Bluetooth no longer recognized after update to 5.14.0-1013

[Erich Eickmeyer]

On many CLEVO chassis computers with the recent update to the OEM 5.14.0-1013 kernel, bluetooth is no longer recognized by KDE Plasma. We are not sure why this has happened, but this is likely an API change.

ProblemType: Bug
DistroRelease: Ubuntu 20.04
Package: linux-image-5.14.0-1013-oem 5.14.0-1013.13
ProcVersionSignature: Ubuntu 5.14.0-1013.13-oem 5.14.20
Uname: Linux 5.14.0-1013-oem x86_64
ApportVersion: 2.20.11-0ubuntu27.21
Architecture: amd64
CasperMD5CheckResult: skip
CurrentDesktop: KDE
Date: Tue Jan 4 13:47:39 2022
InstallationDate: Installed on 2021-08-05 (151 days ago)
InstallationMedia: Kubuntu 20.04.2 "Focal Fossa" (20210805)
SourcePackage: linux-signed-oem-5.14
UpgradeStatus: No upgrade log present (probably fresh install)

[Michael Mikowski (kfocus)]

Here is some additional information that may be helpful.

1. Prior to upgrade with kernel 5.14.0-1011-oem, KDE Control Center > Bluetooth configuration shows full control over Bluetooth
2. After update to 5.14.0-1013-oem KDE Control Center > Bluetooth configuration shows message similar to 'No Bluetooth Device Found'
3. Reverting to kernel 5.14-0-1011-oem restores the expected capability
4. The following show differences between kernels:
4.1 `systemctl status bluetooth` shows more endpoints with 5.14.0-1011-oem
4.2 `lsusb` shows an Intel device with 5.14.0-1011-oem
4.3 `dmesg | egrep -i 'blue|firm'` shows more output with 5.14.0-1011-oem
5. We have reproduced on three systems, all with Intel AX201 Wifi Cards
5.1 First system:
  #> 00:14.3 Network controller: Intel Corporation Comet Lake PCH CNVi WiFi
  #> Subsystem: Intel Corporation Wi-Fi 6 AX201 160MHz ...
5.2 Second System:
  #> 00:14.3 Network controller: Intel Corporation Wi-Fi 6 AX201 (rev 20)
  #> Subsystem: Intel Corporation Device 0074

#> Updated Packages:
apt-rollback --last 2
#> 1: 2022-01-04 13:09:02 - Upgrade of libsystemd0:amd64 libsystemd0:i386 udev:amd64 libudev1:amd64 libudev1:i386 systemd-timesyncd:amd64 libudev-dev:amd64 systemd-sysv:amd64 libpam-systemd:amd64 systemd:amd64 libnss-systemd:amd64
#> 2: 2022-01-04 08:51:30 - Upgrade of linux-image-oem-20.04d:amd64 linux-libc-dev:amd64 libnvidia-compute-470:amd64 libnvidia-compute-470:i386 libnvidia-encode-470:amd64 libnvidia-encode-470:i386 nvidia-kernel-common-470:amd64 xserver-xorg-video-nvidia-470:amd64 libnvidia-gl-470:amd64 libnvidia-gl-470:i386 azure-cli:amd64 libnvidia-fbc1-470:amd64 libnvidia-fbc1-470:i386 libnvidia-decode-470:amd64 libnvidia-decode-470:i386 linux-headers-oem-20.04d:amd64 libnvidia-cfg1-470:amd64 nvidia-utils-470:amd64 nvidia-dkms-470:amd64 linux-tools-oem-20.04d:amd64 nvidia-compute-utils-470:amd64 libnvidia-ifr1-470:amd64 libnvidia-ifr1-470:i386 nvidia-driver-470:amd64 libnvidia-common-470:amd64 linux-firmware:amd64 linux-tools-common:amd64 linux-oem-20.04d:amd64 libnvidia-extra-470:amd64 nvidia-kernel-source-470:amd64

systemctl status bluetooth # with 5.14.0-1013-oem kernel
#> ● bluetooth.service - Bluetooth service
#> Loaded: loaded (/lib/systemd/system/bluetooth.service; enabled; vendor preset: enabled)
#> Active: active (running) since Tue 2022-01-04 13:26:24 PST; 10min ago
#> Docs: man:bluetoothd(8)
#> Main PID: 17644 (bluetoothd)
#> Status: "Running"
#> Tasks: 1 (limit: 38132)
#> Memory: 504.0K
#> CGroup: /system.slice/bluetooth.service
#> └─17644 /usr/lib/bluetooth/bluetoothd
#>
#> Jan 04 13:26:24 kfocus systemd[1]: Starting Bluetooth service...
#> Jan 04 13:26:24 kfocus bluetoothd[17644]: Bluetooth daemon 5.53
#> Jan 04 13:26:24 kfocus systemd[1]: Started Bluetooth service.
#> Jan 04 13:26:24 kfocus bluetoothd[17644]: Starting SDP server
#> Jan 04 13:26:24 kfocus bluetoothd[17644]: Bluetooth management interface 1.21 initialized

systemctl status bluetooth # with 5.14.0-1011-oem kernel
#> ● bluetooth.service - Bluetooth service
#> Loaded: loaded (/lib/systemd/system/bluetooth.service; enabled; vendor preset: enabled)
#> Act...

[jeremyszu (os369510)]

Hi,

Could you please provide:

# lspci -nnv
# journalctl -b # boot with 5.14-1013

[You-Sheng Yang (vicamo)]

Hi, I cannot reproduce this with an AX201 on TGL platform. Is it possible to run `sudo apport-collect -p linux-firmware 1956393` to collect some more details? I'd need dmesg from a cold boot & output of `lspci -nnk`, `lscpu`, `lsusb -v -v -v` and `hciconfig` specifically.

[Michael Mikowski (kfocus)]

ProblemType: Bug
ApportVersion: 2.20.11-0ubuntu27.21
Architecture: amd64
CasperMD5CheckResult: skip
Dependencies:

DistroRelease: Ubuntu 20.04
InstallationDate: Installed on 2020-06-10 (573 days ago)
InstallationMedia: Kubuntu 20.04 LTS "Focal Fossa" - Release amd64 (20200423)
MachineType: Notebook PCX0DX
Package: linux-firmware 1.187.24
PackageArchitecture: all
ProcEnviron:
 LANGUAGE=
 TERM=xterm-256color
 PATH=(custom, no user)
 LANG=en_US.UTF-8
 SHELL=/bin/bash
ProcFB: 0 i915
ProcKernelCmdLine: BOOT_IMAGE=/vmlinuz-5.14.0-1013-oem root=/dev/mapper/vgkubuntu-root ro quiet splash acpi=force vt.handoff=7
ProcVersionSignature: Ubuntu 5.14.0-1013.13-oem 5.14.20
PulseList: Error: command ['pacmd', 'list'] failed with exit code 1: No PulseAudio daemon running, or not running as session daemon.
RelatedPackageVersions:
 linux-restricted-modules-5.14.0-1013-oem N/A
 linux-backports-modules-5.14.0-1013-oem N/A
 linux-firmware 1.187.24
RfKill:
 0: phy0: Wireless LAN
  Soft blocked: no
  Hard blocked: no
Tags: focal
Uname: Linux 5.14.0-1013-oem x86_64
UpgradeStatus: No upgrade log present (probably fresh install)
UserGroups: N/A
_MarkForUpload: True
dmi.bios.date: 06/17/2020
dmi.bios.release: 7.7
dmi.bios.vendor: INSYDE Corp.
dmi.bios.version: 1.07.07MIN29
dmi.board.asset.tag: Tag 12345
dmi.board.name: PCX0DX
dmi.board.vendor: Notebook
dmi.board.version: Not Applicable
dmi.chassis.asset.tag: No Asset Tag
dmi.chassis.type: 10
dmi.chassis.vendor: Notebook
dmi.chassis.version: N/A
dmi.ec.firmware.release: 7.2
dmi.modalias: dmi:bvnINSYDECorp.:bvr1.07.07MIN29:bd06/17/2020:br7.7:efr7.2:svnNotebook:pnPCX0DX:pvrNotApplicable:rvnNotebook:rnPCX0DX:rvrNotApplicable:cvnNotebook:ct10:cvrN/A:skuNotApplicable:
dmi.product.family: Not Applicable
dmi.product.name: PCX0DX
dmi.product.sku: Not Applicable
dmi.product.version: Not Applicable
dmi.sys.vendor: Notebook

[Michael Mikowski (kfocus)]

apport information

[Michael Mikowski (kfocus)]

apport information

[Michael Mikowski (kfocus)]

apport information

[Michael Mikowski (kfocus)]

apport information

[Michael Mikowski (kfocus)]

apport information

[Michael Mikowski (kfocus)]

apport information

[Michael Mikowski (kfocus)]

apport information

[Michael Mikowski (kfocus)]

apport information

[Michael Mikowski (kfocus)]

apport information

[Michael Mikowski (kfocus)]

apport information

[Michael Mikowski (kfocus)]

apport information

[Michael Mikowski (kfocus)]

apport information

[Michael Mikowski (kfocus)]

apport information

[Michael Mikowski (kfocus)]

apport information

[Michael Mikowski (kfocus)]

apport information

[Michael Mikowski (kfocus)]

apport information

[Michael Mikowski (kfocus)]

apport information

[Launchpad Janitor (janitor)]

Status changed to 'Confirmed' because the bug affects multiple users.

[Michael Mikowski (kfocus)]

Hi @vicamo and @os369510:

At present, we have advised the owners of the devices that require Bluetooth to use the older 5.14.0-1011-oem kernel which we have confirmed fixes the problem. Besides the testing on two different Clevo development devices, we have also received 4 reports of failed bluetooth from users running the 5.14.0-1013-oem kernel. The affected devices use an i7-10875H and an i7-1165g7. An i7-9750H model does NOT have this issue. I will test an i7-11800H processor when back in the office. All devices are running 20.04 LTS.

I have run 'sudo apport-collect -p linux-firmware 1956393' on a device running the 5.14.0-1013-oem kernel. Please let me know if you would also like apport-collect run on the 5.14.0-1011-oem kernel for comparison. If there is a better way than adding all the messy attachments above, please let me know.

I am also collecting the other information you requested for BOTH the non-working kernel (5.14.0-1013) and the working kernel (5.14.0-1011) and will attach soon. I propose to attach a tarball, but please let me know if you prefer a different mechanism. Here are the proposed files:

  1956393.1_5.14.0-1011-oem_hciconfig.txt # sudo hciconfig
  1956393.1_5.14.0-1011-oem_journalctl-b.txt # sudo journalctl -b
  1956393.1_5.14.0-1011-oem_lscpu.txt # sudo lscpu
  1956393.1_5.14.0-1011-oem_lspci-nnk.txt # sudo lscpi -nnk
  1956393.1_5.14.0-1011-oem_lspci-nnv.txt # sudo lspci -nnv
  1956393.1_5.14.0-1011-oem_lsusb-v-v-v.txt # sudo lsusb -v -v -v

  1956393.1_5.14.0-1013-oem_hciconfig.txt # sudo hciconfig
  1956393.1_5.14.0-1013-oem_journalctl-b.txt # sudo journalctl -b
  1956393.1_5.14.0-1013-oem_lscpu.txt # sudo lscpu
  1956393.1_5.14.0-1013-oem_lspci-nnk.txt # sudo lscpi -nnk
  1956393.1_5.14.0-1013-oem_lspci-nnv.txt # sudo lspci -nnv
  1956393.1_5.14.0-1013-oem_lsusb-v-v-v.txt # sudo lsusb -v -v -v

[Michael Mikowski (kfocus)]

Hi @vicamo and @os369510:

I have attached the files in 1956393.1_checks.tgz. We have inspected the
differences and found the following which we hope you find useful:

meld 1956393.1_5.14.0-1011-oem_hciconfig.txt \
  1956393.1_5.14.0-1013-oem_hciconfig.txt;
This shows no output for the 1013 kernel, but sustantial output for 1011.

meld 1956393.1_5.14.0-1011-oem_journalctl-b.txt \
  1956393.1_5.14.0-1013-oem_journalctl-b.txt;
This is substantially different (as expected).

meld 1956393.1_5.14.0-1011-oem_lscpu.txt \
  1956393.1_5.14.0-1013-oem_lscpu.txt;
Only current freq differs.

meld 1956393.1_5.14.0-1011-oem_lspci-nnk.txt \
  1956393.1_5.14.0-1013-oem_lspci-nnk.txt;
No changes.

meld 1956393.1_5.14.0-1011-oem_lspci-nnv.txt \
  1956393.1_5.14.0-1013-oem_lspci-nnv.txt;
Some flags differ with IRQs changed.

meld 1956393.1_5.14.0-1011-oem_lsusb-v-v-v.txt \
  1956393.1_5.14.0-1013-oem_lsusb-v-v-v.txt;
Bus 001 Device 005: ID 8087:0026 Intel Corp IS NOT FOUND for 1013 kernel.

Please let us know if there is anything else we can provide.

Sincerely, Mike

[Michael Mikowski (kfocus)]

We just tested kernel 5.14.0-1015 from proposed on the i7-10875H system and Bluetooth is still not available. However, there is an additional hub shown which you can see in the attched file, 1956393.1_5.14.0-1015-oem_lsusb-v-v-v.txt.

[You-Sheng Yang (vicamo)]

There is a "usb 1-14: USB disconnect, device number 5" in the 1013 dmesg, so your bluetooth device is done. Let's see...

[Timo Aaltonen (tjaalton)]

could you test latest stable mainline build for 5.14.21:

https://kernel.ubuntu.com/~kernel-ppa/mainline/v5.14.21/

this would show if it's a regression there or in the distro backports

[Timo Aaltonen (tjaalton)]

another one to test is v5.15.8

[Timo Aaltonen (tjaalton)]

and here's oem-5.14 1015 minus one patch to see if it caused the bug:

http://aaltoset.kapsi.fi/oem-debug/

[Michael Mikowski (kfocus)]

@tjaalton, @vicamo, thank you for such rapid assistance. I will test these immediately and report back in a few minutes.

[Michael Mikowski (kfocus)]

@tjaalton and @vicamo, there's an issue with the last set of packages. Here's what happens when I try to install:

sudo apt install ./linux-headers-5.14.0-1015-oem_5.14.0-1015.15+revert_amd64.deb ./linux-image-unsigned-5.14.0-1015-oem_5.14.0-1015.15+revert_amd64.deb ./linux-modules-5.14.0-1015-oem_5.14.0-1015.15+revert_amd64.deb
Reading package lists... Done
Building dependency tree
Reading state information... Done
Note, selecting 'linux-headers-5.14.0-1015-oem' instead of './linux-headers-5.14.0-1015-oem_5.14.0-1015.15+revert_amd64.deb'
Note, selecting 'linux-image-unsigned-5.14.0-1015-oem' instead of './linux-image-unsigned-5.14.0-1015-oem_5.14.0-1015.15+revert_amd64.deb'
Note, selecting 'linux-modules-5.14.0-1015-oem' instead of './linux-modules-5.14.0-1015-oem_5.14.0-1015.15+revert_amd64.deb'
Some packages could not be installed. This may mean that you have
requested an impossible situation or if you are using the unstable
distribution that some required packages have not yet been created
or been moved out of Incoming.
The following information may help to resolve the situation:

The following packages have unmet dependencies:
 linux-headers-5.14.0-1015-oem : Depends: linux-oem-5.14-headers-5.14.0-1015 but it is not installable
E: Unable to correct problems, you have held broken packages.

EDIT: installed without headers and tested. See below.

[Michael Mikowski (kfocus)]

@tjaalton and @vicamo, attempting to install the first set, 5.14.21, results in this error:

The following packages have unmet dependencies:
 linux-headers-5.14.21-051421-generic : Depends: libc6 (>= 2.34) but 2.31-0ubuntu9.3 is to be installed

EDIT: installed without headers and tested. See below.

[Michael Mikowski (kfocus)]

@tjaalton and @vicamo, I installed just the image and modules for 5.14.0-1015 comment 31 without the headers (so no DKMS), and got the same error: "No Bluetooth adapters have been found." I just installed 5.14.21 and will report back here too. I don't believe DKMS is required for Bluetooth.

[Michael Mikowski (kfocus)]

@tjaalton and @vicamo, I have tested with 5.14.21 without the headers and the bluetooth device is recognized and appears to work. I attached a bluetooth headset and listen to audio without issue.

[Michael Mikowski (kfocus)]

@tjaalton and @vicamo, I have tested with 5.15.8 without the headers and the bluetooth device is recognized and appears to work. I again attached a bluetooth headset and listened to audio without issue.

So in summary:

FAIL 5.14.0-1013 "No Bluetooth adapters have been found"
FAIL 5.14.0-1015 "No Bluetooth adapters have been found"
FAIL 5.14.0-1015 without patch (from comment 31) "No Bluetooth adapters have been found"
FAIL 5.15.0-1018 "No Bluetooth adapters have been found"
FAIL 5.14.20-051420-generic "No Bluetooth adapters have been found"
pass 5.14.21-051421-generic Bluetooth adapter available; test audio device works
pass 5.15.8-051508-generic Bluetooth adapter available; test audio device works

I hope this is helpful, and if there is anything else I can do, please let me know.

[Timo Aaltonen (tjaalton)]

looks like 1015 didn't actually contain 5.14.21 but .20, though I doubt the fix was there but you may also try .20 mainline to see if it fails

[Michael Mikowski (kfocus)]

Hi @tjaalton and @vicamo:

I tested the 5.14.20 from the mainline as suggested in comment #8, and it FAILED, where as 5.14.21 worked.

FAIL 5.14.20-051420-generic "No Bluetooth adapters have been found"

I added this to the table in comment #37.

[Michael Mikowski (kfocus)]

I was surprised to find a 5.14.0-1018 upgrade just a few minutes ago. I installed and tested; it FAILED, and I have updated the table in comment #37 as well.

[Timo Aaltonen (tjaalton)]

1018 was a security update, but anyway it looks like the fix for the regression is in 5.14.21 so you'll have that in the next bigger update which will be released to -updates on jan 31st

[Michael Mikowski (kfocus)]

Hi @tjaalton and @vicamo:

So 5.14.21 will be coming on Jan 31st. Have you been able to identify what went wrong? If so, is there any chance the fix will appear before the 31st?

Thanks again for rapid assistance!

[Erich Eickmeyer (eeickmeyer)]

Tested with 1020 kernel in focal-updates, still no-good.
Tested with 1021 kernel in focal-proposed, seems to be OK.

[Erich Eickmeyer (eeickmeyer)]

Tested with 1022 kernel in focal-proposed, looks fine except that since 1021 the kernel seems to be introducing a regression causing virtualbox dkms drivers to fail in the version in the repository.

Recommendation: either kernel needs to be patched for the regression or virtualbox needs to be updated.

[Timo Aaltonen (tjaalton)]

closing as -1021 seems to work

[Michael Mikowski (kfocus)]

@tjaalton: The kernel 5.14.0-1022 has been tested on 6 laptop types and passes on all of them. Tested subsystems included USB, Bluetooth, Wi-Fi, Ethernet and many others. We can get you a full report if you like. Thank you!