[alsa-devel] Microphone detected, but no output for all ASUS G751xx with ALC668 chipset

[Connor McAdams]

You get that terminal output while running the program? Those
vfio_region_writes shouldn't be visible from the terminal... they
should be fed into QemuHDADump, and then it outputs it's own stuff. If
that is the output you're seeing, you might not have set up logging to
go to stderr.

On Wed, Sep 12, 2018 at 1:05 PM, Håvard <hovardslill at gmail.com> wrote:
> I think this is right!
>
> I first booted up, then tested speakers (That I could not hear) and then
> tested microphone (Which worked perfectly) in windows settings.
>
> -Håvard
>
> Den ons. 12. sep. 2018 kl. 18:49 skrev Connor McAdams
> <conmanx360 at gmail.com>:
>>
>> Also, sorry to keep sending more replies, make sure the only trace you
>> have on is vfio_region_write, and not the read one. The read spams up
>> the console.
>>
>> On Wed, Sep 12, 2018 at 12:43 PM, Connor McAdams <conmanx360 at gmail.com>
>> wrote:
>> > Reason being, it may not have identified the CORB buffer properly or
>> > something. That's very odd to have no data. Never seen that before.
>> >
>> > On Wed, Sep 12, 2018 at 12:42 PM, Connor McAdams <conmanx360 at gmail.com>
>> > wrote:
>> >> Could you try again and make a copy of your terminal output?
>> >>
>> >> On Wed, Sep 12, 2018 at 12:41 PM, Håvard <hovardslill at gmail.com> wrote:
>> >>> Hi again.
>> >>> Sorry for not noticing all files were empty, did it not run for long
>> >>> enough?
>> >>>
>> >>> -Håvard
>> >>>
>> >>> Den ons. 12. sep. 2018 kl. 18:39 skrev Håvard <hovardslill at gmail.com>:
>> >>>>
>> >>>> Sorry my bad again it was working perfectly, just missed a step.
>> >>>>
>> >>>> I'll add the dumps as an attatchment. I had both external headset and
>> >>>> microphone plugged in and let the VM run in the background for ~30min
>> >>>> while
>> >>>> doing other stuff. I did not play music or test my microphone in the
>> >>>> VM
>> >>>> either, If i did anything wrong, please tell me!
>> >>>>
>> >>>> It stopped at 0x104f4
>> >>>>
>> >>>> -Håvard
>> >>>>
>> >>>> Den ons. 12. sep. 2018 kl. 00:31 skrev Connor McAdams
>> >>>> <conmanx360 at gmail.com>:
>> >>>>>
>> >>>>> Hm.... that's odd. They should show up in the folder you ran the
>> >>>>> command from. Does your console show any "DumpMem entered..." or
>> >>>>> something like that? You may have a permissions error. I've had some
>> >>>>> people report that as an issue before.
>> >>>>>
>> >>>>> On Tue, Sep 11, 2018 at 5:08 PM, Håvard <hovardslill at gmail.com>
>> >>>>> wrote:
>> >>>>> > Hi!
>> >>>>> >
>> >>>>> > Took a while, but I think I got it working. Did not see any
>> >>>>> > "frame[xx]"
>> >>>>> > files though. What dumps do you want?
>> >>>>> >
>> >>>>> > -Håvard
>> >>>>> >
>> >>>>> > Den tir. 11. sep. 2018 kl. 21:27 skrev Håvard
>> >>>>> > <hovardslill at gmail.com>:
>> >>>>> >>
>> >>>>> >> Sorry about that, I forgot to enable a kernel config...
>> >>>>> >>
>> >>>>> >> -Håvard
>> >>>>> >>
>> >>>>> >> Den tir. 11. sep. 2018 kl. 21:20 skrev Connor McAdams
>> >>>>> >> <conmanx360 at gmail.com>:
>> >>>>> >>>
>> >>>>> >>> When it's bound as a stub, it shouldn't show up in alsamixer
>> >>>>> >>> controls.
>> >>>>> >>> You can check what module is loaded by doing lspci -v . It
>> >>>>> >>> should say
>> >>>>> >>> kernel driver in use: pci-stub. Also, did you run the vfio-bind
>> >>>>> >>> script
>> >>>>> >>> before trying it?
>> >>>>> >>>
>> >>>>> >>> On Tue, Sep 11, 2018 at 3:18 PM, Håvard <hovardslill at gmail.com>
>> >>>>> >>> wrote:
>> >>>>> >>> > Thank you for the offer!
>> >>>>> >>> >
>> >>>>> >>> > I run gentoo, but your github guide is very useful. I can't
>> >>>>> >>> > seem to
>> >>>>> >>> > bind my
>> >>>>> >>> > audio driver to a pci-stud. Is the Sound card still supposed
>> >>>>> >>> > to
>> >>>>> >>> > function as
>> >>>>> >>> > normal when I (think I) have bound it to a stud.
>> >>>>> >>> >
>> >>>>> >>> > -Håvard
>> >>>>> >>> >
>> >>>>> >>> > Den tir. 11. sep. 2018 kl. 20:26 skrev Connor McAdams
>> >>>>> >>> > <conmanx360 at gmail.com>:
>> >>>>> >>> >>
>> >>>>> >>> >> One thing you could try, is using the program I used to
>> >>>>> >>> >> reverse
>> >>>>> >>> >> engineer the Sound Blaster Z series of cards, QemuHDADump. If
>> >>>>> >>> >> your
>> >>>>> >>> >> processor supports pci-passthrough with a virtual machine,
>> >>>>> >>> >> you
>> >>>>> >>> >> could
>> >>>>> >>> >> run a Windows virtual machine with the sound card in it and
>> >>>>> >>> >> capture
>> >>>>> >>> >> the commands. I'd be willing to look through the dumps to see
>> >>>>> >>> >> if
>> >>>>> >>> >> there
>> >>>>> >>> >> are any special verbs or anything.
>> >>>>> >>> >>
>> >>>>> >>> >> You can find the program here:
>> >>>>> >>> >> https://github.com/Conmanx360/QemuHDADump
>> >>>>> >>> >>
>> >>>>> >>> >> Let me know.
>> >>>>> >>> >>
>> >>>>> >>> >> On Tue, Sep 11, 2018 at 2:15 PM, Håvard
>> >>>>> >>> >> <hovardslill at gmail.com>
>> >>>>> >>> >> wrote:
>> >>>>> >>> >> > Sorry, my gmail didn't update so I wrote my response before
>> >>>>> >>> >> > I
>> >>>>> >>> >> > read
>> >>>>> >>> >> > your
>> >>>>> >>> >> > last one.
>> >>>>> >>> >> >
>> >>>>> >>> >> > It is a separate mic port as the G751JT doesn't have any
>> >>>>> >>> >> > headset
>> >>>>> >>> >> > multijacks.
>> >>>>> >>> >> >
>> >>>>> >>> >> > I'll try and dig around in the kernel! Thanks for the tip!
>> >>>>> >>> >> >
>> >>>>> >>> >> > -Håvard
>> >>>>> >>> >> >
>> >>>>> >>> >> > Den tir. 11. sep. 2018 kl. 20:09 skrev Håvard
>> >>>>> >>> >> > <hovardslill at gmail.com>:
>> >>>>> >>> >> >
>> >>>>> >>> >> >> Thank you for taking your time and trying to help! :)
>> >>>>> >>> >> >>
>> >>>>> >>> >> >> Do you know where I can ask around for more help on the
>> >>>>> >>> >> >> issue?
>> >>>>> >>> >> >> I
>> >>>>> >>> >> >> don't
>> >>>>> >>> >> >> want to give up yet.
>> >>>>> >>> >> >>
>> >>>>> >>> >> >> -Håvard
>> >>>>> >>> >> >>
>> >>>>> >>> >> >> Den tir. 11. sep. 2018 kl. 20:02 skrev Takashi Iwai
>> >>>>> >>> >> >> <tiwai at suse.de>:
>> >>>>> >>> >> >>
>> >>>>> >>> >> >>> On Tue, 11 Sep 2018 19:14:37 +0200,
>> >>>>> >>> >> >>> Håvard wrote:
>> >>>>> >>> >> >>> >
>> >>>>> >>> >> >>> > Yes, microphone gets detected instantly and it
>> >>>>> >>> >> >>> > automatically
>> >>>>> >>> >> >>> > changes
>> >>>>> >>> >> >>> > to
>> >>>>> >>> >> >>> it
>> >>>>> >>> >> >>> > in pavucontrol.
>> >>>>> >>> >> >>>
>> >>>>> >>> >> >>> Then it likely requires some additional initialization
>> >>>>> >>> >> >>> outside
>> >>>>> >>> >> >>> HD-audio.  It's hard to know, as it's pretty much
>> >>>>> >>> >> >>> vendor-specific.
>> >>>>> >>> >> >>> You can dig down the Windows, but I have no idea about
>> >>>>> >>> >> >>> Windows
>> >>>>> >>> >> >>> implementation, so can't give any hints, unfortunately.
>> >>>>> >>> >> >>>
>> >>>>> >>> >> >>>
>> >>>>> >>> >> >>> Takashi
>> >>>>> >>> >> >>>
>> >>>>> >>> >> >>> >
>> >>>>> >>> >> >>> > Den tir. 11. sep. 2018 kl. 18:52 skrev Takashi Iwai
>> >>>>> >>> >> >>> > <tiwai at suse.de>:
>> >>>>> >>> >> >>> >
>> >>>>> >>> >> >>> > > On Tue, 11 Sep 2018 18:40:23 +0200,
>> >>>>> >>> >> >>> > > Håvard wrote:
>> >>>>> >>> >> >>> > > >
>> >>>>> >>> >> >>> > > > Thank you for replying!
>> >>>>> >>> >> >>> > > >
>> >>>>> >>> >> >>> > > > Enabling loopback in alsamixer:
>> >>>>> >>> >> >>> > > > http://i.imgur.com/lNo6e7T.png
>> >>>>> >>> >> >>> > > >
>> >>>>> >>> >> >>> > > > And unmuting more and more things in "Mic Playback
>> >>>>> >>> >> >>> > > > Volume"
>> >>>>> >>> >> >>> > > > in
>> >>>>> >>> >> >>> > > hdaanalyzer:
>> >>>>> >>> >> >>> > > > http://i.imgur.com/H0HiOhy.png
>> >>>>> >>> >> >>> > > > made white noise come from the headset. However it
>> >>>>> >>> >> >>> > > > did
>> >>>>> >>> >> >>> > > > not
>> >>>>> >>> >> >>> > > > change or
>> >>>>> >>> >> >>> > > react
>> >>>>> >>> >> >>> > > > at all when I talked or even muted the microphone
>> >>>>> >>> >> >>> > > > physically.
>> >>>>> >>> >> >>> > > >
>> >>>>> >>> >> >>> > > > I couldn't find "Mic Playback Switch" anywhere in
>> >>>>> >>> >> >>> > > > either
>> >>>>> >>> >> >>> > > > alsamixer
>> >>>>> >>> >> >>> or
>> >>>>> >>> >> >>> > > > hdaanalyzer.
>> >>>>> >>> >> >>> > >
>> >>>>> >>> >> >>> > > It's a mixer mute switch.
>> >>>>> >>> >> >>> > >
>> >>>>> >>> >> >>> > > > The microphone works perfectly fine under Windows,
>> >>>>> >>> >> >>> > > > so I
>> >>>>> >>> >> >>> > > > don't
>> >>>>> >>> >> >>> > > > think
>> >>>>> >>> >> >>> it is
>> >>>>> >>> >> >>> > > > the mic pin.
>> >>>>> >>> >> >>> > >
>> >>>>> >>> >> >>> > > But the fact above indicates the possibility of the
>> >>>>> >>> >> >>> > > wrong
>> >>>>> >>> >> >>> > > pin,
>> >>>>> >>> >> >>> > > too.
>> >>>>> >>> >> >>> > >
>> >>>>> >>> >> >>> > > Does the jack detection of the ext mic pin work?
>> >>>>> >>> >> >>> > >
>> >>>>> >>> >> >>> > >
>> >>>>> >>> >> >>> > > Takashi
>> >>>>> >>> >> >>> > >
>> >>>>> >>> >> >>> > > >
>> >>>>> >>> >> >>> > > > -Håvard
>> >>>>> >>> >> >>> > > >
>> >>>>> >>> >> >>> > > > Den man. 10. sep. 2018 kl. 22:39 skrev Takashi Iwai
>> >>>>> >>> >> >>> > > > <tiwai at suse.de
>> >>>>> >>> >> >>> >:
>> >>>>> >>> >> >>> > > >
>> >>>>> >>> >> >>> > > > > On Thu, 06 Sep 2018 20:44:30 +0200,
>> >>>>> >>> >> >>> > > > > Håvard wrote:
>> >>>>> >>> >> >>> > > > > >
>> >>>>> >>> >> >>> > > > > > Additional relevant info:
>> >>>>> >>> >> >>> > > > > >
>> >>>>> >>> >> >>> > > > > > A similar issue was also discussed three years
>> >>>>> >>> >> >>> > > > > > ago
>> >>>>> >>> >> >>> > > > > > on
>> >>>>> >>> >> >>> > > > > > Sun
>> >>>>> >>> >> >>> > > > > > Jun
>> >>>>> >>> >> >>> > > 17:15:54
>> >>>>> >>> >> >>> > > > > CEST
>> >>>>> >>> >> >>> > > > > > 2015 and was about his surround sound setup,
>> >>>>> >>> >> >>> > > > > > but did
>> >>>>> >>> >> >>> > > > > > not
>> >>>>> >>> >> >>> > > > > > touch
>> >>>>> >>> >> >>> on the
>> >>>>> >>> >> >>> > > > > > external microphone problem:
>> >>>>> >>> >> >>> > > > > >
>> >>>>> >>> >> >>> > > > >
>> >>>>> >>> >> >>> > >
>> >>>>> >>> >> >>>
>> >>>>> >>> >> >>>
>> >>>>> >>> >> >>>
>> >>>>> >>> >> >>>
>> >>>>> >>> >> >>> http://mailman.alsa-project.org/pipermail/alsa-devel/2015-June/093317.html
>> >>>>> >>> >> >>> > > > > >
>> >>>>> >>> >> >>> > > > > > alsa-info.sh:
>> >>>>> >>> >> >>> > > > > >
>> >>>>> >>> >> >>> > > > >
>> >>>>> >>> >> >>> > >
>> >>>>> >>> >> >>>
>> >>>>> >>> >> >>>
>> >>>>> >>> >> >>>
>> >>>>> >>> >> >>>
>> >>>>> >>> >> >>> http://www.alsa-project.org/db/?f=1d8616ba5977308e03db6c3a86e36e9e9b38d6f0
>> >>>>> >>> >> >>> > > > > >
>> >>>>> >>> >> >>> > > > > > Graph of ALC668 chipset from hdaanalyzer:
>> >>>>> >>> >> >>> > > > > > http://i.imgur.com/c08DNJW.png
>> >>>>> >>> >> >>> > > > > >
>> >>>>> >>> >> >>> > > > > > setting alsa-mode[1-8] does nothing to help the
>> >>>>> >>> >> >>> > > > > > issue.
>> >>>>> >>> >> >>> > > > > >
>> >>>>> >>> >> >>> > > > > > There are several of similar bug reports around
>> >>>>> >>> >> >>> > > > > > the
>> >>>>> >>> >> >>> > > > > > web
>> >>>>> >>> >> >>> experiencing
>> >>>>> >>> >> >>> > > > > > similar issues, and on different distros.
>> >>>>> >>> >> >>> > > > > >
>> >>>>> >>> >> >>> > > > > > Microphone works perfectly in windows
>> >>>>> >>> >> >>> > > > > >
>> >>>>> >>> >> >>> > > > > > I have a ASUS ROG G751JT, but this problem
>> >>>>> >>> >> >>> > > > > > seems to
>> >>>>> >>> >> >>> > > > > > happen
>> >>>>> >>> >> >>> > > > > > with
>> >>>>> >>> >> >>> all
>> >>>>> >>> >> >>> > > > > laptops
>> >>>>> >>> >> >>> > > > > > under the G751 name.
>> >>>>> >>> >> >>> > > > >
>> >>>>> >>> >> >>> > > > > When you enable the loopback volume and switch,
>> >>>>> >>> >> >>> > > > > and
>> >>>>> >>> >> >>> > > > > unmute/adjust
>> >>>>> >>> >> >>> "Mic
>> >>>>> >>> >> >>> > > > > Playback Volume", and "Mic Playback Switch", do
>> >>>>> >>> >> >>> > > > > you
>> >>>>> >>> >> >>> > > > > hear
>> >>>>> >>> >> >>> > > > > the
>> >>>>> >>> >> >>> > > > > input
>> >>>>> >>> >> >>> > > > > from the ext mic?  It's a route directly from NID
>> >>>>> >>> >> >>> > > > > 0x18
>> >>>>> >>> >> >>> > > > > to
>> >>>>> >>> >> >>> > > > > the
>> >>>>> >>> >> >>> mixer
>> >>>>> >>> >> >>> > > > > NID 0x0b, then output mixer NID 0x0c, then
>> >>>>> >>> >> >>> > > > > outputs.
>> >>>>> >>> >> >>> > > > > So
>> >>>>> >>> >> >>> > > > > this
>> >>>>> >>> >> >>> > > > > can
>> >>>>> >>> >> >>> be
>> >>>>> >>> >> >>> > > > > used to verify the hardware routing.
>> >>>>> >>> >> >>> > > > >
>> >>>>> >>> >> >>> > > > > If you don't hear via this route, it means that
>> >>>>> >>> >> >>> > > > > the
>> >>>>> >>> >> >>> > > > > input
>> >>>>> >>> >> >>> > > > > from
>> >>>>> >>> >> >>> the ext
>> >>>>> >>> >> >>> > > > > mic pin itself is broken, and it implies that
>> >>>>> >>> >> >>> > > > > something
>> >>>>> >>> >> >>> > > > > outside
>> >>>>> >>> >> >>> > > > > HD-audio codec.
>> >>>>> >>> >> >>> > > > >
>> >>>>> >>> >> >>> > > > >
>> >>>>> >>> >> >>> > > > > Takashi
>> >>>>> >>> >> >>> > > > >
>> >>>>> >>> >> >>> > > > [2  <text/html; UTF-8 (quoted-printable)>]
>> >>>>> >>> >> >>> > > >
>> >>>>> >>> >> >>> > >
>> >>>>> >>> >> >>> > [2  <text/html; UTF-8 (quoted-printable)>]
>> >>>>> >>> >> >>> >
>> >>>>> >>> >> >>>
>> >>>>> >>> >> >>
>> >>>>> >>> >> > _______________________________________________
>> >>>>> >>> >> > Alsa-devel mailing list
>> >>>>> >>> >> > Alsa-devel at alsa-project.org
>> >>>>> >>> >> > http://mailman.alsa-project.org/mailman/listinfo/alsa-devel