical.com>
> Date:   Mon Jun 4 09:33:51 2012 +0200
> 
>     ALSA: HDA: Create phantom jacks for fixed inputs and outputs

Could you check the patch below whether it fixes your problem?
It essentially disables what the commit above introduced.

If the problem persists with the patch, it means that the commit above
is no real culprit, thus bisection wasn't correct.


Takashi

---
diff --git a/sound/pci/hda/hda_jack.c b/sound/pci/hda/hda_jack.c
index 5c690cb..d3f4ae6 100644
--- a/sound/pci/hda/hda_jack.c
+++ b/sound/pci/hda/hda_jack.c
@@ -359,6 +359,8 @@ static int add_jack_kctl(struct hda_codec *codec, hda_nid_t nid,
 		return 0;
 	phantom_jack = (conn != AC_JACK_PORT_COMPLEX) ||
 		       !is_jack_detectable(codec, nid);
+	if (phantom_jack)
+		return 0;
 
 	snd_hda_get_pin_label(codec, nid, cfg, name, sizeof(name), &idx);
 	if (phantom_jack)

Re: Bug 48381 - [Regression][Bisected]Sound gets permanently muted a few minutes after booting

[David Henningsson]

[-- Attachment #1: Type: text/plain, Size: 5350 bytes --]

On 10/10/2012 03:39 PM, Da Fox wrote:
> Dear Ben,

Actually, it's David :-)

> I mostly use mplayer to play audio and video files, using the 'alsa'
> output driver.
> I have attached the requested information to the bugtracker in full,
> the following is a quick diff of the two files:

After some attempts I think was able to find the culprit. Please test 
the attached patch and see if it resolves the problem.

>
> ---8<---------
> --- alsa-info_fresh-boot-speakers-working.txt	2012-10-10
> 15:31:13.787442105 +0200
> +++ alsa-info_after-a-minute-speakers-NOT-working.txt	2012-10-10
> 15:32:29.395605762 +0200
> @@ -3,7 +3,7 @@
>   !!ALSA Information Script v 0.4.61
>   !!################################
>
> -!!Script ran on: Wed Oct 10 13:31:13 UTC 2012
> +!!Script ran on: Wed Oct 10 13:32:28 UTC 2012
>
>
>   !!Linux Distribution
> @@ -275,7 +275,7 @@
>       Conn = ATAPI, Color = Unknown
>       DefAssociation = 0x1, Sequence = 0x0
>       Misc = NO_PRESENCE
> -  Pin-ctls: 0x40: OUT
> +  Pin-ctls: 0x00:
>     Unsolicited: tag=00, enabled=0
>     Power states:  D0 D1 D2 D3 EPSS
>     Power: setting=D0, actual=D0
> @@ -530,7 +530,7 @@
>
>   crw-rw---- 1 root audio 116,  6 Oct 10  2012 /dev/snd/controlC0
>   crw-rw---- 1 root audio 116,  5 Oct 10  2012 /dev/snd/pcmC0D0c
> -crw-rw---- 1 root audio 116,  4 Oct 10 15:30 /dev/snd/pcmC0D0p
> +crw-rw---- 1 root audio 116,  4 Oct 10 15:32 /dev/snd/pcmC0D0p
>   crw-rw---- 1 root audio 116,  3 Oct 10  2012 /dev/snd/pcmC0D1p
>   crw-rw---- 1 root audio 116,  2 Oct 10  2012 /dev/snd/pcmC0D3p
>   crw-rw---- 1 root audio 116,  1 Oct 10  2012 /dev/snd/seq
> @@ -550,7 +550,7 @@
>
>   **** List of PLAYBACK Hardware Devices ****
>   card 0: PCH [HDA Intel PCH], device 0: ALC665 Analog [ALC665 Analog]
> -  Subdevices: 0/1
> +  Subdevices: 1/1
>     Subdevice #0: subdevice #0
>   card 0: PCH [HDA Intel PCH], device 1: ALC665 Digital [ALC665 Digital]
>     Subdevices: 1/1
> --->8---------
>
> Best regards,
>    Da Fox
>
> On Wed, Oct 10, 2012 at 3:20 PM, David Henningsson
> <david.henningsson@canonical.com> wrote:
>> On 10/10/2012 02:59 PM, Da Fox wrote:
>>>
>>> Hi all,
>>>
>>> Since the official bug-tracker at
>>> https://bugtrack.alsa-project.org/alsa-bug/ seems to be down (and
>>> moreover this appears to be a kernel bug), I have filed a bug at the
>>> kernel's bug-tracker last week (
>>> https://bugzilla.kernel.org/show_bug.cgi?id=48381 ). However, it seems
>>> to have gone unnoticed thus-far.
>>>
>>> Quick summary of the issue:
>>> After a while of inactivity the sound-card seems to go into a sort of
>>> power-saving mode (you can hear a faint 'pop' from the speakers).
>>> Before the sound would automatically turn back on again as soon as a
>>> program attempted to play sound. However now it seems that after
>>> resuming from the power-saving mode the driver thinks that there is
>>> always a headphone connected, even when it is not. I think this is
>>> what happens because toggling the 'Auto-Mute Mode' (setting it to
>>> 'Disabled') in alsamixer allows the speakers to produce sound again.
>>> This is all on a laptop (please see the linked bug report for more
>>> detailed hardware description).
>>>
>>> Would a developer please take a look at it? I have bisected the issue
>>> down to a commit made by David Henningsson and/or Takashi Iwai:
>>> ---8<---------
>>> 80c8bfbe76869bfd6bdf3d260d316e7a32f318c3 is the first bad commit
>>> commit 80c8bfbe76869bfd6bdf3d260d316e7a32f318c3
>>> Author: David Henningsson <david.henningsson@canonical.com>
>>> Date:   Mon Jun 4 09:33:51 2012 +0200
>>>
>>>       ALSA: HDA: Create phantom jacks for fixed inputs and outputs
>>>
>>>       PulseAudio sometimes have difficulties knowing that there is a
>>>       "Speaker" or "Internal Mic", if they have no individual volume
>>>       controls or selectors. As a result, only e g "Headphone" might
>>>       be created for a laptop, but no "Speaker".
>>>       To help out, create phantom jacks (that are always present,
>>>       at least for now) for "Speaker", "Internal Mic" etc, in case we
>>>       detect them.
>>>       The naming convention is e g "Speaker Phantom Jack".
>>>
>>>       In order not to pollute the /dev/input namespace with even more
>>>       devices, these are added to the kcontrols only, not the input
>>> devices.
>>>
>>>       Signed-off-by: David Henningsson <david.henningsson@canonical.com>
>>>       Signed-off-by: Takashi Iwai <tiwai@suse.de>
>>>
>>> :040000 040000 7a954b731bef80cee731763521f64a424a7058a0
>>> 7d176b02ebb4553b0b0f97d200f7527cb69c122a M    sound
>>> --->8---------
>>>
>>> If any additional information is required please let me know.
>>
>>
>> Thanks for the heads up. I would need the following additional information:
>>
>> 1) One alsa-info taken during playback through internal speakers, when the
>> internal speakers are working, and
>> 2) One alsa-info taken during playback through internal speakers, when the
>> internal speakers are not working.
>>
>> (See https://wiki.ubuntu.com/Audio/AlsaInfo if you don't know what alsa-info
>> is.)
>>
>> If you're using PulseAudio for playback, please also include the output of
>> "pacmd list" in the above two scenarios.
>>
>>
>> --
>> David Henningsson, Canonical Ltd.
>> https://launchpad.net/~diwic
>



-- 
David Henningsson, Canonical Ltd.
https://launchpad.net/~diwic

[-- Attachment #2: 0001-ALSA-hda-do-not-detect-jack-on-internal-speakers-for.patch --]
[-- Type: text/x-patch, Size: 1155 bytes --]

>From 17d858f06b5b03c733940a4c3e45dab0108c425b Mon Sep 17 00:00:00 2001
From: David Henningsson <david.henningsson@canonical.com>
Date: Wed, 10 Oct 2012 16:32:09 +0200
Subject: [PATCH] ALSA: hda - do not detect jack on internal speakers for
 Realtek

This caused the internal speaker to mute itself because it was
present, which happened after powersave.
It was found on Dell XPS 15 (L502x), ALC665.

Reported-by: Da Fox <da.fox.mail@gmail.com>
Cc: stable@vger.kernel.org
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
---
 sound/pci/hda/patch_realtek.c |    2 ++
 1 file changed, 2 insertions(+)

diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c
index 8568aee..12d16e2 100644
--- a/sound/pci/hda/patch_realtek.c
+++ b/sound/pci/hda/patch_realtek.c
@@ -611,6 +611,8 @@ static void alc_line_automute(struct hda_codec *codec, struct hda_jack_tbl *jack
 {
 	struct alc_spec *spec = codec->spec;
 
+	if (spec->autocfg.line_out_type == AUTO_PIN_SPEAKER_OUT)
+		return;
 	/* check LO jack only when it's different from HP */
 	if (spec->autocfg.line_out_pins[0] == spec->autocfg.hp_pins[0])
 		return;
-- 
1.7.9.5


[-- Attachment #3: Type: text/plain, Size: 0 bytes --]

Re: Bug 48381 - [Regression][Bisected]Sound gets permanently muted a few minutes after booting

[Takashi Iwai]

At Wed, 10 Oct 2012 16:37:31 +0200,
David Henningsson wrote:
> 
> [1  <text/plain; ISO-8859-1 (7bit)>]
> On 10/10/2012 03:39 PM, Da Fox wrote:
> > Dear Ben,
> 
> Actually, it's David :-)
> 
> > I mostly use mplayer to play audio and video files, using the 'alsa'
> > output driver.
> > I have attached the requested information to the bugtracker in full,
> > the following is a quick diff of the two files:
> 
> After some attempts I think was able to find the culprit. Please test 
> the attached patch and see if it resolves the problem.

I queued it now since the patch looks obviously correct.


thanks,

Takashi

Re: Bug 48381 - [Regression][Bisected]Sound gets permanently muted a few minutes after booting

[Da Fox]

On Wed, Oct 10, 2012 at 5:14 PM, Takashi Iwai <tiwai@suse.de> wrote:
> At Wed, 10 Oct 2012 16:37:31 +0200,
> David Henningsson wrote:
>>
>> [1  <text/plain; ISO-8859-1 (7bit)>]
>> On 10/10/2012 03:39 PM, Da Fox wrote:
>> > Dear Ben,
>>
>> Actually, it's David :-)
Whoopsy, I'm sorry for the mix-up! I was probably (trying to)
multi-task too much :)

>> > I mostly use mplayer to play audio and video files, using the 'alsa'
>> > output driver.
>> > I have attached the requested information to the bugtracker in full,
>> > the following is a quick diff of the two files:
>>
>> After some attempts I think was able to find the culprit. Please test
>> the attached patch and see if it resolves the problem.
>
> I queued it now since the patch looks obviously correct.

I've tested David's patch (applied on top of plain kernel v3.6), and
confirm that this fixes the issue. Auto-mute when plugging in
headphones still works as expected too. Thank you for the quick fix!
Will you be closing the bug at the kernel's bug-tracker once the fix
is upstream (or should I do it now)?

Best regards,
  Da Fox