Easynote MX65, AD1986A, No sound output in speakers/headphones, some with tricks

[LukasThyWalls]

I have a fresh install of KUbuntu 14.10 with this problem and i come of a KUbuntu 13.10 without problems.

I don't have any sound from speakers or headphone, i think the driver isn't loaded, but the microphone works, it records. The sound playback works too, but the sound doesn't go through speakers or headphones.

I was testing with a lot of LiveCD KUbuntu when the problem comes. With LiveCDs KUbuntu 12.04.5 LTS, 12.10, 13.04 the sound works, they configs the sound system as "HDA Intel AD198x". Since LiveCDs 13.10, 14.04.1 LTS and, of course, 14.10 Live CD and my install, i have this problem, because the sound system is now "AD1986A". I will attach the alsa-info output for all this systems in my computer.

But with HDA_Analyzer http://www.alsa-project.org/main/index.php/HDA_Analyzer i found the details of the problem: i can create a python script to brought back the sound working in a basic mode, like as previously. The details are activating EAPD in Card 0, Codec 0, Node 0x1b PIN (In my computer, it could be different in another) what brought back the sound to the speakers, and unmuting "Output Amplifier" in Card 0, Codec 0, Node 0x1a VAL[0] and VAL[1] to brought back sound to the headphones when they are plugged. I will attach too the output of this diff and the python script created by HDA_Analyzer for my system.

At least, the basic working state should be back, because with a untoching fresh install there isn't any sound. There are related bugs here and other post in internet, but i collected all information i found and can give more if it's necessary.

For additional information: Searching the solution, i tested too all the models availables for HDA Intel AD1986A (editing /etc/modprobe.d/alsa-base.conf with AD1986A models /usr/share/doc/alsa-base/driver/HD-Audio-Models.txt.gz) and i found the default one doesn't fit my system and anothers are better, but have problems of no sound too and i don't found how to fix these problem. When this bug is corrected, i want to help to found a better model for my computer, i don't know how now, but first it brought back the sound. Also, i test with snd-hda-tools https://launchpad.net/~diwic/+archive/ubuntu/hda and with hda-jack-sense-test the Green Line Out sense doesn't work (always NO present) but the Pink Mic sense works without problems.

ProblemType: Bug
DistroRelease: Ubuntu 14.10
Package: alsa-base 1.0.25+dfsg-0ubuntu4
ProcVersionSignature: Ubuntu 3.16.0-24.32-generic 3.16.4
Uname: Linux 3.16.0-24-generic i686
NonfreeKernelModules: nvidia
ApportVersion: 2.14.7-0ubuntu8
Architecture: i386
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC0: thywalls 2380 F.... pulseaudio
CurrentDesktop: KDE
Date: Sun Nov 2 16:24:36 2014
InstallationDate: Installed on 2014-10-30 (3 days ago)
InstallationMedia: Kubuntu 14.10 "Utopic Unicorn" - Release i386 (20141022.1)
PackageArchitecture: all
SourcePackage: alsa-driver
Symptom: audio
Title: Bluetooth sound card not detected
UpgradeStatus: No upgrade log present (probably fresh install)
dmi.bios.date: 09/28/2006
dmi.bios.vendor: American Megatrends Inc.
dmi.bios.version: T12JNC.206
dmi.board.asset.tag: To Be Filled By O.E.M.
dmi.board.name: EasyNote_MX65
dmi.board.vendor: Packard Bell BV
dmi.board.version: 1.0
dmi.chassis.type: 10
dmi.chassis.vendor: Packard Bell BV
dmi.modalias: dmi:bvnAmericanMegatrendsInc.:bvrT12JNC.206:bd09/28/2006:svnPackardBellBV:pnEasyNote_MX65:pvrPB63E03104:rvnPackardBellBV:rnEasyNote_MX65:rvr1.0:cvnPackardBellBV:ct10:cvr:
dmi.product.name: EasyNote_MX65
dmi.product.version: PB63E03104
dmi.sys.vendor: Packard Bell BV

[Raymond (superquad-vortex2)]

seem missing hp nodes

is this a notebook?

 18.446135] sound hdaudioC0D0: autoconfig: line_outs=1 (0x1b/0x0/0x0/0x0/0x0) type:line
[ 18.446141] sound hdaudioC0D0: speaker_outs=0 (0x0/0x0/0x0/0x0/0x0)
[ 18.446144] sound hdaudioC0D0: hp_outs=0 (0x0/0x0/0x0/0x0/0x0)
[ 18.446147] sound hdaudioC0D0: mono: mono_out=0x0
[ 18.446150] sound hdaudioC0D0: dig-out=0x25/0x0
[ 18.446153] sound hdaudioC0D0: inputs:
[ 18.446156] sound hdaudioC0D0: Mic=0x1f
[ 18.446159] sound hdaudioC0D0: CD=0x22

[LukasThyWalls (thywalls)]

Packard Bell Easynote MX65-042 is a Notebook. It have only two jacks: Green Line Out and Pink Mic. It have too the internal speakers, and an internal mic what never was detected and used in the past.

In HDA_Analyzer, I can view the HP option in "Widget Control" in both Card 0, Codec 0, Node 0x1b PIN and Card 0, Codec 0, Node 0x1a, but activating/deactivating them don't make any difference. I can make it work with the option related above in report, and with HDA_Analyzer or its script.

Maybe related: How i said, i tried all AD1986A models config names, including "laptop-automute (2-channel with EAPD and HP-automute)", but all "laptop", "laptop-eadp" and "laptop-automute" (more information here /usr/share/doc/alsa-base/driver/HD-Audio-Models.txt.gz) work in the same way: Only with one of these options models name configured in /etc/modprobe.d/alsa-base.conf, without using HDA_Analyzer, i have sound with the headphones, but i don't have any sound through the speakers, and with HDA_Analyzer i haven't been able to bring back the sound through speakers with any option i tested.
However, and for that i said in the last paragraph that "the default model config doesn't fit my computer" in the report, with the model option "laptop-eapd (2-channel with EAPD)", it detect all the conectors (like it said in KDE Phonon configuration) when i plug them: Speakers and Headphones for output, and Mic and Internal Mic for input, this last one i never see it before and works, but i can't bring back the sound to the speakers, only with the headphones. With the default config without models, the only connectors i have is "Analog Input" and "Microphone" (The Pink mic Jack, the internal mic doesn't do nothing), but i can bring back the sound to normal like always with HDA_Analyzer and its script. Like i said, first, i only want to bring back a normal and working basic configuration, and after, if i can make bring back the Internal Mic working, it would be better, but first it's first.

[LukasThyWalls (thywalls)]

Another detail i found today:

I have now using the sound system after using the HDA_Analyzer script, and because that i have sound, and i see the "Master" volume do nothing. I am playing a song in VLC, and with alsamixer, i can change the volume with "PCM", but with "Master" nothing happens.

It's evident that the driver can work with my soundcard but it's totally misconfigured, the interlan options and nodes, and there isn't a model option what fits my system when it changes to AD1986A, and when it's AD198x at least worked with the basic functionality, but with the new not.

I will test more with HDA_Analyzer and maybe i found more nodes and options what makes the sound works better, but i need to learn more about what i'm doing with this program.

[Raymond (superquad-vortex2)]

you need early patching or hda jack retask to change

node 0x1a to [jack] hp and

node from [jack] line out to [fixed] speaker at int

[Raymond (superquad-vortex2)]

https://bugzilla.kernel.org/show_bug.cgi?id=67481

[LukasThyWalls (thywalls)]

First, i found today that the "Master" Volume works with the speakers as expected, and doesn't with the headphones like i said yesterday, when i missed this detail.

Second, i found too other bug reports related to this, but that bugzilla.kernel report it's very near, but i can't trust it because they are talking about other specific notebook, and i don't know if that work for me. But i will to read and test all info to apply and to find solutions to my problem. And maybe, bugzilla kernel should be the site for this report, or not?

Third, your solution with hdajackretask (with advanced override activated), i don't have in jack any "HP" option, and there isn't any 0x1a connected. Green Line Out is 0x1b. Maybe i need to config to use another AD1986A model (laptop-automute?) and not the default? Maybe PINs 0x1a (headphones) and 0x1b (speakers) are inverted and 0x1a jack detection isn't detection properly (see hda-jack-sense-test above)?

I don't know it it's related: In hda_analyzer, 0x1a HP option is enables, but in 0x1b HP option is disables, but both don't make any differences.

At least the only thing i have to came now for the basic functionality is the mixer for the headphones, because all the other changes in hda_analyzer are working. I want to put the original text dump of hda_analyzer in the affected pins, it could be useful:

-----

Original and default AD1986A Configuration in HDA_Analyzer for PIN 0x1a (Change headphones) and PIN 0x1b (Change speakers)

Node 0x1a [Pin Complex] wcaps 0x400185: Stereo Amp-Out
  Amp-Out caps: ofs=0x1f, nsteps=0x1f, stepsize=0x05, mute=1
  Amp-Out vals: [0x9f 0x9f]
  Pincap 0x0000001f: OUT HP Detect Trigger ImpSense
  Pin Default 0x4221401f: [N/A] HP Out at Ext Front
    Conn = 1/8, Color = Green
    DefAssociation = 0x1, Sequence = 0xf
  Pin-ctls: 0xc0: OUT HP
  Unsolicited: tag=0x00, enabled=0
  Connection: 1
     0x0a

Node 0x1b [Pin Complex] wcaps 0x400185: Stereo Amp-Out
  Control: name="Master Playback Volume", index=0, device=0
    ControlAmp: chs=3, dir=1, idx=0, ofs=0
  Control: name="Master Playback Switch", index=0, device=0
    ControlAmp: chs=3, dir=1, idx=0, ofs=0
  Control: iface="card", name="Line Out Jack", index=0, device=0
  Amp-Out caps: ofs=0x1f, nsteps=0x1f, stepsize=0x05, mute=1
  Amp-Out vals: [0x14 0x14]
  Pincap 0x0001001f: OUT HP EAPD Detect Trigger ImpSense
  EAPD 0x0:
  Pin Default 0x01014010: [Jack] Line Out at Ext Rear
    Conn = 1/8, Color = Green
    DefAssociation = 0x1, Sequence = 0x0
  Pin-ctls: 0x40: OUT
  Unsolicited: tag=0x03, enabled=1
  Connection: 1
     0x0b

-----

PIN 0x1a (Change headphones) and PIN 0x1b (Change speakers) text dump from HDA_Analyzer when i made the changes to fix sound problems (except the not-working mixer for headphones)
Changes:
- In "Node [0x1b] PIN": EAPD option activated (Bring back sound to speakers)
- In "Node [0x1a] PIN": Output Amplifier VAL[0] and VAL[1] Pass from muted to unmuted (Bring back sound to headphones)

Node 0x1a [Pin Complex] wcaps 0x400185: Stereo Amp-Out
  Amp-Out caps: ofs=0x1f, nsteps=0x1f, stepsize=0x05, mute=1
  Amp-Out vals: [0x9f 0x9f]
  Pincap 0x0000001f: OUT HP Detect Trigger ImpSense
  Pin Default 0x4221401f...

[LukasThyWalls (thywalls)]

In the last comment, the second 0x1a text dump is incorrect (the HDA_Analyzer edited one). Sorry

Node 0x1a [Pin Complex] wcaps 0x400185: Stereo Amp-Out
  Amp-Out caps: ofs=0x1f, nsteps=0x1f, stepsize=0x05, mute=1
  Amp-Out vals: [0x1f 0x1f]
  Pincap 0x0000001f: OUT HP Detect Trigger ImpSense
  Pin Default 0x4221401f: [N/A] HP Out at Ext Front
    Conn = 1/8, Color = Green
    DefAssociation = 0x1, Sequence = 0xf
  Pin-ctls: 0xc0: OUT HP
  Unsolicited: tag=0x00, enabled=0
  Connection: 1
     0x0a

Node 0x1b [Pin Complex] wcaps 0x400185: Stereo Amp-Out
  Control: name="Master Playback Volume", index=0, device=0
    ControlAmp: chs=3, dir=1, idx=0, ofs=0
  Control: name="Master Playback Switch", index=0, device=0
    ControlAmp: chs=3, dir=1, idx=0, ofs=0
  Control: iface="card", name="Line Out Jack", index=0, device=0
  Amp-Out caps: ofs=0x1f, nsteps=0x1f, stepsize=0x05, mute=1
  Amp-Out vals: [0x14 0x14]
  Pincap 0x0001001f: OUT HP EAPD Detect Trigger ImpSense
  EAPD 0x2: EAPD
  Pin Default 0x01014010: [Jack] Line Out at Ext Rear
    Conn = 1/8, Color = Green
    DefAssociation = 0x1, Sequence = 0x0
  Pin-ctls: 0x40: OUT
  Unsolicited: tag=0x03, enabled=1
  Connection: 1
     0x0b

[Raymond (superquad-vortex2)]

did you select show unconnected nodes and using advanced override option?

need to set boot override option since dynamic reconfig may not work on some hda codec

there common bug reports for notebook using ad1986a in bugzilla :

1) auto mute

2) auto mic - the topology of this codec is similar to ac97 codec's mic1 / mic2

take a look at ad1986a datasheet

in the functional block diagram

port A is the headphone

[Raymond (superquad-vortex2)]

https://git.kernel.org/cgit/linux/kernel/git/tiwai/sound.git/log/sound/pci/hda/patch_analog.c?qt=grep&q=ad1986a

[LukasThyWalls (thywalls)]

I'm not saying what is 0x1a node or 0x1b node or not, i'm only saying that changing this parameters in 0x1a node configuration in HDA_Analyzer made a change in headphones, and doing it in 0x1b do something in speakers, and with hda-jack-sense-test for example, i have always this output:

$ sudo hda-jack-sense-test
[sudo] password for ******:
Pin 0x1b (Green Line Out): present = No
Pin 0x1f (Pink Mic): present = Yes

(Both are plugged)

I copy the text dump in both cases, i don't understand a lot of things here, but i'm giving all the best information to give us light about the problem and for learn something and search what is happening in this case with this hardware. If only i had mixer control of the headphones volume, it would work in a reasonable functionaly, like in the past with AD198x, and the rest would be details.

But about your suggest to use hdajackretask, i don't have any "HP" option in any of the options (including the advanced), i only see the "HP" option like i said above in HDA_Analyzer. And 0x1a is not connected but because i haven't that "HP" option i don't know what to do. If you give me a more specific changes to test... In the first try, I mess i little with hdajackretask and the only thing i get is broke the KDE Mixer until i restart the computer, and in the second try, i have always this message "tee: /sys/class/sound/hwC0D0/reconfig: Dispositivo o recurso ocupado" (Busy) and this terminal output:

E: [pulseaudio] core-util.c: Home directory not accessible: Permiso denegado
E: [pulseaudio] main.c: No se ha podido detener el demonio: Conseguido
0x1a 0x02214110
0x1b 0x82170110
0x1c 0x41013012
0x1d 0x41019011
0x1e 0x511711f0
0x1f 0x02a19020
0x20 0x41813021
0x21 0x5997e1f0
0x22 0x9933112e
0x23 0x59b771f0
0x24 0x91f711f0
0x25 0x0145f1f0
1

(Although, if i close and enter again in the program the options are remembered, i don't know if it means they are saved anyway!)
I will upload hdajackretask' screenshot without any change to guide you, maybe it can help.

Remembering some testing i do, i put appart the default model option, and i add in /etc/modprobe.d/alsa-base.conf "options snd-hda-intel model=laptop-eapd". With this model i have, by default, sound through the headphones (with master volume control) and not through the speakers, and also appears Internal Mic along with the jack Mic what is always missing. I will upload alsa-info with laptop-eapd option.
The problems in this case, i don't have sound in speakers, but with HDA_Analyzer i can brought back it, in Node 0x1b PIN -> Output Amplifier -> Unmuting and rising volume for both VAL[0] and VAL[1], BUT if it should be with the green jack unplugged and the sound comes through the speakers, but if i plug it comes to headphones (it's normal) but if i unplug it again the sound doesn't come throught speakers again until i click with the mouse in another node in HDA_Analyzer and click again in Node 0x1b (some refreshing?). Another detail is when i do this with the green jack plugged, when i unplug it, the sound comes to the speakers for less than a second until it mutes. It remembered me the hda-jack-sense-test, when it said the Green Line Out is always NO present.

Befor...

[Raymond (superquad-vortex2)]

http://git.alsa-project.org/?p=alsa-tools.git;a=blob;f=hdajackretask/README;hb=HEAD

http://voices.canonical.com/david.henningsson/2011/11/29/turn-your-mic-jack-into-a-headphone-jack/

you need to check the override box of the node which you want to change

check "set as boot default" box

and reboot to test

[LukasThyWalls (thywalls)]

Thanks to give me the clue to make work hdajackretask. Now is working without problem.

I was testing a lot with it. First, i put the default model (no model) in alsa-base.conf, and forgot for now laptop-eapd.

With hdajackretask and the default model, i can replicate the behavior (or better) of the laptop-eapd config model:
- 0x1a set as Headphone
- 0x1d set as Internal Mic
(Both of them said there aren't connected, and i don't need advanced override)
With that, when the SO starts, i have the internal mic and the jack microphone (not a priority but there are they), and the headphones working, but i haven't the speakers. With hda_analyzer, activating EAPD in 0x1b i have sound, and in both work the mixer volume, BUT if i plug in the headphones and plug it out, the sound doesn't come through speakers again until i "refresh" the EAPD option in hda_analyzer (refresh -> deactivate/activate, click in another node and return back to 0x1b...) and it's exactly the same happened with laptop-eapd model.

I tested override 0x1b "Green Line Out" with hdajackretask, but i don't find any change what did something different (in some cases i lost all sound in speakers and headphones, but i supposed it's a normal incompatibility like said in hdajackretask readme). The first thing i change is try to override the jack detection of 0x1b putting it in "No Present" (Default is "Present"), because like said with "hda-jack-sense-test -a" it doesn't do nothing:

$ sudo hda-jack-sense-test -a
Pin 0x1a (Not connected): present = Yes
Pin 0x1b (Green Line Out): present = No
Pin 0x1c (Not connected): present = No
Pin 0x1d (Not connected): present = No
Pin 0x1e (Not connected): present = No
Pin 0x1f (Pink Mic): present = Yes
Pin 0x20 (Not connected): present = No
Pin 0x21 (Not connected): present = No
Pin 0x22 (Internal CD): present = No
Pin 0x23 (Not connected): present = No
Pin 0x24 (Internal Other): present = No
Pin 0x25 (Other SPDIF Out): present = No

(0x1a and 0x1f changes to NO when i unplug headphones jack and headphones mic jack)

But even with that i have no changes in the behavior of the sound. I will add the actual config, i now some things can be changed (Jack in 0x1a, for example) but with that i have the actual behavior..
What can i do? What can i change to give another behavior? How to fix that problem with i plug the jack headphones and lost the sound of the speakers (with hdajackretask or hda_analyzer) and make it permanent at least?

[Raymond (superquad-vortex2)]

https://bugzilla.kernel.org/enter_bug.cgi?product=Drivers

[Raymond (superquad-vortex2)]

have you look at last page of ad1986a datasheet and git log of patch_analog.c related to ad1986a?

some laptop seem to use 3stack instead of laptop model in the past since the mic select is quite complicated

[Raymond (superquad-vortex2)]

https://answers.launchpad.net/ubuntu/+source/alsa-driver/+question/247989

[LukasThyWalls (thywalls)]

- #23: With 3stack model, i have the same behavior of missing sound in speakers when i plug-in/plug-out, and even is worse, because appears 3 inputs (Mic, Internal Mic, Line In) recording only Mic with a lot of noise.

- #24: About the cuestion, I think it's a problem of upgrading VS. clean install. I have a similar problem, i have 13.10 (Installed first in 12.04 and upgrading until 13.10) with sound working but if i use the 13.10 liveCD the sound isn't working. Maybe some config files kept or else makes using AD198x instead AD1986A in my case, and when i made a clean install i discover the problem. In his case, the sound don't work properly in 12.04... but in my case with the liveCD 12.04.5 works, maybe there is something different there...

- #22: do you think i should send this "bug" to the kernel bugtracker with all the information? I want first be sure it isn't a mismatch configuration, common bug related to the distrubution, or a fixed bug in a newer kernel version. Isn't it?

[Raymond (superquad-vortex2)]

static models were dropped

https://git.kernel.org/cgit/linux/kernel/git/tiwai/sound.git/commit/sound/pci/hda/patch_analog.c?id=632408adfe70be6706cb89522b0d5b3dce188d84

you need to report bug to kernel bugzilla by providing output of alsa-info.sh when ad1986a is working and not working

[LukasThyWalls (thywalls)]

https://bugzilla.kernel.org/show_bug.cgi?id=88251

[madbiologist (me-again)]

This should be fixed in the upstream 3.19 kernel. Ubuntu 15.04 "Vivid Vervet" is based on the 15.04 kernel.

[LukasThyWalls (thywalls)]

I'm stuck in 3.16.0-24 with a working patch provided in the bugzilla kernel entry (https://bugzilla.kernel.org/show_bug.cgi?id=88251) and i am aware that patch is applied to the 3.19 kernel. I'm glad the next release of ubuntu should use that kernel's version, i will do a fresh install and later post if all is working well here.

[LukasThyWalls (thywalls)]

With KUbuntu 15.04 (Kernel 3.19), the sound problem seems to be resolved and now the sound is working directly. I have another problems, but i don't think they aren't related.