No sound device (Intel) found with kernel 5.6.bpo

[jwiz]

Seems the 5.6 kernel from buster-backports has issues with detecting the internal Intel sound chip (Realtek ALC233) on my machine.
Both 'aplay -l' and 'alsactl init' have not found the sound device.
Had to create /etc/modprobe.d/alsa-base.conf and put 'options snd-hda-intel dmic_detect=0' in it to enable the legacy mode.
The 5.4 kernel from buster-backports and the default 4.19 kernel had no issues with that.

[cpedretti]

Seems the 5.6 kernel from buster-backports has issues with detecting the internal Intel sound chip (Realtek ALC233) on my machine.
Both 'aplay -l' and 'alsactl init' have not found the sound device.
Had to create /etc/modprobe.d/alsa-base.conf and put 'options snd-hda-intel dmic_detect=0' in it to enable the legacy mode.
The 5.4 kernel from buster-backports and the default 4.19 kernel had no issues with that.

Possible that the firmware has not yet been compiled with the new kernel.


Try to install the latest firmwares and test if it works like this:

Code: Select all

sudo apt install -t buster-backports linux-firmware intel-microcode -y

You need to update the initramfs to compile the firmware with the new kernel, as root:

Code: Select all

update-initramfs -u

[jwiz]

Both packages were already the latest ones and I also had 'firmware-intel-sound' installed.
update-initramfs didn't make any change.

[MrEen]

This; options snd-hda-intel dmic_detect=0 has been deprecated somewhere in the 5.3 kernel series. You should be using this instead:

Code: Select all

options snd-intel-dspcfg dsp_driver=1

If you'd rather try to solve the issue, remove that edit, reboot then post the output of this:

Code: Select all

dmesg | grep -C1 -E 'ALSA|HDA|sof|HDMI|snd[_-]|sound|hda.codec|hda.intel'

I not saying we can definitely solve it, but I've helped a few get sound working properly on the new audio driver.

[jwiz]

Hmm..., seems like a firmware problem:

Code: Select all

jwiz@minibox42:~$ dmesg | grep -C1 -E 'ALSA|HDA|sof|HDMI|snd[_-]|sound|hda.codec|hda.intel'
[    0.354806] ACPI: Added _OSI(Linux-Dell-Video)
[    0.354806] ACPI: Added _OSI(Linux-Lenovo-NV-HDMI-Audio)
[    0.354806] ACPI: Added _OSI(Linux-HPI-Hybrid-Graphics)
--
[    1.499365] integrity: Loading X.509 certificate: UEFI:db
[    1.499380] integrity: Loaded X.509 cert 'Microsoft Corporation UEFI CA 2011: 13adbf4309bd82709c8cd54f316ed522988a1bd4'
[    1.499380] integrity: Loading X.509 certificate: UEFI:db
[    1.499395] integrity: Loaded X.509 cert 'Microsoft Windows Production PCA 2011: a92902398e16c49778cd90f99e4f9ae17c55af53'
[    1.502811] rtc_cmos rtc_cmos: setting system clock to 2020-07-24T05:31:13 UTC (1595568673)
--
[    4.032750] cryptd: max_cpu_qlen set to 1000
[    4.035254] snd_hda_intel 0000:00:1f.3: DSP detected with PCI class/subclass/prog-if info 0x040380
[    4.043350] snd_hda_intel 0000:00:1f.3: Digital mics found on Skylake+ platform, using SOF driver
[    4.044094] cfg80211: Loading compiled-in X.509 certificates for regulatory database
--
[    4.381829] iwlwifi 0000:02:00.0: base HW address: 04:33:c2:6a:b7:33
[    4.381908] snd_soc_skl 0000:00:1f.3: DSP detected with PCI class/subclass/prog-if info 0x040380
[    4.382673] [drm] Supports vblank timestamp caching Rev 2 (21.10.2013).
[    4.382675] [drm] Driver supports precise vblank timestamp query.
[    4.390070] snd_soc_skl 0000:00:1f.3: Digital mics found on Skylake+ platform, using SOF driver
[    4.390635] i915 0000:00:02.0: vgaarb: changed VGA decodes: olddecodes=io+mem,decodes=io+mem:owns=io+mem
--
[    4.538160] Bluetooth: hci0: Firmware revision 0.1 build 39 week 43 2018
[    4.539346] sof-audio-pci 0000:00:1f.3: DSP detected with PCI class/subclass/prog-if info 0x040380
[    4.539947] sof-audio-pci 0000:00:1f.3: Digital mics found on Skylake+ platform, using SOF driver
[    4.539958] sof-audio-pci 0000:00:1f.3: enabling device (0000 -> 0002)
[    4.540058] sof-audio-pci 0000:00:1f.3: DSP detected with PCI class/subclass/prog-if 0x040380
[    4.664631] [drm] failed to retrieve link info, disabling eDP
--
[    4.685296] input: Video Bus as /devices/LNXSYSTM:00/LNXSYBUS:00/PNP0A08:00/LNXVIDEO:00/input/input10
[    4.685513] sof-audio-pci 0000:00:1f.3: bound 0000:00:02.0 (ops i915_audio_component_bind_ops [i915])
[    4.719969] sof-audio-pci 0000:00:1f.3: use msi interrupt mode
[    4.735853] fbcon: i915drmfb (fb0) is primary device
--
[    4.787152] i915 0000:00:02.0: fb0: i915drmfb frame buffer device
[    4.879833] sof-audio-pci 0000:00:1f.3: hda codecs found, mask 5
[    4.879837] sof-audio-pci 0000:00:1f.3: using HDA machine driver skl_hda_dsp_generic now
[    4.884265] intel_rapl_common: Found RAPL domain package
--
[    4.884272] intel_rapl_common: Found RAPL domain dram
[    4.886055] sof-audio-pci 0000:00:1f.3: firmware: failed to load intel/sof/sof-cfl.ri (-2)
[    4.886081] firmware_class: See https://wiki.debian.org/Firmware for information about missing firmware
[    4.886084] sof-audio-pci 0000:00:1f.3: Direct firmware load for intel/sof/sof-cfl.ri failed with error -2
[    4.886089] sof-audio-pci 0000:00:1f.3: error: request firmware intel/sof/sof-cfl.ri failed err: -2
[    4.886091] sof-audio-pci 0000:00:1f.3: error: failed to load DSP firmware -2
[    4.930221] sof-audio-pci 0000:00:1f.3: error: sof_probe_work failed err: -2
[    5.577209] Bluetooth: BNEP (Ethernet Emulation) ver 1.3

[jwiz]

The sof firmware appears to be missing.
I don't see anything in /lib/firmware/intel/sof*
Installing alsa-firmware-loaders didnt help too.
So it's the first option for the moment.

[Moonstone Man]

So it's the first option for the moment.

Good luck getting it fixed if you can't be bothered following the instructions you're given.

[jwiz]

So it's the first option for the moment.

Good luck getting it fixed if you can't be bothered following the instructions you're given.

???
Your reaction seems a bit over the top, dont you think?

[jwiz]

Been reading up a bit about the SOF (Sound Open Firmware) project and it seems that SOF is supported since kernel 5.2 and part of the LF.
But the whole thing seems to be under a BSD license.
I wonder, if that could be a reason why we don't see the DSP firmware in Debian and thus disabling the dsp driver being the only option for now?

[antikythera]

I wonder, if that could be a reason why we don't see the DSP firmware in Debian and thus disabling the dsp driver being the only option for now?

Could be, but it is something Mint could offer for LMDE if someone with appropriate knowledge to compile, plus relevant Intel hardware (not to mention time and inclination) to test it packaged and maintained it.

[jwiz]

Well, its obviously not the problem that it can't be done.
I just pulled the latest linux-firmware_1.187.2 from Ubuntu, extracted the sof and sof-tlpg and pushed them into /lib/firmware/intel/.
Undid the workaround in alsa-base.conf and rebooted.
Voila, things work.

Code: Select all

[    5.051478] skl_hda_dsp_generic skl_hda_dsp_generic: ASoC: source widget Alt Analog Codec Capture overwritten
[    5.065032] input: sof-hda-dsp Mic as /devices/pci0000:00/0000:00:1f.3/skl_hda_dsp_generic/sound/card0/input11
[    5.065148] input: sof-hda-dsp Front Mic as /devices/pci0000:00/0000:00:1f.3/skl_hda_dsp_generic/sound/card0/input12
[    5.065247] input: sof-hda-dsp Front Headphone as /devices/pci0000:00/0000:00:1f.3/skl_hda_dsp_generic/sound/card0/input13
[    5.065353] input: sof-hda-dsp HDMI/DP,pcm=3 as /devices/pci0000:00/0000:00:1f.3/skl_hda_dsp_generic/sound/card0/input14
[    5.065452] input: sof-hda-dsp HDMI/DP,pcm=4 as /devices/pci0000:00/0000:00:1f.3/skl_hda_dsp_generic/sound/card0/input15
[    5.065545] input: sof-hda-dsp HDMI/DP,pcm=5 as /devices/pci0000:00/0000:00:1f.3/skl_hda_dsp_generic/sound/card0/input16
[    5.545398] Bluetooth: BNEP (Ethernet Emulation) ver 1.3
--
[   11.757686] IPv6: ADDRCONF(NETDEV_CHANGE): wlp2s0: link becomes ready
[   13.022769] sof-audio-pci 0000:00:1f.3: error: no reply expected, received 0x0
[   13.114940] sof-audio-pci 0000:00:1f.3: firmware boot complete
[   13.218751] show_signal: 4 callbacks suppressed

[antikythera]

Good stuff! Has the GPU freeze happened on the new kernel yet or is that looking resolved too?

[jwiz]

No, the GPU freeze hasn't happened with the 5.6 kernel so far, but I'm wary to say it's resolved, as it happened only sporadically.

[cpedretti]

Well, its obviously not the problem that it can't be done.
I just pulled the latest linux-firmware_1.187.2 from Ubuntu, extracted the sof and sof-tlpg and pushed them into /lib/firmware/intel/.
Undid the workaround in alsa-base.conf and rebooted.
Voila, things work.

Code: Select all

[    5.051478] skl_hda_dsp_generic skl_hda_dsp_generic: ASoC: source widget Alt Analog Codec Capture overwritten
[    5.065032] input: sof-hda-dsp Mic as /devices/pci0000:00/0000:00:1f.3/skl_hda_dsp_generic/sound/card0/input11
[    5.065148] input: sof-hda-dsp Front Mic as /devices/pci0000:00/0000:00:1f.3/skl_hda_dsp_generic/sound/card0/input12
[    5.065247] input: sof-hda-dsp Front Headphone as /devices/pci0000:00/0000:00:1f.3/skl_hda_dsp_generic/sound/card0/input13
[    5.065353] input: sof-hda-dsp HDMI/DP,pcm=3 as /devices/pci0000:00/0000:00:1f.3/skl_hda_dsp_generic/sound/card0/input14
[    5.065452] input: sof-hda-dsp HDMI/DP,pcm=4 as /devices/pci0000:00/0000:00:1f.3/skl_hda_dsp_generic/sound/card0/input15
[    5.065545] input: sof-hda-dsp HDMI/DP,pcm=5 as /devices/pci0000:00/0000:00:1f.3/skl_hda_dsp_generic/sound/card0/input16
[    5.545398] Bluetooth: BNEP (Ethernet Emulation) ver 1.3
--
[   11.757686] IPv6: ADDRCONF(NETDEV_CHANGE): wlp2s0: link becomes ready
[   13.022769] sof-audio-pci 0000:00:1f.3: error: no reply expected, received 0x0
[   13.114940] sof-audio-pci 0000:00:1f.3: firmware boot complete
[   13.218751] show_signal: 4 callbacks suppressed

sound_intel.png

With this link you have all the latest linux-firmware under linux:

https://git.kernel.org/pub/scm/linux/ke ... mware.git/

Code: Select all

apt install git -y

Code: Select all

cd /tmp

Code: Select all

git clone git://git.kernel.org/pub/scm/linux/kernel/git/firmware/linux-firmware.git

Code: Select all

cd linux-firmware/

Code: Select all

cp -va intel/ /lib/firmware/

You need to update the initramfs to compile the firmware with the new kernel:

Code: Select all

update-initramfs -u

Reboot to take effects

Code: Select all

reboot

Enjoy!

[jwiz]

Wouldn't help in this case, as the sof architecture isn't merged into the intel firmware package.
You'd have to grab it from here: https://github.com/thesofproject/linux- ... ster/intel

[MrEen]

Well, its obviously not the problem that it can't be done.
I just pulled the latest linux-firmware_1.187.2 from Ubuntu, extracted the sof and sof-tlpg and pushed them into /lib/firmware/intel/.
Undid the workaround in alsa-base.conf and rebooted.
Voila, things work.

Well done!

Usually that firmware error just needs a symlink to fix, but that's on Ubuntu-based systems it seems. I'm a bit surprised you had nothing in the firmware/intel/sof folder, but I don't use LMDE so wasn't aware what it was going to take.

[jwiz]

...
Usually that firmware error just needs a symlink to fix, but that's on Ubuntu-based systems it seems. I'm a bit surprised you had nothing in the firmware/intel/sof folder, but I don't use LMDE so wasn't aware what it was going to take.

From what I gathered do far on this topic, it seems that Intel is unwilling to provide the binary topology with their firmware package, though opinions exists that they should do that.
Intel oth wants the distro maintainers to build the binary topology for their distros from the sof sources on github.
Maintainers for distros like Arch, Fedora, Ubuntu have done that up to now.

[MrEen]

Well, I do not know the full scope, but one of the main sof developers is an Intel employee, and from reading some his posts in the sof bug reports and elsewhere, basically everything is there in the sof sources, and they've given instructions to the distros on how to implement it all. I know Ubuntu made a lot of mistakes at first, but they're getting pretty close to having it all working properly.

The issue I was expecting in your case was similar to this: viewtopic.php?p=1842203#p1842203

In that thread, both the firmware and the topology file needed symlinking. Same issue you originally presented with using that Coffee Lake chip. Simply symlinking to the Cannon Lake firmware fixes that up, then in some cases the topology file then throws a similar error with a similar cure.