[INTEL DP55WG,Realtek ALC889] - Audio Playback Unavailable

[Natalia Bidart]

This is a fresh upgrade to raring. During the weekend, I updated from precise to quantal, and from there to raring. I did not use sound while the system was in quantal, so no idea if the bug was present there as well.

The symptom is:

* after some period of sound usage, playback stops working

What I mean with sound usage is:

I tried having a meeting with: mumble, skype, and google hangout, and in all three cases, audio input/outout will work just fine for ~5 minutes, and the playback just dies (people tell me they keep listening to me, so mic works fine).

The only way to solve this is by rebooting. After one of the reboots, I was able to play a 20 minute video with mplayer with no further issue. Then opened skype and after ~3-5 minute talk, audio playback died again.

ProblemType: Bug
DistroRelease: Ubuntu 13.04
Package: alsa-base 1.0.25+dfsg-0ubuntu4
ProcVersionSignature: Ubuntu 3.8.0-26.38-generic 3.8.13.2
Uname: Linux 3.8.0-26-generic x86_64
NonfreeKernelModules: nvidia
ApportVersion: 2.9.2-0ubuntu8.1
Architecture: amd64
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC0: nessita 2627 F.... pulseaudio
Date: Mon Jul 15 14:40:20 2013
InstallationDate: Installed on 2011-12-20 (572 days ago)
InstallationMedia: Ubuntu 12.04 LTS "Precise Pangolin" - Alpha amd64 (20111129.1)
MarkForUpload: True
PackageArchitecture: all
ProcEnviron:
 TERM=xterm
 PATH=(custom, no user)
 XDG_RUNTIME_DIR=<set>
 LANG=en_US.UTF-8
 SHELL=/bin/bash
SourcePackage: alsa-driver
Symptom: audio
Symptom_AlsaPlaybackTest: ALSA playback test through plughw:MID failed
Symptom_AlsaPlaybackTestStderr:
 W r i t e e r r o r : - 5 , I n p u t / o u t p u t e r r o r
  x r u n _ r e c o v e r y f a i l e d : - 5 , I n p u t / o u t p u t e r r o r
  T r a n s f e r f a i l e d : O p e r a t i o n n o t p e r m i t t e d
Symptom_Card: Built-in Audio - HDA Intel MID
Symptom_Jack: Green Headphone Out, Front
Symptom_Type: None of the above
Title: [, Realtek ALC889, Green Headphone Out, Front] Playback problem
UpgradeStatus: Upgraded to raring on 2013-07-13 (2 days ago)
dmi.bios.date: 08/05/2009
dmi.bios.vendor: Intel Corp.
dmi.bios.version: KGIBX10J.86A.3206.2009.0805.1855
dmi.board.asset.tag: Base Board Asset Tag
dmi.board.name: DP55WG
dmi.board.vendor: Intel Corporation
dmi.board.version: AAE57269-404
dmi.chassis.type: 2
dmi.modalias: dmi:bvnIntelCorp.:bvrKGIBX10J.86A.3206.2009.0805.1855:bd08/05/2009:svn:pn:pvr:rvnIntelCorporation:rnDP55WG:rvrAAE57269-404:cvn:ct2:cvr:
---
ApportVersion: 2.17.2-0ubuntu1
Architecture: amd64
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC0: nessita 13188 F.... pulseaudio
CurrentDesktop: Unity
DistroRelease: Ubuntu 15.04
HibernationDevice: RESUME=UUID=bd987ac2-eb4b-43e7-881b-4cf2b5078e4b
InstallationDate: Installed on 2014-05-11 (366 days ago)
InstallationMedia: Ubuntu 14.04 LTS "Trusty Tahr" - Release amd64 (20140417)
Package: linux (not installed)
ProcFB: 0 nouveaufb
ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-3.19.0-16-generic root=UUID=e91bd3a4-787b-404b-9f19-aa6dcbe707b0 ro quiet splash
ProcVersionSignature: Ubuntu 3.19.0-16.16-generic 3.19.3
RelatedPackageVersions:
 linux-restricted-modules-3.19.0-16-generic N/A
 linux-backports-modules-3.19.0-16-generic N/A
 linux-firmware 1.143
RfKill:

Tags: vivid vivid vivid vivid
Uname: Linux 3.19.0-16-generic x86_64
UpgradeStatus: Upgraded to vivid on 2015-03-11 (61 days ago)
UserGroups: adm cdrom dip libvirtd lpadmin plugdev sambashare sudo
_MarkForUpload: True
dmi.bios.date: 08/05/2009
dmi.bios.vendor: Intel Corp.
dmi.bios.version: KGIBX10J.86A.3206.2009.0805.1855
dmi.board.asset.tag: Base Board Asset Tag
dmi.board.name: DP55WG
dmi.board.vendor: Intel Corporation
dmi.board.version: AAE57269-404
dmi.chassis.type: 2
dmi.modalias: dmi:bvnIntelCorp.:bvrKGIBX10J.86A.3206.2009.0805.1855:bd08/05/2009:svn:pn:pvr:rvnIntelCorporation:rnDP55WG:rvrAAE57269-404:cvn:ct2:cvr:

[Natalia Bidart (nataliabidart)]

Tested latest kernel available:

linux-image-3.8.13-03081304-generic_3.8.13-03081304.201307032035_amd64.deb

Installed, rebooted in the new kernel, opened skype, make a test call (all good, as before), called someone and got the same audio playback dying issue.

[Luke Yelavich (themuso)]

Please try to get a log from PulseAudio when this happens, see https://wiki.ubuntu.com/PulseAudio/Log for instructions on how to do this.

[Natalia Bidart (nataliabidart)]

Attached verbose log during reproduction of the issue as requested by Luke.

[Natalia Bidart (nataliabidart)]

I also tested a clean raring by booting it in mem from a pendrive. I managed to reproduce the same unwanted effect by using google hangouts and tweaking sound levels from the "Sound settings" app below the sound icon in the systray.

[Natalia Bidart (nataliabidart)]

As per the advice from #ubuntu-kernel, following https://wiki.ubuntu.com/Audio/PositionReporting, seems like setting options snd-hda-intel position_fix=2 did the trick for me. Will re-open if audio breaks again.

[Natalia Bidart (nataliabidart)]

Update: after using this setup (position_fix=2) for a while, audio broke again, not sure when or how.

[Natalia Bidart (nataliabidart)]

Following instructions from https://wiki.ubuntu.com/Audio/UpgradingAlsa/DKMS:

nessita@dali:~$ sudo dpkg -i oem-audio-hda-daily-dkms_0.201307180559~raring1_all.deb
Selecting previously unselected package oem-audio-hda-daily-dkms.
(Reading database ... 277693 files and directories currently installed.)
Unpacking oem-audio-hda-daily-dkms (from oem-audio-hda-daily-dkms_0.201307180559~raring1_all.deb) ...
Setting up oem-audio-hda-daily-dkms (0.201307180559~raring1) ...
Loading new oem-audio-hda-daily-0.201307180559~raring1 DKMS files...
First Installation: checking all kernels...
Building only for 3.8.0-26-generic
Building for architecture x86_64
Building initial module for 3.8.0-26-generic
Done.

snd-hda-codec-via:
Running module version sanity check.
 - Original module
 - Installation
   - Installing to /lib/modules/3.8.0-26-generic/updates/dkms/

snd-hda-codec-si3054.ko:
Running module version sanity check.
 - Original module
 - Installation
   - Installing to /lib/modules/3.8.0-26-generic/updates/dkms/

snd-hda-codec-ca0110.ko:
Running module version sanity check.
 - Original module
 - Installation
   - Installing to /lib/modules/3.8.0-26-generic/updates/dkms/

snd-hda-codec-ca0132.ko:
Running module version sanity check.
 - Original module
 - Installation
   - Installing to /lib/modules/3.8.0-26-generic/updates/dkms/

snd-hda-codec-analog.ko:
Running module version sanity check.
 - Original module
 - Installation
   - Installing to /lib/modules/3.8.0-26-generic/updates/dkms/

snd-hda-codec-realtek.ko:
Running module version sanity check.
 - Original module
 - Installation
   - Installing to /lib/modules/3.8.0-26-generic/updates/dkms/

snd-hda-codec.ko:
Running module version sanity check.
 - Original module
 - Installation
   - Installing to /lib/modules/3.8.0-26-generic/updates/dkms/

snd-hda-codec-conexant.ko:
Running module version sanity check.
 - Original module
 - Installation
   - Installing to /lib/modules/3.8.0-26-generic/updates/dkms/

snd-hda-codec-hdmi.ko:
Running module version sanity check.
 - Original module
 - Installation
   - Installing to /lib/modules/3.8.0-26-generic/updates/dkms/

snd-hda-intel.ko:
Running module version sanity check.
 - Original module
 - Installation
   - Installing to /lib/modules/3.8.0-26-generic/updates/dkms/

snd-hda-codec-idt.ko:
Running module version sanity check.
 - Original module
 - Installation
   - Installing to /lib/modules/3.8.0-26-generic/updates/dkms/

snd-hda-codec-cirrus.ko:
Running module version sanity check.
 - Original module
 - Installation
   - Installing to /lib/modules/3.8.0-26-generic/updates/dkms/

snd-hda-codec-cmedia.ko:
Running module version sanity check.
 - Original module
 - Installation
   - Installing to /lib/modules/3.8.0-26-generic/updates/dkms/

depmod........

DKMS: install completed.

[Natalia Bidart (nataliabidart)]

The update to dkms did not help. A piece of output from pulseaudio, once audio playback is broken, when playing a video with mplayer:

# tail -f /var/log/syslog
Jul 19 12:45:34 pulseaudio[4457]: last message repeated 10 times
Jul 19 12:45:34 dali pulseaudio[4457]: [pulseaudio] client.c: Created 4 "Native client (UNIX socket client)"
Jul 19 12:45:34 dali pulseaudio[4457]: [pulseaudio] protocol-native.c: Protocol version: remote 27, local 27
Jul 19 12:45:34 dali pulseaudio[4457]: [pulseaudio] protocol-native.c: Got credentials: uid=1000 gid=1000 success=1
Jul 19 12:45:34 dali pulseaudio[4457]: [pulseaudio] protocol-native.c: SHM possible: yes
Jul 19 12:45:34 dali pulseaudio[4457]: [pulseaudio] protocol-native.c: Negotiated SHM: yes
Jul 19 12:45:34 dali pulseaudio[4457]: [pulseaudio] module-augment-properties.c: Looking for .desktop file for mplayer
Jul 19 12:45:34 dali pulseaudio[4457]: [pulseaudio] module-stream-restore.c: Restoring device for stream sink-input-by-application-name:MPlayer.
Jul 19 12:45:34 dali pulseaudio[4457]: [pulseaudio] module-intended-roles.c: Not setting device for stream audio stream, because already set.
Jul 19 12:45:34 dali pulseaudio[4457]: [pulseaudio] sink-input.c: Trying to change sample rate
Jul 19 12:45:34 dali pulseaudio[4457]: [pulseaudio] sink.c: Cannot update rate, monitor source is RUNNING
Jul 19 12:45:34 dali pulseaudio[4457]: [pulseaudio] module-stream-restore.c: Restoring volume for sink input sink-input-by-application-name:MPlayer.
Jul 19 12:45:34 dali pulseaudio[4457]: [pulseaudio] module-stream-restore.c: Restoring mute state for sink input sink-input-by-application-name:MPlayer.
Jul 19 12:45:34 dali pulseaudio[4457]: [pulseaudio] module-suspend-on-idle.c: Sink alsa_output.pci-0000_00_1b.0.analog-stereo becomes busy.
Jul 19 12:45:34 dali pulseaudio[4457]: [pulseaudio] resampler.c: Using resampler 'speex-float-1'
Jul 19 12:45:34 dali pulseaudio[4457]: [pulseaudio] resampler.c: Using float32le as working format.
Jul 19 12:45:34 dali pulseaudio[4457]: [pulseaudio] resampler.c: Choosing speex quality setting 1.
Jul 19 12:45:34 dali pulseaudio[4457]: [pulseaudio] memblockq.c: memblockq requested: maxlength=33554432, tlength=0, base=4, prebuf=0, minreq=1 maxrewind=0
Jul 19 12:45:34 dali pulseaudio[4457]: [pulseaudio] memblockq.c: memblockq sanitized: maxlength=33554432, tlength=33554432, base=4, prebuf=0, minreq=4 maxrewind=0
Jul 19 12:45:34 dali pulseaudio[4457]: [pulseaudio] sink-input.c: Created input 1 "audio stream" on alsa_output.pci-0000_00_1b.0.analog-stereo with sample spec float32le 2ch 48000Hz and channel map front-left,front-right
Jul 19 12:45:34 dali pulseaudio[4457]: [pulseaudio] sink-input.c: media.name = "audio stream"
Jul 19 12:45:34 dali pulseaudio[4457]: [pulseaudio] sink-input.c: application.name = "MPlayer"
Jul 19 12:45:34 dali pulseaudio[4457]: [pulseaudio] sink-input.c: native-protocol.peer = "UNIX socket client"
Jul 19 12:45:34 dali pulseaudio[4457]: [pulseaudio] sink-input.c: native-protocol.version = "27"
Jul 19 12:45:34 dali pulseaudio[4457]: [pulseaudio] sink-input.c: application.process.id = "4466"
Jul 19 12:45:34 dali pulseaudio[4457]: [pulseaudio] sink-inp...

[David Henningsson (diwic)]

After some debugging on IRC, I'm still not sure what's causing the problem. We should try a build with DEBUG_TIMING (in module-alsa-sink.c) on for even more verbose logging in PulseAudio.

[Raymond (superquad-vortex2)]

do your computer chassis have hda front audio panel with Jack detection ?

do your BIOS have any option to set front audio panel type :hda or AC97 ?

MISC = NO_PRESENCE mean front audio panel has no Jack detection circuit

http://www.intel.com/support/motherboards/desktop/sb/cs-015851.htm

control.45 {
  iface CARD
  name 'Front Headphone Phantom Jack'
  value true
  comment {
   access read
   type BOOLEAN
   count 1
  }
 }
 control.46 {
  iface CARD
  name 'Front Mic Phantom Jack'
  value true
  comment {
   access read
   type BOOLEAN
   count 1
  }
 }

Node 0x15 [Pin Complex] wcaps 0x40018d: Stereo Amp-Out
  Control: name="Headphone Playback Switch", index=0, device=0
    ControlAmp: chs=3, dir=Out, idx=0, ofs=0
  Control: name="Front Headphone Phantom Jack", index=0, device=0
  Amp-Out caps: ofs=0x00, nsteps=0x00, stepsize=0x00, mute=1
  Amp-Out vals: [0x00 0x00]
  Pincap 0x0000003c: IN OUT HP Detect
  Pin Default 0x02214520: [Jack] HP Out at Ext Front
    Conn = 1/8, Color = Green
    DefAssociation = 0x2, Sequence = 0x0
    Misc = NO_PRESENCE
  Pin-ctls: 0xc0: OUT HP
  Unsolicited: tag=00, enabled=0
  Connection: 5
     0x0c 0x0d 0x0e 0x0f 0x26*

[Natalia Bidart (nataliabidart)]

Raymond, below some answers:

* do your computer chassis have hda front audio panel with Jack detection ?

The computer chassis does have front audio panel, I'm not sure how to check "visually" if it has Jack detection or not. The chassis is exactly this one: http://www.sentey.com/en_productos_details.php?tipo=Gabinetes&ctg=Professional%20Series&prod=ps2-3270

* do your BIOS have any option to set front audio panel type :hda or AC97?

Will check next time I reboot; motherboard is Intel DP55WG (perhaps that answers your question?)

In any case, is worth noting that audio playback works perfectly both in the headset using the front audio connectot and in the speakers plugged in the rear audio output (they work perfectly *at the same time*, until they both stop working altogether).
I have also tried unplugging one of each at a time (ie have only headset, have only speakers). This does not look like a hardware issue, since audio works for some random amount of time after a reboot.

[Raymond (superquad-vortex2)]

you can use hda-jack-sense-test to find out whether the front audio panel headphone and mic have Jack detection or not ?

it is strange that dB min and dBmax of virtual master is different from it slaves (e.g. front, suroound, center, pre and headphone)

state.MID {
 control.1 {
  iface MIXER
  name 'Front Playback Volume'
  value.0 62
  value.1 62
  comment {
   access 'read write'
   type INTEGER
   count 2
   range '0 - 64'
   dbmin -6200
   dbmax 200
   dbvalue.0 0
   dbvalue.1 0
  }
 }

control.39 {
  iface MIXER
  name 'Master Playback Volume'
  value 60
  comment {
   access 'read write'
   type INTEGER
   count 1
   range '0 - 64'
   dbmin -6400
   dbmax 0
   dbvalue.0 -400
  }
 }

[Natalia Bidart (nataliabidart)]

nessita@dali:~$ sudo hda-jack-sense-test
Traceback (most recent call last):
  File "/usr/bin/hda-jack-sense-test", line 47, in <module>
    codec = HDACodec(cardindex, codecindex)
  File "/usr/share/snd-hda-tools/hda_codec.py", line 1023, in __init__
    self.fd = os.open("/dev/snd/hwC%sD%s" % (card, device), os.O_RDWR)
OSError: [Errno 2] No such file or directory: '/dev/snd/hwC0D0'

nessita@dali:~$ ls /dev/snd/
by-path/ controlC0 hwC0D2 pcmC0D0c pcmC0D0p pcmC0D1c pcmC0D1p pcmC0D2c seq timer

Edited the tool so /dev/snd/hwC0D2 was used and got:

nessita@dali:~$ sudo hda-jack-sense-test
Unable to find proc file '/proc/asound/card0/codec#0'
Pin 0x14 (Green Line Out): present = No
Pin 0x16 (Orange Line Out): present = No
Pin 0x17 (Grey Line Out): present = No
Pin 0x19 (Black Line Out): present = Yes
Pin 0x1a (Blue Line In): present = No
Pin 0x1b (Pink Mic): present = No

At the time the test was run, I had the speakers connected to the black line out, and the headset in the front panel. Audio playback is not working at the time the test was run.

Still need to check the BIOS setting, need to reboot to do that, will try to get that done today.

[Natalia Bidart (nataliabidart)]

Rebooted and checked the BIOS. Couldn't find something specifically related to the options AC97 or :hda, but took a screenshot of the peripheral devices screen (shows "Legacy Front Panel Audio -> <enabled>":

http://ubuntuone.com/3y5DR6FKZdWS3WKU1GgQ73

[Raymond (superquad-vortex2)]

you may need to disable Legacy front audio panel in BIOS setup

 check whether misc bit is clear in node 0x15 and 0x18

[Raymond (superquad-vortex2)]

the availablityy of headphone should be either yes or no instead of unknown

ports:
  analog-output: Analog Output (priority 9900, latency offset 0 usec, available: unknown)
   properties:

  analog-output-headphones: Headphones (priority 9000, latency offset 0 usec, available: unknown)
   properties:
    device.icon_name = "audio-headphones"
 active port: <analog-output>
2 source(s) available.

[Natalia Bidart (nataliabidart)]

Raymond:

Re: "you may need to disable Legacy front audio panel in BIOS setup" -> does this mean I will not be able to use the headset from the front panel?

Re: "check whether misc bit is clear in node 0x15 and 0x18" -> no idea how to do this, would you please advice what do I need to do?

And is the comment #18 for me is it technical info for the bug report? If the former, I'm not sure what that means either.

Thanks!

[Raymond (superquad-vortex2)]

leagcy front audio panel is AC97 AUDIO PANEL without Jack detection

you have to disable it to use HDA front audio panel with Jack detection

the rear panel jacks are auto muted when you plugged the headphone by the driver

the codec info is in the output of alsa-info.sh

the headphone won`t be available in sound preference since BIOS set misc bit of pin default of front panel jacks for AC97 audio panel

[Raymond (superquad-vortex2)]

control.41 {
  iface CARD
  name 'Line Out Front Jack'
  value true
  comment {
   access read
   type BOOLEAN
   count 1
  }
 }
 control.42 {
  iface CARD
  name 'Line Out Surround Jack'
  value false
  comment {
   access read
   type BOOLEAN
   count 1
  }
 }
 control.43 {
  iface CARD
  name 'Line Out CLFE Jack'
  value false
  comment {
   access read
   type BOOLEAN
   count 1
  }
 }
 control.44 {
  iface CARD
  name 'Line Out Side Jack'
  value false
  comment {
   access read
   type BOOLEAN
   count 1
  }
 }

 control.47 {
  iface CARD
  name 'Rear Mic Jack'
  value false
  comment {
   access read
   type BOOLEAN
   count 1
  }
 }
 control.48 {
  iface CARD
  name 'Line Jack'
  value false
  comment {
   access read
   type BOOLEAN
   count 1
  }

[Raymond (superquad-vortex2)]

you have to ask the author since he assume no gain in vmaster but slaves can have gain in your case

https://git.kernel.org/cgit/linux/kernel/git/tiwai/sound.git/commit/?id=3b0a5f22d4649433a5842ffc7313803292e95718

+/*
+ * Add a slave control to the group with the given master control
+ *
+ * All slaves must be the same type (returning the same information
+ * via info callback). The fucntion doesn't check it, so it's your
+ * responsibility.
+ *
+ * Also, some additional limitations:
+ * - at most two channels
+ * - logarithmic volume control (dB level), no linear volume
+ * - master can only attenuate the volume, no gain

[David Henningsson (diwic)]

While it would be interesting to see if jack detection works differently if "Legacy Front Panel Audio" was disabled, it is unrelated to this issue.

This issue is about when sound suddenly stops to work, and does so for all output jacks.

[David Henningsson (diwic)]

One thing to check is what's causing the > 10 ms system latencies here. In a root shell, execute this command:

cd /sys/kernel/debug/tracing
echo wakeup_rt > current_tracer
echo 1 > tracing_on

Run for a minute or two, then

cat trace > /tmp/trace.txt
echo 0 > tracing on

Then attach /tmp/trace.txt here. Thanks!

[David Henningsson (diwic)]

Hi,

I've made a special build in this PPA:

https://launchpad.net/~diwic/+archive/pulse-timing-test

Once it has finished building, could you try to give me a pulseaudio verbose log? Try to reproduce the hangout/mumble/skype bug with the pulseaudio package in this ppa, and the log enabled. Thanks!

[Raymond (superquad-vortex2)]

> sudo hda-jack-sense-test
Unable to find proc file '/proc/asound/card0/codec#0'
Pin 0x14 (Green Line Out): present = No
Pin 0x16 (Orange Line Out): present = No
Pin 0x17 (Grey Line Out): present = No
Pin 0x19 (Black Line Out): present = Yes
Pin 0x1a (Blue Line In): present = No
Pin 0x1b (Pink Mic): present = No

If the objective of hda-jack-sense-test is to figure out if your BIOS is lying to you
about what pin is used for what.

it should check those pin complex which pincap support DETECT instead of ignoring those Misc = NO_PRESENCE in order to find out whether the jack support jack sense or not

AC_PINCAP_PRES_DETECT (1<<2) /* presence detect capable */

Node 0x15 [Pin Complex] wcaps 0x40018d: Stereo Amp-Out
  Pincap 0x0000003c: IN OUT HP Detect
  Pin Default 0x02214520: [Jack] HP Out at Ext Front

[Natalia Bidart (nataliabidart)]

nessita@dali:~$ apt-cache policy pulseaudio
pulseaudio:
  Installed: 1:4.0-0ubuntu1+debugtiming
  Candidate: 1:4.0-0ubuntu1+debugtiming
  Version table:
 *** 1:4.0-0ubuntu1+debugtiming 0
        500 http://ppa.launchpad.net/diwic/pulse-timing-test/ubuntu/ raring/main amd64 Packages
        100 /var/lib/dpkg/status
     1:3.0-0ubuntu6 0
        500 http://us.archive.ubuntu.com/ubuntu/ raring/main amd64 Packages

[Natalia Bidart (nataliabidart)]

Removed verbose log to syslog and tried again with local verbose log as per wiki instrucions. Booted computer, logged in, ran pulseaudio, and opened mumble. Got this:

nessita@dali:~$ LANG=C pulseaudio -vvvv --log-time=1 > ~/pulseverbose.log 2>&1
Trace/breakpoint trap (core dumped)

Attaching verbose.log.

Ran pulseaudio with gdb, got:

Program received signal SIGTRAP, Trace/breakpoint trap.
[Switching to Thread 0x7fffed380700 (LWP 3580)]
0x00007fffedbad451 in ?? () from /usr/lib/pulse-4.0/modules/libalsa-util.so
(gdb) w
Ambiguous command "w": watch, wh, whatis, where, while, while-stepping, winheight, ws.
(gdb) where
#0 0x00007fffedbad451 in ?? () from /usr/lib/pulse-4.0/modules/libalsa-util.so
#1 0x00007fffedbaf164 in ?? () from /usr/lib/pulse-4.0/modules/libalsa-util.so
#2 0x00007ffff792bd63 in ?? () from /usr/lib/x86_64-linux-gnu/pulseaudio/libpulsecommon-4.0.so
#3 0x00007ffff6e34f8e in start_thread (arg=0x7fffed380700) at pthread_create.c:311
#4 0x00007ffff644de1d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:113
(gdb)

[Natalia Bidart (nataliabidart)]

Installed updated pulseaudio from PPA and re-tested with:

LANG=C pulseaudio -vvvv --log-time=1 > ~/pulseverbose.log 2>&1

Got no crashes, but no audio playback either (mix did work this time). The pulseaudio log is 84M, so uploading it compressed with bzip2.

[David Henningsson (diwic)]

Looking at the latest logs, there are terrible delays, probably for which we cannot blame the graphics driver.

( 61.711| 0.007) D: [alsa-sink-ALC889 Analog] alsa-sink.c: Expected sleep: 9.27ms, real sleep: 7.91ms (diff -1.36 ms)
( 61.920| 0.204) D: [alsa-sink-ALC889 Analog] alsa-sink.c: Loop
( 61.920| 0.000) D: [alsa-sink-ALC889 Analog] alsa-sink.c: Buffer time: 46 ms; Sleep time: 10 ms; Process time: 35 ms
( 61.716| 0.005) D: [alsa-source-ALC889 Analog] alsa-source.c: Expected sleep: 19.99ms, real sleep: 20.09ms (diff 0.10 ms)
( 61.920| 0.000) D: [alsa-sink-ALC889 Analog] alsa-sink.c: avail: 95888
( 61.921| 0.000) D: [alsa-source-ALC889 Analog] alsa-source.c: Loop
( 61.921| 0.000) I: [alsa-sink-ALC889 Analog] alsa-sink.c: Underrun!

There is 200 ms of delay between the "Expected sleep" line and "Loop" line, which is not normal. Any chance you can look at the system as a whole to see what's causing these delays? I'm not really sure how to do this, but maybe "top" could give an initial indication if there's anything that's taking a lot of CPU power.

[David Henningsson (diwic)]

To sum up our IRC discussions, a few things to try are:

 - Install precise, verify it's still working correctly
 - Upgrade to a raring kernel in precise (install linux-generic-lts-raring-eol-upgrade and reboot) and see if this causes the problem to appear
 - If not, also upgrade to raring X stack in precise (install xserver-xorg-lts-raring and libgl1-mesa-glx-lts-raring, reboot) and see if this causes the problem to appear

Also, does anything appear in dmesg when these problems appear, that could give us a hint of what is happening?

[Natalia Bidart (nataliabidart)]

Booted in the raring PC with the audio issues. dmesg output before reproducing the issue has the following line that looks suspicious to me:

[13828.248950] hrtimer: interrupt took 15288 ns

When started pulseaudio with the -vvvv option dmesg showed (did not open any audio program yet):

[27622.091908] hda-intel 0000:00:1b.0: Unstable LPIB (65536 >= 8192); disabling LPIB delay counting

After reproducing the issue by using mumble, dmesg shows *nothing* new.
At this stage, top shows:

nessita@dali:~$ top

top - 15:50:13 up 7:45, 8 users, load average: 0.14, 0.22, 0.23
Tasks: 289 total, 3 running, 284 sleeping, 0 stopped, 2 zombie
%Cpu(s): 2.0 us, 1.6 sy, 0.0 ni, 95.9 id, 0.0 wa, 0.0 hi, 0.5 si, 0.0 st
KiB Mem: 8167864 total, 3780108 used, 4387756 free, 377504 buffers
KiB Swap: 6290428 total, 0 used, 6290428 free, 866680 cached

  PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
11852 nessita 20 0 1461m 55m 32m S 5.3 0.7 0:13.38 mumble
10633 nessita 9 -11 330m 7100 4364 R 2.3 0.1 0:05.73 pulseaudio

Last few lines from pulse log are:

.... a lot ....
( 168.054| 0.000) E: [alsa-sink-ALC889 Analog] protocol-native.c: request_bytes(192)
( 168.054| 0.000) E: [alsa-sink-ALC889 Analog] protocol-native.c: ALSA Playback, pop(): 0
( 168.054| 0.000) E: [pulseaudio] protocol-native.c: Requesting 2048 bytes
( 168.054| 0.000) E: [pulseaudio] protocol-native.c: signalling underflow
( 168.059| 0.004) E: [alsa-sink-ALC889 Analog] protocol-native.c: ALSA Playback, pop(): 0
( 169.389| 1.330) I: [alsa-sink-ALC889 Analog] alsa-sink.c: Scheduling delay of 1.02ms > 0.96ms, you might want to investigate this to improve latency...
( 173.060| 3.671) I: [pulseaudio] module-suspend-on-idle.c: Sink alsa_output.pci-0000_00_1b.0.analog-stereo idle for too long, suspending ...
( 173.061| 0.000) I: [alsa-sink-ALC889 Analog] alsa-sink.c: Device suspended...
( 251.342| 78.281) I: [alsa-source-ALC889 Analog] alsa-source.c: Scheduling delay of 4.75ms, you might want to investigate this to improve latency...
( 251.409| 0.066) I: [alsa-source-ALC889 Analog] alsa-source.c: Scheduling delay of 12.15ms, you might want to investigate this to improve latency...
( 251.409| 0.000) I: [alsa-source-ALC889 Analog] alsa-source.c: Overrun!
( 251.409| 0.000) I: [alsa-source-ALC889 Analog] alsa-source.c: Increasing wakeup watermark to 16.00 ms
( 256.575| 5.166) I: [alsa-source-ALC889 Analog] alsa-source.c: Scheduling delay of 6.29ms, you might want to investigate this to improve latency...
( 256.591| 0.015) I: [alsa-source-ALC889 Analog] alsa-source.c: Scheduling delay of 5.32ms, you might want to investigate this to improve latency...
( 269.375| 12.784) I: [alsa-source-ALC889 Analog] alsa-source.c: Scheduling delay of 9.03ms, you might want to investigate this to improve latency...
( 277.341| 7.966) I: [alsa-source-ALC889 Analog] alsa-source.c: Scheduling delay of 6.62ms, you might want to investigate this to improve latency...
( 285.666| 8.324) I: [alsa-source-ALC889 Analog] alsa-source.c: ...

[David Henningsson (diwic)]

> dmesg output before reproducing the issue has the following line that looks suspicious to me:
> [13828.248950] hrtimer: interrupt took 15288 ns

It seems a bit suspicious to me too, but 15288 ns = 0.015 ms, which should not be enough to disturb the overall system latency.

> [27622.091908] hda-intel 0000:00:1b.0: Unstable LPIB (65536 >= 8192); disabling LPIB delay counting

That is okay. It indicates that "position_fix=1" won't work reliably, which we already knew.

[Natalia Bidart (nataliabidart)]

(sort of) Good news!

* Installed an Ubuntu Precise LTS system, from scratch, in an empty partition:

foobar@dale:~$ lsb_release -a
No LSB modules are available.
Distributor ID: Ubuntu
Description: Ubuntu 12.04.2 LTS
Release: 12.04
Codename: precise

* Tried to break audio by any means, I couldn't.

* Installed the suggested kernel from raring, as in:

foobar@dale:~$ uname -a
Linux dale 3.8.0-27-generic #40~precise3-Ubuntu SMP Fri Jul 19 14:38:30 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux

foobar@dale:~$ apt-cache policy linux-image-3.8.0-27-generic
linux-image-3.8.0-27-generic:
  Installed: 3.8.0-27.40~precise3
  Candidate: 3.8.0-27.40~precise3
  Version table:
 *** 3.8.0-27.40~precise3 0
        500 http://ar.archive.ubuntu.com/ubuntu/ precise-updates/main amd64 Packages
        500 http://security.ubuntu.com/ubuntu/ precise-security/main amd64 Packages
        100 /var/lib/dpkg/status

* Opened mumble, ran the audio wizard a couple of times, and also played the sample ogg from the default "Examples" folder, and audio playback died.

Attaching pulseaudioverbose.log, obtained by following the ubuntu wiki instructions.

[Natalia Bidart (nataliabidart)]

nessita@dali:~$ cat /proc/interrupts
           CPU0 CPU1 CPU2 CPU3
  0: 43 0 0 0 IO-APIC-edge timer
  1: 1 0 0 2 IO-APIC-edge i8042
  8: 0 0 0 1 IO-APIC-edge rtc0
  9: 0 0 0 0 IO-APIC-fasteoi acpi
 12: 1 0 0 4 IO-APIC-edge i8042
 16: 10152 13612 110778 15223 IO-APIC-fasteoi ehci_hcd:usb1, nvidia
 19: 6122 117488 0 0 IO-APIC-fasteoi ata_piix, ata_piix, firewire_ohci
 21: 117236 0 186 0 IO-APIC-fasteoi eth0
 23: 0 3433 31343 5566 IO-APIC-fasteoi ehci_hcd:usb2
 40: 918436 0 0 0 HPET_MSI-edge hpet2
 41: 0 1007770 0 0 HPET_MSI-edge hpet3
 42: 0 0 689446 0 HPET_MSI-edge hpet4
 43: 0 0 0 603787 HPET_MSI-edge hpet5
 45: 167339473 0 0 0 PCI-MSI-edge snd_hda_intel
NMI: 1906 405 304 263 Non-maskable interrupts
LOC: 20 17 13 10 Local timer interrupts
SPU: 0 0 0 0 Spurious interrupts
PMI: 1906 405 304 263 Performance monitoring interrupts
IWI: 0 0 0 0 IRQ work interrupts
RTR: 3 0 0 0 APIC ICR read retries
RES: 159721 65093 10550 6350 Rescheduling interrupts
CAL: 417 593 717 673 Function call interrupts
TLB: 21067 4427 4571 3131 TLB shootdowns
TRM: 0 0 0 0 Thermal event interrupts
THR: 0 0 0 0 Threshold APIC interrupts
MCE: 0 0 0 0 Machine check exceptions
MCP: 26 26 26 26 Machine check polls
ERR: 0
MIS: 0

[Natalia Bidart (nataliabidart)]

* Booted a live Saucy from pendrive:

ubuntu@ubuntu:~$ uname -a
Linux ubuntu 3.10.0-6-generic #17-Ubuntu SMP Fri Jul 26 18:29:23 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux

ubuntu@ubuntu:~$ lsb_release -a
No LSB modules are available.
Distributor ID: Ubuntu
Description: Ubuntu Saucy Salamander (development branch)
Release: 13.10
Codename: saucy

* Disabled pulseaudio autospawn and ran with verbose logging as per wiki instructions (log attached).

* Installed mumble and ran it from a terminal, audio playback broke immediatly (the faster I ever seen). Mumble crashed with:

ubuntu@ubuntu:~$ mumble
G15LCDEngineUnix: Unable to connect to G15Daemon.
CELT bitstream 8000000b from /usr/lib/mumble/libcelt0.so.0.7.0
Locale is en_US
Database SQLite: "3.7.17"
Overlay: Listening on "/home/ubuntu/.MumbleOverlayPipe"
GlobalShortcutX: Unable to open any keyboard input devices under /dev/input, falling back to XInput
GlobalShortcutX: Using XI2 2.0
AudioInput: 40000 bits/s, 48000 hz, 480 sample CELT
PulseAudio: Starting output: alsa_output.pci-0000_00_1b.0.analog-stereo
PulseAudio: Starting input alsa_input.pci-0000_00_1b.0.analog-stereo
AudioOutput: Initialized 2 channel 44100 hz mixer
AudioInput: Initialized mixer for 1 channel 44100 hz mic and 0 channel 48000 hz echo
warning: The VAD has been replaced by a hack pending a complete rewrite
AudioInput: 40000 bits/s, 48000 hz, 480 sample CELT
AudioInput: 40000 bits/s, 48000 hz, 480 sample CELT
AudioInput: Initialized mixer for 1 channel 44100 hz mic and 0 channel 48000 hz echo
AudioOutput: Initialized 2 channel 44100 hz mixer
warning: The VAD has been replaced by a hack pending a complete rewrite
AudioInput: 40000 bits/s, 48000 hz, 480 sample CELT
AudioInput: Initialized mixer for 1 channel 44100 hz mic and 0 channel 48000 hz echo
warning: The VAD has been replaced by a hack pending a complete rewrite
AudioOutput: Initialized 2 channel 44100 hz mixer
AudioInput: 40000 bits/s, 48000 hz, 480 sample CELT
AudioInput: Initialized mixer for 1 channel 44100 hz mic and 0 channel 48000 hz echo
warning: The VAD has been replaced by a hack pending a complete rewrite
AudioOutput: Initialized 2 channel 44100 hz mixer
AudioInput: 40000 bits/s, 48000 hz, 480 sample CELT
AudioInput: Initialized mixer for 1 channel 44100 hz mic and 0 channel 48000 hz echo
warning: The VAD has been replaced by a hack pending a complete rewrite
AudioOutput: Initialized 2 channel 44100 hz mixer
Bus error (core dumped)

* Interrupts are:

ubuntu@ubuntu:~$ cat /proc/interrupts
           CPU0 CPU1 CPU2 CPU3
  0: 43 0 0 0 IO-APIC-edge timer
  1: 1 0 0 2 IO-APIC-edge i8042
  8: 0 0 0 1 IO-APIC-edge rtc0
  9: 0 0 0 0 IO-APIC-fasteoi acpi
 12: 1 0 0 4 IO-APIC-edge i8042
 16: 0 377 74074 0 IO-APIC-fasteoi ehci_hcd:usb1, nouveau
 19: 1808 10 0 0 IO-APIC-fasteoi ata_piix, ata_piix, firewire_ohci
 21: 24882 0 84 0 IO-APIC-fas...

[David Henningsson (diwic)]

Looking at the latest logs, I think we actually have two different bugs to deal with.

One being the long system delays, which might or might not be related to the graphics driver.

The other one being the audio driver. The regular underruns at 194 ms, and the high interrupt count, both seem to indicate some problem with the controller.

[Brad Figg (brad-figg)]

This bug is missing log files that will aid in diagnosing the problem. From a terminal window please run:

apport-collect 1201528

and then change the status of the bug to 'Confirmed'.

If, due to the nature of the issue you have encountered, you are unable to run this command, please add a comment stating that fact and change the bug status to 'Confirmed'.

This change has been made by an automated script, maintained by the Ubuntu Kernel Team.

[Joseph Salisbury (jsalisbury)]

I'd like to perform a bisect to figure out what commit caused this regression. We need to identify the earliest kernel where the issue started happening as well as the latest kernel that did not have this issue.

Can you test the following kernels and report back? We are looking for the first kernel version that exhibits this bug:
v3.5 final: http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.5-quantal/
v3.6 final: http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.6-quantal/
v3.7 final: http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.7-raring/
v3.8 final: http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.8-raring/
v3.9-rc1: http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.9-rc1-saucy

You don't have to test every kernel, just up until the kernel that first has this bug.

Thanks in advance!

[Joseph Salisbury (jsalisbury)]

Also, it would be good to know if this bug is already fixed in the current mainline kernel. Can you also test the following kernel:

http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.11-rc4-saucy/

[Joseph Salisbury (jsalisbury)]

Sorry the v3.9-rc1 link should be:
http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.9-rc1-raring/

[Natalia Bidart (nataliabidart)]

Tested kernel 3.5.0-030500, was not able to break (tried my best). Attached pulseudio verbose log.

[Natalia Bidart (nataliabidart)]

Tested kernel 3.6.0-030600, was not able to break audio, still trying my best. Attached pulseudio verbose log.

[Natalia Bidart (nataliabidart)]

Tested kernel 3.7.0-030700, audio playback survived all sorts of stress test. No breakage. Attaching pulseudio verbose log.

[Natalia Bidart (nataliabidart)]

Tested kernel 3.8.0-030800, made the same tests, no audio breakage. Note that 3.8.0-27 does break audio, quite easily. Attaching verbose pulseaidio logs from both kernels.

[Natalia Bidart (nataliabidart)]

After booting 3.9.0-030900rc1 could not login to desktop session, keep getting constant kernel panics, apparently caused by nouveau module. From /var/log/kern.log:

Aug 6 16:28:55 dale kernel: [ 48.092580] traps: gdu-notificatio[1878] trap int3 ip:7f9e5cda4fdb sp:7fff193ca1b0 error:0
Aug 6 16:28:57 dale kernel: [ 49.249223] nouveau E[ PFB][0000:01:00.0] trapped read at 0x002001e020 on channel 0x0001fb24 [unknown] SEMAPHORE_BG/PFIFO_READ/00 reason: PAGE_NOT_PRESENT
Aug 6 16:28:57 dale kernel: [ 49.249326] nouveau E[ PFIFO][0000:01:00.0] CACHE_ERROR - ch 2 [Xorg[1070]] subc 0 mthd 0x0060 data 0xbeef0201
Aug 6 16:28:57 dale kernel: [ 49.252849] nouveau E[ PGRAPH][0000:01:00.0] TRAP_M2MF NOTIFY
Aug 6 16:28:57 dale kernel: [ 49.252885] nouveau E[ PGRAPH][0000:01:00.0] TRAP_M2MF 00304041 00004360 00000000 06000432
Aug 6 16:28:57 dale kernel: [ 49.252916] nouveau E[ PGRAPH][0000:01:00.0] TRAP
Aug 6 16:28:57 dale kernel: [ 49.252946] nouveau E[ PGRAPH][0000:01:00.0] ch 2 [0x001fb24000 Xorg[1070]] subc 0 class 0x5039 mthd 0x0328 data 0x00000000
Aug 6 16:28:57 dale kernel: [ 49.252994] nouveau E[ PFB][0000:01:00.0] trapped write at 0x002001e010 on channel 0x0001fb24 [Xorg[1070]] PGRAPH/DISPATCH/M2M_NOTIFY reason: PAGE_NOT_PRESENT
Aug 6 16:28:57 dale kernel: [ 49.253088] nouveau E[ PGRAPH][0000:01:00.0] TRAP_M2MF IN
Aug 6 16:28:57 dale kernel: [ 49.253124] nouveau E[ PGRAPH][0000:01:00.0] TRAP_M2MF 003200d1 2039a980 00000000 04000000
Aug 6 16:28:57 dale kernel: [ 49.253154] nouveau E[ PGRAPH][0000:01:00.0] TRAP
Aug 6 16:28:57 dale kernel: [ 49.253184] nouveau E[ PGRAPH][0000:01:00.0] ch 2 [0x001fb24000 Xorg[1070]] subc 0 class 0x5039 mthd 0x0328 data 0x00000000
Aug 6 16:29:18 dale kernel: [ 70.935563] nouveau E[ PFB][0000:01:00.0] trapped read at 0x002001e020 on channel 0x0001fb24 [unknown] SEMAPHORE_BG/PFIFO_READ/00 reason: PAGE_NOT_PRESENT
Aug 6 16:29:19 dale kernel: [ 71.300930] traps: gdu-notificatio[2198] trap int3 ip:7fc07ed52fdb sp:7fffc5051b40 error:0
Aug 6 16:29:21 dale kernel: [ 73.932288] nouveau E[Xorg[1894]] failed to idle channel 0xcccc0000 [Xorg[1894]]
Aug 6 16:29:21 dale kernel: [ 73.932340] nouveau E[ PFB][0000:01:00.0] trapped read at 0x002001c100 on channel 0x0001fb24 [unknown] SEMAPHORE_BG/PFIFO_READ/00 reason: PAGE_NOT_PRESENT
Aug 6 16:29:21 dale kernel: [ 73.932732] nouveau E[ PGRAPH][0000:01:00.0] TRAP_DISPATCH (unknown 0x00000004)
Aug 6 16:29:21 dale kernel: [ 73.932828] nouveau E[ PFB][0000:01:00.0] trapped read at 0x0000000000 on channel 0x0001fb24 [Xorg[1894]] PGRAPH/DISPATCH/GRCTX reason: DMAOBJ_LIMIT
Aug 6 16:29:21 dale kernel: [ 73.932915] nouveau E[ PGRAPH][0000:01:00.0] ILLEGAL_MTHD ILLEGAL_CLASS
Aug 6 16:29:21 dale kernel: [ 73.932951] nouveau E[ PGRAPH][0000:01:00.0] ch -1 [0x001fb24000 unknown] subc 2 class 0x0000 mthd 0x0200 data 0x000000cf

(a LOT more errors like the above).

[Natalia Bidart (nataliabidart)]

Tested kernel 3.9.0-030900 instead of 3.9.0-030900rc1 (see previous comment). Could not break audio. Verbose pulseaudio log attached.

This is pretty weird, because 3.8.0-27 and other kernels installed from "repo" (ie, using apt), does break, easily. Will test latest from mainline and report back.

[Natalia Bidart (nataliabidart)]

Tested kernel 3.11.0-031100rc4, was able to break audio (but it was not trivial). Attaching verbose pulseaudio log.

[Natalia Bidart (nataliabidart)]

Rebooted in kernel 3.9.0-030900 to re-try to break audio. Made everything I made in kernel 3.11.0-031100rc4 and more, can not break it. Attaching this second attempt's verbose pulseaudio log.

[Natalia Bidart (nataliabidart)]

Trying to make some sense of the results from earlier comments, installed kernel 3.8.0-19-generic from precise repo, booted into it, and was able to break audio. Verbose pulseaudio log attached.

[Joseph Salisbury (jsalisbury)]

It sounds like a Sauce patch could be the cause of the regression. Can you test the following Ubuntu kernels on your current Precise install:

Current Precise: https://launchpad.net/ubuntu/precise/+source/linux/3.2.0-51.77
Current Quantal: https://launchpad.net/ubuntu/quantal/+source/linux/3.5.0-37.58
Current Raring: https://launchpad.net/ubuntu/raring/+source/linux/3.8.0-27.40

The kernel .debs can be found under the section "Builds" for your arch on the previous pages.

I expect the bug to show up in the Current Quantal. If it does, we can test some earlier Quantal kernels to identify exactly when it was introduced.

[Natalia Bidart (nataliabidart)]

Report of my findings:

* kernel 3.2.0-51, works fine, audio playback did not break (attaching pulseaudio verbose log).

* kernel 3.5.0-37, works fine: I made the impossible to break audio, tried 2 different reboots, opened mumble while playing music, installed skype on top of that and use it, audio did not break. There was some glitches between skype and mumble stepping into each other (as in, the first one to start would "lock" the audio device and the other will not start properly), but closing one of the programs will make the othwr works fine. Attaching pulseaudio verbose log.

* kernel 3.8.0-27, breaks audio playback, rather easily. Just opening mumble and connecting to server, and starting a conversation will break it. Attaching pulseaudio verbose log, this is the biggest log file, lots of "Underrun!" in it.

Joseph: I know you expected kernel 3.5 to break audio but I did my best and could not break it. Any more tests I can do?

[Joseph Salisbury (jsalisbury)]

Thanks for testing, Natalia. So it sounds like this bug was introduced in a Sauce patch added in Raring. However, it also sounds like this is broken in the current mainline kernel, since you were able to reproduce the bug with 3.11-rc4.

It would be helpful if you could test some additional kernels. First the -proposed Raring kernel, just to confirm it wasn't fixed yet:
https://launchpad.net/ubuntu/+source/linux/3.8.0-28.41

If it still exists in -proposed, we have two options, we can bisect the Raring kernels to find the commit that introduced the regrssion and/or we can bisect the current mainline kernel to find the commit that breaks it there. It may be the same patch that breaks audio in both instances.

It might be best to find the commit that breaks this in mainline first, since it is going to affect Saucy and we can work with upstream to get it fixed before release. We can then see if that same patch was applied to Raring as a Sauce patch.

Can you test the v3.11-rc1 kernel to see if the bug also exists there:
http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.11-rc1-saucy/

If 3.11-rc1 is good, can you test v3.11-rc2:
http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.11-rc2-saucy/

If 3.11-rc2 is good, we should next test v3.11-rc3:
http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.11-rc3-saucy/

Based on these tests, we should know the two 3.11 kernels we should bisect between.

[Natalia Bidart (nataliabidart)]

Tested 3.8.0-28.41, audio playback broke. Then, tested 3.11-rc1 and audio playback easily broke, so I did not test any other kernel. Attaching pulseaudio verbose log of this two runs.

Let me know how to continue from here, and thanks a lot for the help!

[Natalia Bidart (nataliabidart)]

Given that 3.11-rc1 broke, I tried to gain some debugging time and went ahead and downloaded latest kernel 3.10 from mainline, since I didn't try that version yet. Tested 3.10.5-031005 and audio did not break (used mumble, audacity, rhythmbox and skype at the same time). Mumble showed a couple of messages like this:

PulseAudio: Sinks or inputs changed (inserted or removed sound card)
ALSAAudio: Non-critical: w=snd_pcm_writei(pcm_handle, outbuff, period_size): Broken pipe

but audio playback never stopped working. Attaching verbose log from pulseaudio.

[Joseph Salisbury (jsalisbury)]

Can you next test upstream v3.10 final:
http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.10-saucy/

[Natalia Bidart (nataliabidart)]

Tested kernels:

3.10.0 -> audio breaks
3.10.6 -> audio breaks

Attaching pulseaudio verbose log for all these 2.

[Joseph Salisbury (jsalisbury)]

Per comment #63, this bug does not exist in v3.9 final. We now also know v3.10 final has the bug. That would indicate this bug was introduced in one of the v3.9 release candidates.

In comment #58 you mentined that you tested v3.9-rc1, but were unable to test due to issues with the desktop, so we should try some other release candidates. Can you test v3.9-rc4:

http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.9-rc4-raring/

[Raymond (superquad-vortex2)]

Advanced information - PCI Vendor/Device/Subsystem ID's
!!--------------------------------------------------------

00:1b.0 0403: 8086:3b56 (rev 05)
 Subsystem: 8086:0033

is there any relationship with disable runtime PM for Intel 5 Series/3400 ?

http://git.kernel.org/cgit/linux/kernel/git/tiwai/sound.git/commit/sound/pci/hda/hda_intel.c?id=2c1350fdeaefefe1a149d3b083383409f43f0daa

[Natalia Bidart (nataliabidart)]

Joseph:

During the last weekend I upgraded my main installation from raring to saucy, and with that a new grub was installed (2.00-17ubuntu1).

So now, when I reboot into my secondary partition (the precise installation I use to test these kernels), despite I install the new kernels, the grub menu does not show the newly added kernels for the precise install.

Any ideas?

[Natalia Bidart (nataliabidart)]

I rebooted into saucy and ran sugo grub-update, that made the new kernels show in the grub menu.

Tested kernel 3.9-rc4 and audio broke when I used mumble. Attaching pulseaudio verbose log.

[Joseph Salisbury (jsalisbury)]

Maybe try re-installing the kernels from the partition you use for these tests? Or you could also try running update-grub again?

[Natalia Bidart (nataliabidart)]

On Tue, Aug 20, 2013 at 5:14 PM, Joseph Salisbury <
<email address hidden>> wrote:

> Maybe try re-installing the kernels from the partition you use for these
> tests? Or you could also try running update-grub again?
>

I fixed by running update-grub from my saucy, as per my last comment on the
bug.

Tested the 3.9-rc4 and audio broke.

[Joseph Salisbury (jsalisbury)]

I built a test kernel with commit 2c1350fdeaefefe1a149d3b083383409f43f0daa reverted, as suggested in comment #78. Can you test this kernel to see if it still exhibits this bug? It can be downloaded from:

http://kernel.ubuntu.com/~jsalisbury/lp1201528/

[Natalia Bidart (nataliabidart)]

Sorry for the delay on testing this, I almost had no chance to reboot into my "dummy" precise installation. Report as follow:

Tested kernel 3.11.0-3-generic #6~lp1201528v1, audio broke as soon as I logged in into mumble. Attaching pulseaudio verbose log.

For future debugging, if possible, it would be easier for me to have a list of several kernels to tests (in order), so I can do all the possible testing in one "reboot" ("a reboot" meaning leaving my main system).

Thanks!

[Joseph Salisbury (jsalisbury)]

The 3.11 final kernel is now available and the Ubuntu kernel has been rebased to it. Before we continue the bisect, can you test the latest Ubuntu updates and see if this bug still exists or not? The Ubuntu kernel you should test is version 3.11.0-7

[Natalia Bidart (nataliabidart)]

Hello Joseph,

Sorry for not replying sooner, I was on a leave for 2 weeks, so I'm catching up with this.

Question: I'm not sure where to get kernel 3.11.0-7 from, on my saucy installation I have 3.11.0-11-generic, and on the precise installation I'm not sure where to install it from (I can not find 3.11.0-7 on http://kernel.ubuntu.com/~kernel-ppa/mainline/).

In any case, I tested audio on my saucy main install with

nessita@dali:~$ uname -a
Linux dali 3.11.0-11-generic #17-Ubuntu SMP Tue Oct 1 19:42:04 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux

and audio playback breaks as soon as I open mumble and join a channel.

Happy to start the bisect with the kernels you'll link.

[Joseph Salisbury (jsalisbury)]

The 3.12-rc4 kernel is now out. Can you first test this kernel, to see if the bug was already fixed upstream:
http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.12-rc4-saucy/

If the bug still exists in v3.12-rc4, then we should continue bisecting. Per comment #80, 3.9-rc4 exhibits the bug. Can you also test the following kernel:

v3.9-rc2: http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.9-rc2-raring/
v3.9-rc3: http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.9-rc3-raring/

[Natalia Bidart (nataliabidart)]

I'm not being able to login to kernel v3.9-rc2, there is a video issue that prevents me to login. Shall I test older versions?

[Joseph Salisbury (jsalisbury)]

It might be difficult to bisect in 3.9 with a video issue. We know the issue was also introduced in a 3.8 stable update, so lets try bisecting in that tree. You confirmed 3.8.0-27 has the but, and it is based off of upstream 3.8.13.3. Can you test the following earlier upstream kernels:

3.8.13.1: http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.8.13.1-raring/
3.8.13.2: http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.8.13.2-raring/

[Natalia Bidart (nataliabidart)]

Tested kernel v3.8.13-03081302-generic and audio playback broke: it was not quick, had to use mumble for about 10 minutes and request the other party to talk a lot.

[Natalia Bidart (nataliabidart)]

Tested kernel 3.8.13-03081301-generic and was NOT able to break audio playback. Used mumble for 15 minutes getting music broadcasted, and audio never stopped working.

[Natalia Bidart (nataliabidart)]

Hey, is there any update on this bug? I'm happy to test more kernels, and I still can not use audio on my main computer.

[Raymond (superquad-vortex2)]

you mplayer did not require low latency playback

Jul 24 09:20:47 dali pulseaudio[2800]: [pulseaudio] protocol-native.c: Requested tlength=250.00 ms, minreq=20.00 ms
Jul 24 09:20:47 dali pulseaudio[2800]: [pulseaudio] protocol-native.c: Traditional mode enabled, modifying sink usec only for compat with minreq.
Jul 24 09:20:47 dali pulseaudio[2800]: [alsa-sink-ALC889 Analog] alsa-sink.c: Expected sleep: 160.98ms, real sleep: 0.22ms (diff -160.76 ms)

but mumble require low latency playback

1383327649.384264-1444531632"
( 7.432| 0.000) I: [pulseaudio] sink-input.c: module-stream-restore.id = "sink-input-by-media-role:phone"
( 7.433| 0.000) I: [pulseaudio] protocol-native.c: Requested tlength=60.00 ms, minreq=10.00 ms
( 7.433| 0.000) D: [pulseaudio] protocol-native.c: Adjust latency mode enabled, configuring sink latency to half of overall latency.

do you know whether mumble receive enough voice audio from the internet voice chat ?

under run can occur if your sound card playback faster than your partner 's sound card capture

or

buffer overflow when your sound card playback slower than your partner's sound card capture

[Joseph Salisbury (jsalisbury)]

It sounds like we can bisect between 3.8.13.1 and 3.8.13.2 per your comments in #92 and #93. I'll start the bisect and post a test kernel shortly.

[Joseph Salisbury (jsalisbury)]

I started a kernel bisect between 3.8.13.1 and 3.8.13.2 . The kernel bisect will require testing of about 7-10 test kernels.

I built the first test kernel, up to the following commit:
90856193fc94eec039ab17bbde2906d6fc6919c6

The test kernel can be downloaded from:
http://kernel.ubuntu.com/~jsalisbury/lp1201528

Can you test that kernel and report back if it has the bug or not. I will build the next test kernel based on your test results.

Thanks in advance

[Natalia Bidart (nataliabidart)]

Hello Joseph!

Sorry it too so long for me to respond, but I was traveling and I wasn't near the desktop PC that is suffering from this bug.

I installed the kernel you compiled for me and I was able to reproduce the audio playback issue. Attaching the verbos pulseaudio log.

I'm now using the desktop PC regularly, so I should be able to test new kernel much, much faster.

[Natalia Bidart (nataliabidart)]

Just to be sure I'm running the right kernel, this is uname -a:

foobar@dale:~$ uname -a
Linux dale 3.8.13-03081301-generic #201312101408 SMP Tue Dec 10 19:46:32 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux

[Joseph Salisbury (jsalisbury)]

Since it's been some time since we started the bisect, would it be possible for you to apply the latest updates and install the latest kernel to see if this bug still exists?

[Natalia Bidart (nataliabidart)]

Sure!

Would that be the latest for precise, or the kernel 3.14 from the mainline directory ?

[Joseph Salisbury (jsalisbury)]

I would say just apply the latest updates for whatever release you are running. So either precise, saucy or trusty.

[Natalia Bidart (nataliabidart)]

Tested kernel from saucy in my daily desktop:

nessita@dali:~$ uname -a
Linux dali 3.11.0-19-generic #33-Ubuntu SMP Tue Mar 11 18:48:34 UTC 2014 x86_64 x86_64 x86_64 GNU/Linux

Audio keeps breaking. Updating to trusty soon to re-check.

[Raymond (superquad-vortex2)]

you may try limiting the Max volume of mixer 0x0c, 0x0d, 0x0e, 0x0f and 0x26 to 0dB instead of 2dB by overwrite nsteps from 0x40 to 0x3e

Amp-Out caps: ofs=0x3e, nsteps=0x40, stepsize=0x03, mute=0
  Amp-Out vals: [0x3a 0x3a]

https://git.kernel.org/cgit/linux/kernel/git/tiwai/sound.git/log/sound/pci/hda?qt=grep&q=fix+max+pcm

[Natalia Bidart (nataliabidart)]

Re tested on latest trusty:

Linux dali 3.13.0-24-generic #46-Ubuntu SMP Thu Apr 10 19:11:08 UTC 2014 x86_64 x86_64 x86_64 GNU/Linux

Audio broke as soon as I opened a hangout.

[Raymond (superquad-vortex2)]

http://www.intel.com/content/www/us/en/motherboards/desktop-motherboards/desktop-board-dp55wg.html

10-channel (7.1+ 2 independent multi-streaming) audio subsystem with five analog audio outputs and two optical S/PDIF digital audio for input and output using the Realtek * ALC889 audio codec

do it has RCA coaxial ?

Node 0x11 [Pin Complex] wcaps 0x400300: Mono Digital
  Control: name="SPDIF Phantom Jack", index=0, device=0
  Pincap 0x00000010: OUT
  Pin Default 0x01442130: [Jack] SPDIF Out at Ext Rear
    Conn = RCA, Color = Grey
    DefAssociation = 0x3, Sequence = 0x0
    Misc = NO_PRESENCE
  Pin-ctls: 0x40: OUT
  Connection: 1
     0x10

Node 0x1e [Pin Complex] wcaps 0x400300: Mono Digital
  Control: name="SPDIF Phantom Jack", index=1, device=0
  Pincap 0x00000010: OUT
  Pin Default 0x01451140: [Jack] SPDIF Out at Ext Rear
    Conn = Optical, Color = Black
    DefAssociation = 0x4, Sequence = 0x0
    Misc = NO_PRESENCE
  Pin-ctls: 0x40: OUT
  Connection: 1
     0x06
Node 0x1f [Pin Complex] wcaps 0x400280: Mono Digital
  Control: name="SPDIF In Phantom Jack", index=0, device=0
  Pincap 0x00000020: IN
  Pin Default 0x01c51170: [Jack] SPDIF In at Ext Rear
    Conn = Optical, Color = Black
    DefAssociation = 0x7, Sequence = 0x0
    Misc = NO_PRESENCE
  Pin-ctls: 0x20: IN
  Unsolicited: tag=00, enabled=0

https://git.kernel.org/cgit/linux/kernel/git/tiwai/sound.git/commit/sound/pci/hda/patch_realtek.c?id=6732bd0d15893d20359396309adc238e0af10b58

this patch add alc889_coef_init for motherboard with 6 jacks

https://git.kernel.org/cgit/linux/kernel/git/tiwai/sound.git/commit/sound/pci/hda/patch_realtek.c?id=20b67dddcc5f29d3d0c900225d85e0ac655bc69d

https://git.kernel.org/cgit/linux/kernel/git/tiwai/sound.git/commit/sound?id=68ef0561efe494143516df38c03a16b837b8e79c

seem add back only to DC58SO

 SND_PCI_QUIRK(0x8086, 0x0022, "DX58SO", ALC889_FIXUP_COEF),

[Joseph Salisbury (jsalisbury)]

This bug was nominated against a series that is no longer supported, ie saucy. The bug task representing the saucy nomination is being closed as Won't Fix.

This change has been made by an automated script, maintained by the Ubuntu Kernel Team.

[Rolf Leggewie (r0lf)]

raring has seen the end of its life and is no longer receiving any updates. Marking the raring task for this ticket as "Won't Fix".

[Rolf Leggewie (r0lf)]

saucy has seen the end of its life and is no longer receiving any updates. Marking the saucy task for this ticket as "Won't Fix".

[Launchpad Janitor (janitor)]

[Expired for linux (Ubuntu) because there has been no activity for 60 days.]

[Launchpad Janitor (janitor)]

[Expired for alsa-driver (Ubuntu) because there has been no activity for 60 days.]

[Natalia Bidart (nataliabidart)]

Hello,

I'm now running latest kernel 3.19 from vivid:

Linux dali 3.19.0-16-generic #16-Ubuntu SMP Thu Apr 30 16:09:58 UTC 2015 x86_64 x86_64 x86_64 GNU/Linux

And this issue is still present for me. I grabbed debug pulseaudio logs again, see attachment. Audio was completely lost when messages such as this started to show up in logs:

( 91.896| 9.804) I: [pulseaudio] module-device-restore.c: Synced.
( 117.291| 25.395) I: [alsa-sink-ALC889 Analog] alsa-sink.c: Underrun!
( 117.291| 0.000) I: [alsa-sink-ALC889 Analog] alsa-sink.c: Increasing minimal latency to 4.00 ms
( 117.291| 0.000) D: [alsa-sink-ALC889 Analog] alsa-sink.c: Latency set to 20.00ms
( 117.291| 0.000) D: [alsa-sink-ALC889 Analog] alsa-sink.c: hwbuf_unused=62008
( 117.291| 0.000) D: [alsa-sink-ALC889 Analog] alsa-sink.c: setting avail_min=15944
( 117.292| 0.000) D: [alsa-sink-ALC889 Analog] alsa-sink.c: Latency set to 20.00ms
( 117.292| 0.000) D: [alsa-sink-ALC889 Analog] alsa-sink.c: hwbuf_unused=62008
( 117.292| 0.000) D: [alsa-sink-ALC889 Analog] alsa-sink.c: setting avail_min=15944
( 117.487| 0.195) I: [alsa-sink-ALC889 Analog] alsa-sink.c: Underrun!
...

[Brad Figg (brad-figg)]

This bug is missing log files that will aid in diagnosing the problem. From a terminal window please run:

apport-collect 1201528

and then change the status of the bug to 'Confirmed'.

If, due to the nature of the issue you have encountered, you are unable to run this command, please add a comment stating that fact and change the bug status to 'Confirmed'.

This change has been made by an automated script, maintained by the Ubuntu Kernel Team.

[Natalia Bidart (nataliabidart)]

apport information

[Natalia Bidart (nataliabidart)]

apport information

[Natalia Bidart (nataliabidart)]

apport information

[Natalia Bidart (nataliabidart)]

apport information

[Natalia Bidart (nataliabidart)]

apport information

[Natalia Bidart (nataliabidart)]

apport information

[Natalia Bidart (nataliabidart)]

apport information

[Natalia Bidart (nataliabidart)]

apport information

[Natalia Bidart (nataliabidart)]

apport information

[Natalia Bidart (nataliabidart)]

apport information

[Natalia Bidart (nataliabidart)]

apport information

[Natalia Bidart (nataliabidart)]

apport information

[Natalia Bidart (nataliabidart)]

apport information

[Natalia Bidart (nataliabidart)]

apport information

[Natalia Bidart (nataliabidart)]

apport information

[Brad Figg (brad-figg)]

This change was made by a bot.

[Joseph Salisbury (jsalisbury)]

Would it be possible for you to test the latest upstream kernel? Refer to https://wiki.ubuntu.com/KernelMainlineBuilds . Please test the latest v4.1 kernel[0].

If this bug is fixed in the mainline kernel, please add the following tag 'kernel-fixed-upstream'.

If the mainline kernel does not fix this bug, please add the tag: 'kernel-bug-exists-upstream'.

If you are unable to test the mainline kernel, for example it will not boot, please add the tag: 'kernel-unable-to-test-upstream'.
Once testing of the upstream kernel is complete, please mark this bug as "Confirmed".

Thanks in advance.

[0] http://kernel.ubuntu.com/~kernel-ppa/mainline/v4.1-rc3-vivid/

[Natalia Bidart (nataliabidart)]

As requested, tested kernel 4.1.0-040100rc3-generic. Audio playback is still lost.

See attached verbose output from pulseaudio, including the line that makes all sound go away:

(2356.093| 0.000) D: [alsa-sink-ALC889 Analog] protocol-native.c: max_request changed, trying to update from 70272 to 71292.
(2356.093| 0.000) D: [alsa-sink-ALC889 Analog] protocol-native.c: Notifying client about increased tlength
(2356.093| 0.000) D: [alsa-sink-ALC889 Analog] alsa-sink.c: Latency set to 341.33ms
(2356.093| 0.000) D: [alsa-sink-ALC889 Analog] alsa-sink.c: hwbuf_unused=4
(2356.093| 0.000) D: [alsa-sink-ALC889 Analog] alsa-sink.c: setting avail_min=737
(2356.268| 0.175) I: [alsa-sink-ALC889 Analog] alsa-sink.c: Underrun!
(2356.268| 0.000) I: [alsa-sink-ALC889 Analog] alsa-sink.c: Increasing wakeup watermark to 331.31 ms
(2356.447| 0.178) I: [alsa-sink-ALC889 Analog] alsa-sink.c: Underrun!

[Joseph Salisbury (jsalisbury)]

It looks like we were bisecting between 3.8.13.1 and 3.8.13.2 a while back. It might be best to pick that up again. I'll set up a bisect and post a test kernel shortly.

[Natalia Bidart (nataliabidart)]

Thanks Joseph. Just FYI, I will have access to this computer for 2 weeks, then I may not be able to access it anymore for a month or two. So as soon as you have the kernels for bisect, I'll be happy to try those.

[Joseph Salisbury (jsalisbury)]

I restarted the bisect between v3.8.13.1 and 3.8.13.2.

I built a test kernel up the following commit:
37e78b62f9895480779ad92dd110a60f6ce51847

The test kernel can be downloaded from:
http://kernel.ubuntu.com/~jsalisbury/lp1201528/

Can you test this kernel and see if it exhibits the bug?

[Natalia Bidart (nataliabidart)]

Tested kernel 3.8.13-03081301-generic

It was not easy, but audio did break. I have to use hangouts for a while. Attaching pulseaudio verbose log.

[Natalia Bidart (nataliabidart)]

I see from comment #93 that back then I couldn't break audio, I guess because I did not try hangout at the time.

From now on I will use hangout for every test, since it seems to be the most effective way of breaking audio. The main issue is that if audio does not break, it may be because I did not stress audio enough, not because is actually working :-/

[Joseph Salisbury (jsalisbury)]

Can you confirm the 3.8.13.1 kernel does not have this bug. That way we know we are bisecting between the correct versions.

3.8.13.1: http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.8.13.1-raring/

[Natalia Bidart (nataliabidart)]

From IRC:

16:20 < nessita> jsalisbury, just wanted to sync about the bisect we're doing for LP: #1201528, since I'm now in 3.8.13.1 and audio died with the underrun error. So I re-read all the comments and noticed that in commnet #39 (https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1201528/comments/39) I reproduced the issue with kernel 3.8.0-27-generic

16:21 < nessita> I'm now installing latest 3.7 from http://kernel.ubuntu.com/~kernel-ppa/mainline/ which is 3.7.10 and see if audio breaks, will report in the bug

So, in summary:

Kernel 3.8.13.1 -> AUDIO BREAKS

Kernel 3.7.10 -> audio does not break

Will try other versions of 3.8.

[Natalia Bidart (nataliabidart)]

As per anoter IRC conversation,

* tested kernel 3.8.0 final (3.8.0-030800) and AUDIO BREAKS (it was hard to reproduced, seems like the underruns are less "extreme"). See verbose pulseaudio log please.

* tested kernel 3.7.0 final (3.7.0-030700) and audio does not break (attached log the same).

[Joseph Salisbury (jsalisbury)]

We should also test some of the 3.8 release candidates to know which exact one introduced the regression. Once we know that, we can start the bisect.

Can you test the following kernels and post back? We are looking for the first kernel version that exhibits this bug:

v3.8-rc4: http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.8-rc4-raring/

If v3.8-rc4 does not exhibit the bug then test v3.8-rc6:
v3.8-rc6: http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.8-rc6-raring/

If v3.8-rc4 does exhibit the bug then test v3.8-rc2:
v3.8-rc2: http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.8-rc2-raring/

Thanks in advance!

[Natalia Bidart (nataliabidart)]

Hello! Reporting back:

Tested kernel 3.8.0-030800rc4 and audio BREAKS

Then, as requested, tested kernel 3.8.0-030800rc4 and audio BREAKS

Attaching pulseaudio verbose logs. In both cases audio breaks when pulseaudio reports:

[alsa-sink-ALC889 Analog] alsa-sink.c: Underrun!

Side note: Just to be super sure, what's the kernel I should have installed to cross check it keeps working? should that be 3.7.0 or 3.7.10?

Thanks

[Natalia Bidart (nataliabidart)]

Sorry, the second rc4 is obviously rc2, see attached log for confirmation, I always run the pulseaudio daemon with:

LANG=C pulseaudio -vvvv --log-time=1 > ~/pulseverbose-`uname -r`.log 2>&1

[Joseph Salisbury (jsalisbury)]

Can you now test v3.8-rc1? If it exhibits the bug we can bisect between 3.7 final and 3.8-rc1. The v3.8-rc1 kernel can be downloaded from:

http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.8-rc1-raring/

[Raymond (superquad-vortex2)]

[alsa-sink-ALC889 Analog] alsa-sink.c: Underrun!

this seem not alsa driver, it occur whe client did not supply audio and server has nothing to write

in theory, the server have to call snd_pcm_foreward to recover

http://cgit.freedesktop.org/pulseaudio/pulseaudio/log/src/modules/alsa/alsa-sink.c?qt=grep&q=underrun

[Raymond (superquad-vortex2)]

you can add pa_alsa_dump() immediately after pa_log_info("Underrun!") to verify appl_ptr is behind hwptr since those pulseaudio clients did not supply sufficient audio to the server

the server seem have no ways to recover since it can not write any data which affect the timing
it just hope the client can supply the missing data at later time

[Natalia Bidart (nataliabidart)]

Audio BREAKS with 3.8.0-030800rc1-generic, attaching verbose log for pulseaudio

[Joseph Salisbury (jsalisbury)]

I started a kernel bisect between v3.7 final and v3.8-rc1. The kernel bisect will require testing of about 7-10 test kernels.

I built the first test kernel, up to the following commit:
6be35c700f742e911ecedd07fcc43d4439922334

The test kernel can be downloaded from:
http://kernel.ubuntu.com/~jsalisbury/lp1201528

Can you test that kernel and report back if it has the bug or not? I will build the next test kernel based on your test results.

Thanks in advance

[Natalia Bidart (nataliabidart)]

Hello Joseph!

Thanks a lot for the bisect work. As I mentioned earlier in the backlog, I'm not at home with the troubling computer for 2 months, but I commit to test this kernel as soon as I get back. I'm very interested to have this issue sorted out (or at least debugged) so I will test the new kernel ASAP, but that will be in early August.

Thanks!

[Andy Whitcroft (apw)]

This bug was nominated against a series that is no longer supported, ie vivid. The bug task representing the vivid nomination is being closed as Won't Fix.

This change has been made by an automated script, maintained by the Ubuntu Kernel Team.

[Natalia Bidart (nataliabidart)]

Hello all,

First of all, sorry I did not try the kernel that Joseph provided over a year ago, I've been traveling and not having my desktop computer close.

I'm now settled in a location with my desktop computer as my daily and main computer, and I'm still seeing this bug occur.

I'm running ubuntu xenial 16.04 with the stock kernel:

nessita@dali:~$ uname -a
Linux dali 4.4.0-109-generic #132-Ubuntu SMP Tue Jan 9 19:52:39 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux

And audio breaks just like I've posted earlier:

( 71.143| 0.000) I: [alsa-sink-ALC889 Analog] alsa-sink.c: Underrun!

I've recorded a new log file from running pulseaudio with verbose output and I've attached it to this bug.