Bug 31965 - bluetooth ‎Realtek mageia9 does not work
Summary: bluetooth ‎Realtek mageia9 does not work
Status: RESOLVED FIXED
Alias: None
Product: Mageia
Classification: Unclassified
Component: RPM Packages (show other bugs)
Version: Cauldron
Hardware: x86_64 Linux
Priority: Normal major
Target Milestone: ---
Assignee: Kernel and Drivers maintainers
QA Contact:
URL:
Whiteboard: MGA9TOO
Keywords:
Depends on: 32623
Blocks:
  Show dependency treegraph
 
Reported: 2023-05-26 21:32 CEST by jean-jacques skap
Modified: 2023-12-26 00:29 CET (History)
5 users (show)

See Also:
Source RPM:
CVE:
Status comment:


Attachments
The instal1_30mai2023.txt file corresponds to the initial installation. (131.11 KB, text/plain)
2023-05-30 19:02 CEST, jean-jacques skap
Details
rpm de l'installtion initiale de mga9RC1 (156.19 KB, text/plain)
2023-08-11 15:31 CEST, jean-jacques skap
Details
rpm installes à la date du 11 aout 2023 pour mga9RC1 (151.52 KB, text/plain)
2023-08-11 15:32 CEST, jean-jacques skap
Details

Description jean-jacques skap 2023-05-26 21:32:29 CEST
Hello,

I'm using Mageia 8 on a Lenovo Ideapad 320 laptop.

The bluetooth is working, here is the information from the CCM:
Identification
Manufacturer: Realtek 
Description: Bluetooth Radio 
Media class: Wireless|Radio Frequency|Bluetooth
Connection
Bus: USB
PCI bus no.: 1
PCI device no.: 3
Manufacturer identifier: 0x0bda
Device identifier: 0x0821
Miscellaneous
Module: btusb <<<<<<<<<<<<<<<<<<<<<

If I test Mageia 9 live (plasma) or install Mageia 9, the bluetooth does not work.
Here is the information from the Mageia 9 CCM:
Identification
Manufacturer: Realtek 
Description: Bluetooth Radio 
Media class: Wireless|Radio Frequency|Bluetooth
Connection
Bus: USB
PCI bus no.: 1
PCI device no.: 3
Manufacturer identifier: 0x0bda
Device identifier: 0x0821
Miscellaneous
Module: rtl8812au <<<<<<<<<<<<<<<<<
This is not the same module as for the Mageia 8.

Please take into account what seems to me to be a bug.

This text is a machine translation, I don't speak English.
Comment 1 jean-jacques skap 2023-05-28 17:31:55 CEST
Hello

I've just updated my mageia 8.

Bad news, I no longer have access to bluetooth, the rtl8812au module is now present in the CCM instead of the initial btusb module (which worked perfectly before the update).
What's more, I no longer have access to wifi! However, it's still the rtl8821ae module that appears in the CCM.

I now have the same problem between the updated mageia 8 and mageia 9.

Sincerely
Comment 2 Morgan Leijström 2023-05-28 19:12:42 CEST
Thank you for reporting.

Can you provide a list of updated packages between when it worked and when it dont?

As a workaround to get it going you may try urpmi --downgrade option on suspicious packages.

CC: (none) => fri

Comment 3 Morgan Leijström 2023-05-28 19:13:53 CEST
Per Comment 1 problem is also in mga8 after update

Whiteboard: (none) => MGA8TOO

Comment 4 jean-jacques skap 2023-05-29 17:23:07 CEST
Hello,
If I remove the following rpm (installed when updating with CCM) :
- kernel-desktop-6.1.27-2.mga8-1-1.mga8.x86_64
- kernel-desktop-latest-6.1.27-2.mga8.x86_64
Bluetooth works with btusb module
Comment 5 Lewis Smith 2023-05-29 21:58:40 CEST
@jean-jacques
> Can you provide a list of updated packages between when it worked and when
> it dont?
It is worth noting that using:
 $ rpm -qa --last | less
lists packages in date order, most recent first; and is good for seeing what packages have been updated since a certain date, or between two dates.
Please do this if you can.

This looks good to assign to kernel/drivers.

Assignee: bugsquad => kernel

Comment 6 jean-jacques skap 2023-05-30 19:01:39 CEST
Hello,
I have carried out a new installation to reproduce the phenomenon, and am sending you the information you requested.

So that you can follow the progress of my installation, I'm sending you several files.

The instal1_30mai2023.txt file corresponds to the initial installation.

The sup_apres_instal1_30mai2023.txt file corresponds to the rpm I've uninstalled that are of no use to me.

The file instal2_30mai2023.txt corresponds to the installation of the rpm pulseaudio-module-bluetooth-14.2-2.1.mga8.x86_64 proposed by the CCM.

The file instal3_30mai2023.txt is the installation of the kernel-desktop-6.1.27-2.mga8-1-1.mga8.x86_64.rpm which seems to be causing problems.
Uninstalling this rpm reactivates the bluetooth.

Good luck with these documents.
Comment 7 jean-jacques skap 2023-05-30 19:02:45 CEST
Created attachment 13859 [details]
The instal1_30mai2023.txt file corresponds to the initial installation.

The instal1_30mai2023.txt file corresponds to the initial installation.
Comment 8 jean-jacques skap 2023-05-30 19:17:36 CEST
Sorry, I don't speak English at all.

I also don't understand how I can send you several files.

any help in french on how to do this would be great.
Comment 9 christian barranco 2023-08-10 10:32:56 CEST
(In reply to jean-jacques skap from comment #8)
> Sorry, I don't speak English at all.
> 
> I also don't understand how I can send you several files.
> 
> any help in french on how to do this would be great.

Bonjour Jean-Jacques
Il faut attacher les fichiers un à un, en cliquant sur « Add an attachment » à chaque fois, en haut de cette page, dans le bloc Attachments.

CC: (none) => chb0

Comment 10 jean-jacques skap 2023-08-11 15:31:38 CEST
Created attachment 13938 [details]
rpm de l'installtion initiale de mga9RC1
Comment 11 jean-jacques skap 2023-08-11 15:32:30 CEST
Created attachment 13939 [details]
rpm installes à la date du 11 aout 2023 pour mga9RC1
Comment 12 jean-jacques skap 2023-08-11 15:48:55 CEST
Bonjour Christian Barranco, merci pour ce message en français.

Je ne peux malheureusement plus transmettre les fichiers de mon message du 2023-05-30 19:01:39.
J'ai effectué une installation de Mageia9 RC1 avec formatage des partitions.
J'ai estimé que la résolution de ce problème est plus importante sous Mga9 que sous Mga8 en fin de vie.

Je confirme que pour Mga8 le fichier kernel-desktop-6.1.27-2.mga8-1-1.mga8.x86_64.rpm bloque le fonctionnement du bluetooth (module rtl8821ae chargé), sa désinstallation réactive le bluetooth (module btusb chargé).

Je confirme pour Mageia9 RC1 que le CCM indique :
Identification
Manufacturer: Realtek 
Description: Bluetooth Radio 
Media class: Wireless|Radio Frequency|Bluetooth
Connection
Bus: USB
PCI bus no.: 1
PCI device no.: 3
Manufacturer identifier: 0x0bda
Device identifier: 0x0821
Miscellaneous
Module: rtl8812au.

Je viens de transmettre 2 fichiers txt obtenus avec la commande rpm -qa --last | less
installation.txt correspond au rpm de l'installation initiale de Mga9
etat_11_08_2023.txt correspond aux rpm actuellement installés aprés désinstallation de rpm sans utilité pour moi puis une mise à jour à partir du Gestionnaire de fichiers de CCM.

_______________________________________________________________
Hello Christian Barranco, thank you for this message in French.

Unfortunately, I can no longer transmit the files of my message of 2023-05-30 19:01:39.
I performed an installation of Mageia9 RC1 with formatting of the partitions.
I felt that the resolution of this problem is more important under Mga9 than under Mga8 at end of life.

I confirm that for Mga8 the kernel-desktop-6.1.27-2.mga8-1-1.mga8.x86_64.rpm file blocks the operation of bluetooth (rtl8821ae module loaded), uninstalling it reactivates bluetooth (btusb module loaded).

I confirm for Mageia9 RC1 that the CCM indicates:
Identification
Manufacturer: Realtek
Description: Bluetooth Radio
Media class: Wireless|Radio Frequency|Bluetooth
Login
Bus: USB
PCI bus no.: 1
PCI device no.: 3
Manufacturer identifier: 0x0bda
Device identifier: 0x0821
Miscellaneous
Module: rtl8812au.

I just transmitted 2 txt files obtained with the command rpm -qa --last | less
installation.txt corresponds to the rpm of the initial installation of Mga9
etat_11_08_2023.txt corresponds to the currently installed rpm after uninstalling rpm without utility for me then an update from the CCM File Manager.

Bon courage, Good luck

Merci aux développeurs pour cette nouvelle Mageia
Thanks to the developers for this new Mageia
Comment 13 jean-jacques skap 2023-08-27 18:06:42 CEST
Hello, I confirm my previous messages.

I'm testing the final version of Mageia9 with the 6.4.9-desktop-4.mga9 (64-bit) kernel, using the live plasma version.

Bluetooth does not work. It is the rtl8821ae module that appears in the CCM.

I remind you that under Mageia 8 the CCM indicates as a btusb module and that the bluetooth works.

Under Mageia9 the Bluetooth section of System Configuration indicates that the bluetooth device is disabled. Pressing the Activate button does not change anything.

In console the iwconfig command returns:
lo no wireless extensions.

enp2s0 no wireless extensions.

wlp3s0 IEEE 802.11 ESSID:"Livebox-C976"
          Mode:Managed Frequency:2.462 GHz Access Point: E8:1B:69:04:3F:86
          Bit Rate=72.2 Mb/s Tx-Power=20 dBm
          Retry short limit:7 RTS thr=2347 B Fragment thr:off
          Power Management:on
          Link Quality=70/70 Signal level=-32dBm
          Rx invalid nwid:0 Rx invalid crypt:0 Rx invalid frag:0
          Excessive retries:0 Invalid misc:1 Missed beacon:0

wlp0s20f0u7 unassociated ESSID:"" Nickname:"<WIFI@REALTEK>"
          Mode:Managed Frequency=2.412 GHz Access Point: Not-Associated
          Sensitivity:0/0
          Retry:off RTS thr:off Fragment thr:off
          Power Management: off
          Link Quality:0 Signal level:0 Noise level:0
          Rx invalid nwid:0 Rx invalid crypt:0 Rx invalid frag:0
          Excessive retries:0 Invalid misc:0 Missed beacon:0

wlp0s20f0u7i1 unassociated ESSID:"" Nickname:"<WIFI@REALTEK>"
          Mode:Managed Frequency=2.412 GHz Access Point: Not-Associated
          Sensitivity:0/0
          Retry:off RTS thr:off Fragment thr:off
          Power Management: off
          Link Quality:0 Signal level:0 Noise level:0
          Rx invalid nwid:0 Rx invalid crypt:0 Rx invalid frag:0
          Excessive retries:0 Invalid misc:0 Missed beacon:0

the rfkill list command returns:
0: ideapad_wlan: Wireless LAN
        Soft blocked: no
        Hard blocked: no
1:ideapad_bluetooth:Bluetooth
        Soft blocked: no
        Hard blocked: no
2: phy0: Wireless LAN
        Soft blocked: no
        Hard blocked: no
3: phy1: Wireless LAN
        Soft blocked: no
        Hard blocked: no
4: phy2: Wireless LAN
        Soft blocked: no
        Hard blocked: no

Bon courage, Good luck

Merci aux développeurs pour cette nouvelle Mageia
Thanks to the developers for this new Mageia

Bonjour, je confirme mes précédents messages.

Je teste la version définitive de Mageia9 avec le kernel 6.4.9-desktop-4.mga9 (64-bit), en utilisant la version live plasma.

Le bluetooth ne fonctionne pas. C'est le module rtl8821ae qui apparait dans le CCM.

Je rappelle que sous Mageia 8 le CCM indique comme module btusb et que le bluetooth fonctionne.

Sous Mageia9 la rubrique Bluetooth de Configuration du système indique que le périphérique bluetooth est désactivé. L'appui sur le bouton Activer ne change rien.

En console la commande iwconfig retourne :
lo        no wireless extensions.

enp2s0    no wireless extensions.

wlp3s0    IEEE 802.11  ESSID:"Livebox-C976"  
          Mode:Managed  Frequency:2.462 GHz  Access Point: E8:1B:69:04:3F:86   
          Bit Rate=72.2 Mb/s   Tx-Power=20 dBm   
          Retry short limit:7   RTS thr=2347 B   Fragment thr:off
          Power Management:on
          Link Quality=70/70  Signal level=-32 dBm  
          Rx invalid nwid:0  Rx invalid crypt:0  Rx invalid frag:0
          Tx excessive retries:0  Invalid misc:1   Missed beacon:0

wlp0s20f0u7  unassociated  ESSID:""  Nickname:"<WIFI@REALTEK>"
          Mode:Managed  Frequency=2.412 GHz  Access Point: Not-Associated   
          Sensitivity:0/0  
          Retry:off   RTS thr:off   Fragment thr:off
          Power Management:off
          Link Quality:0  Signal level:0  Noise level:0
          Rx invalid nwid:0  Rx invalid crypt:0  Rx invalid frag:0
          Tx excessive retries:0  Invalid misc:0   Missed beacon:0

wlp0s20f0u7i1  unassociated  ESSID:""  Nickname:"<WIFI@REALTEK>"
          Mode:Managed  Frequency=2.412 GHz  Access Point: Not-Associated   
          Sensitivity:0/0  
          Retry:off   RTS thr:off   Fragment thr:off
          Power Management:off
          Link Quality:0  Signal level:0  Noise level:0
          Rx invalid nwid:0  Rx invalid crypt:0  Rx invalid frag:0
          Tx excessive retries:0  Invalid misc:0   Missed beacon:0

la commande rfkill list retourne :
0: ideapad_wlan: Wireless LAN
        Soft blocked: no
        Hard blocked: no
1: ideapad_bluetooth: Bluetooth
        Soft blocked: no
        Hard blocked: no
2: phy0: Wireless LAN
        Soft blocked: no
        Hard blocked: no
3: phy1: Wireless LAN
        Soft blocked: no
        Hard blocked: no
4: phy2: Wireless LAN
        Soft blocked: no
        Hard blocked: no

Bon courage, Good luck

Merci aux développeurs pour cette nouvelle Mageia
Thanks to the developers for this new Mageia
Comment 14 christian barranco 2023-08-27 23:54:24 CEST
Hi Jean-Jacques
Do you use PulseAudio or Pipewire?
Comment 15 Morgan Leijström 2023-08-28 00:29:48 CEST
Maybe you can test with a kernel from Mageia 8 backport?
Comment 16 jean-jacques skap 2023-08-28 17:06:06 CEST
Bonjour,

J'utilise Pulse-audio dans le mode Glitch-Free
Comment 17 christian barranco 2023-08-28 18:01:28 CEST
(In reply to jean-jacques skap from comment #16)
> Bonjour,
> 
> J'utilise Pulse-audio dans le mode Glitch-Free

Ok. Because checking your hardware through the CCM leads to the installation of the pulseaudio bluetooth module, which might conflict with pipewire. 

Morgan’s idea to test a kernel from MGA8 backport is a good idea to debug.
Comment 18 jean-jacques skap 2023-08-29 17:12:00 CEST
(In reply to christian barranco from comment #17)
> Ok. Because checking your hardware through the CCM leads to the installation
> of the pulseaudio bluetooth module, which might conflict with pipewire. 
The pulseaudio-module-bluetooth rpm is not installed on my computer.


> Morgan’s idea to test a kernel from MGA8 backport is a good idea to debug.
Morgan Leijström's proposal is beyond my capabilities.
I'm just a basic user with no computer skills.
I'm sorry.

Bonjour,
La proposition de Morgan Leijström dépasse mes compétences.
Je suis un simple utilisateur de base, sans compétence informatique.
Je suis désolé.

Le rpm pulseaudio-module-bluetooth n'est pas installé sur mon ordinateur
Comment 19 jean-jacques skap 2023-09-05 14:27:54 CEST
hello,
I solved the bluetooth problem by deleting the :
/usr/lib/modules/6.x.x-desktop-4.mga9/kernel/drivers/staging/rtl8812au/8812au.ko.xz

After restarting the computer, in the CCM access to Hardware asks me to install pulseaudio-module-bluetooth. I confirm and restart the computer.

In the Hardware entry in the CCM, a new section is available "Bluetooth devices", where my Bluetooth device is present with the btusb module.

Bluetooth now works in the direction computer > smartphone and smartphone > computer.
Comment 20 christian barranco 2023-09-05 17:39:52 CEST
Hi

Could it be that the 6.3.4-2 update was not enough to bring the full support for rtl8812au?

+ Revision: 1958618
- update staging/rtl8812au wifi driver
Comment 21 christian barranco 2023-09-05 17:55:14 CEST
Maybe some hints there?
https://github.com/morrownr/8812au-20210629
Comment 22 jean-jacques skap 2023-09-15 21:09:15 CEST
Hello Christian
I have installed the final version of mageia9 (new installation and not updated) the kernel is 6.4.9-desktop-4.mga9 (64-bit) and the bluetooth problem persists.

As I indicated on 05/09/2023, "I solved the bluetooth problem by deleting the: /usr/lib/modules/6.x.x-desktop-4.mga9/kernel/drivers/staging/rtl8812au/8812au.ko.xz".
Comment 23 jean-jacques skap 2023-09-15 21:12:44 CEST
Christian, 

I consulted the link 
https://github.com/morrownr/8812au-20210629

It seems to be beyond my skills as a simple user.
Comment 24 christian barranco 2023-09-15 22:12:25 CEST
Hi Jean-Jacques.
Indeed, we need tmb (Thomas Backlund) feedback here.

CC: (none) => thomas.backlund

Comment 25 jean-jacques skap 2023-11-09 19:58:09 CET
Bopnjour,

Malgré la mise à jour de Mageia9 et le passage au kernel 6.4.16-desktop-3.mga9, le problème persiste.

C'est toujours le module rtl8821ae qui apparait dans le CCM dans la rubrique Inconnus/Autres du matériel.

J'efface donc le fichier /usr/lib/modules/6.4.16-desktop-3.mga9/kernel/drivers/staging/rtl8812au/8812au.ko.xz
Je relance l'ordnateur et le CCM m'indique que le module btusb est utilisé pour mon périphérique bluetooth.
Un test de connexion avec mon smartphone confirme le bon fonctionnement.

Conclusion le module rtl8821ae est utilisé à tort pour le périphérique bluetooth intégré à mon portable.
Comment 26 christian barranco 2023-11-09 23:29:16 CET
(In reply to jean-jacques skap from comment #25)
> Bopnjour,
> 
> Malgré la mise à jour de Mageia9 et le passage au kernel
> 6.4.16-desktop-3.mga9, le problème persiste.
> 
> C'est toujours le module rtl8821ae qui apparait dans le CCM dans la rubrique
> Inconnus/Autres du matériel.
> 
> J'efface donc le fichier
> /usr/lib/modules/6.4.16-desktop-3.mga9/kernel/drivers/staging/rtl8812au/
> 8812au.ko.xz
> Je relance l'ordnateur et le CCM m'indique que le module btusb est utilisé
> pour mon périphérique bluetooth.
> Un test de connexion avec mon smartphone confirme le bon fonctionnement.
> 
> Conclusion le module rtl8821ae est utilisé à tort pour le périphérique
> bluetooth intégré à mon portable.

8821ae or 8812au ?
Comment 27 jean-jacques skap 2023-11-10 18:53:04 CET
Bonjour,

Erreur de plume de ma part c'est bien rtl8812au qui pose problème.

J'ai inst
Comment 28 jean-jacques skap 2023-11-10 19:01:36 CET
Bonjour,

Erreur de plume de ma part c'est bien rtl8812au qu'il faut lire.

J'ai installé le kernel-linus 6.4.16-3.mga9, hier. Le buetooth fonctionne, bien sur le fichier kernel/drivers/staging/rtl8812au/8812au.ko.xz est absent avec ce kernel.

J'ajoute que sous mageia 8 (mon premier probleme est apparu avec une mise à jour de cette distribution) ce fichier apparait avec le kernel 6.1.15-1 avec un non fonctionnment du bluetooth. Jusqu'au kernel 6.0.12-1 ce module est absent et le bluetooth fonctionnait parfaitement.
Comment 29 jean-jacques skap 2023-11-10 19:26:06 CET
En complément :

Dans votre lien du 2023-09-05 17:55:14 CEST  il est précisé :
8812au ( 8812au.ko ) Linux Driver for USB WiFi Adapters that are based on the RTL8812AU Chipset
or il ne s'agit pas d'une clé USb wifi mais bien  du composant intégré à un ordinateur portable.

De plus le fichier supported-device-IDs donne semble t'il la liste des périphériques supportés par ce module
(Note: Please read the file "supported-device-IDs" for information about how to confirm that this is the correct driver for your adapter.)
Mon périphérique ne figure pas dans ce fichier. voici le résultat de lsusb : Bus 001 Device 003: ID 0bda:0821 Realtek Semiconductor Corp. RTL8821A Bluetooth

Ma conclusion est que c'est à tort que Magiea utilise ce module 8812au depuis Magei8 avec kernel 6.1.15-1 et suivant
christian barranco 2023-11-10 23:57:37 CET

CC: (none) => ghibomgx

Comment 30 jean-jacques skap 2023-12-10 15:35:05 CET
Bonjour,

Après la mise à jour de Mageia9 et le passage au kernel 6.5.11-desktop-5.mga9 (64-bit), le problème persiste.

C'est toujours le module rtl8821au qui apparait dans le CCM dans la rubrique Inconnus/Autres du matériel.

J'efface donc le fichier /usr/lib/modules/6.5.11-desktop-5.mga9/kernel/drivers/staging/rtl8812au/8812au.ko.xz
Je relance l'ordnateur et le CCM m'indique que le module btusb est utilisé pour mon périphérique bluetooth. Le test de connexion avec mon smartphone confirme le bon fonctionnement.

Conclusion: toujours la même, le module rtl8821au est utilisé à tort pour le périphérique bluetooth ID 0bda:0821 (voir mon message du 2023-11-10).
Comment 31 Giuseppe Ghibò 2023-12-12 17:43:29 CET
Please use report in english.

For your problem try also with kernel-desktop-6.5.13-1.mga9 in updates_testing, and be sure to include also kernel-firmware*-2023111* from the same source.
Comment 32 jean-jacques skap 2023-12-13 21:40:07 CET
Hello,

This text is an automatic translation.

The problem is not solved with kernel-desktop-6.5.13-1.mga9 and kernel-firmware*-2023111*.

If I open kernel-desktop-6.5.13-1.mga9.x86_64.rpm with ark I notice that in the file /lib/modules/6.5.13-desktop-1.mga9/modules.alias line 24184 contains "alias usb:v0BDAp0821d*dc*dsc*dp*ic*isc*ip*in* 8812au".
This line seems to associate my device ID 0bda:0821 with the 8812au module, which is incorrect (see my message of 2023-11-10).

I think this info is in the file /lib/modules/6.5.13-desktop-1.mga9/modules.alias.bin file (am I wrong???). If so, it's in this file that the erroneous association must be broken.

Bonjour,

Ce texte est une traduction automatique.

Le problème n'est pas résolu avec kernel-desktop-6.5.13-1.mga9 et kernel-firmware*-2023111*

Si j'ouvre kernel-desktop-6.5.13-1.mga9.x86_64.rpm avec ark je constate que dans le fichier /lib/modules/6.5.13-desktop-1.mga9/modules.alias la ligne 24184 contient "alias usb:v0BDAp0821d*dc*dsc*dp*ic*isc*ip*in* 8812au".
Cette ligne semble associer mon périphérique ID 0bda:0821 au module 8812au ce qui est erroné (voir mon message du 2023-11-10)

Je pense que cette info est reprise dans le fichier /lib/modules/6.5.13-desktop-1.mga9/modules.alias.bin (je me trompe ???). Si c'est le cas c'est dans ce fichier que le lien doit être cassé.
Comment 33 jean-jacques skap 2023-12-13 21:43:22 CET
translation of my message of 2023-11-10 19:26:06 CET

In addition:

In your link of 2023-09-05 17:55:14 CEST it says :
8812au ( 8812au.ko ) Linux Driver for USB WiFi Adapters that are based on the RTL8812AU Chipset
but it's not a USB WiFi key, but a component built into a laptop.

The supported-device-IDs file also appears to list the devices supported by this module.
(Note: Please read the file "supported-device-IDs" for information about how to confirm that this is the correct driver for your adapter.)
My device is not in this file. Here is the result of lsusb: Bus 001 Device 003: ID 0bda:0821 Realtek Semiconductor Corp. RTL8821A Bluetooth
Comment 34 Giuseppe Ghibò 2023-12-13 23:46:47 CET
I think it's not related to the fact that the component is builtin (probably it's  using the same chipset as wifi keys).

Anyway the the 8812au modules is an off-tree modules not part of stock kernel, which was added to the mageia kernel. Probably 8812au needs to be updated to a newer version.

Are you able to do some tiny compilation of modules locally for testing?
Comment 35 Guillaume Bedot 2023-12-14 20:38:31 CET
@jean-jacques:
When you delete the kernel module 8812au.ko , does your wi-fi work too ?

As a last resort, you can create a file named /etc/modprobe.d/blacklist-bad-module.conf containing : blacklist 8812au, and you won't have to delete the module after each kernel update.

@all:
I don't understand why there are so many places with modules for this chipset https://github.com/morrownr/8821au-20210708
https://github.com/morrownr/8812au-20210629

CC: (none) => geex+mageia

Comment 36 Giuseppe Ghibò 2023-12-14 20:45:57 CET
(In reply to Guillaume Bedot from comment #35)

> @all:
> I don't understand why there are so many places with modules for this
> chipset https://github.com/morrownr/8821au-20210708
> https://github.com/morrownr/8812au-20210629

Probably depends on what realtek released and then got unmaintained. AFAIK the one actually used is this one, and not the above:

https://github.com/aircrack-ng/rtl8812au.git

Anyway there is kernel 6.5.13-3.mga9 in backports_testing which should address some update to this module.
Comment 37 Morgan Leijström 2023-12-15 17:54:55 CET
@Guillaume please test and we might put this kernel in normal updates.
I am testing it on my system  (i dont have bluetooth, test other things)
https://bugs.mageia.org/show_bug.cgi?id=32623#c10
Comment 38 jean-jacques skap 2023-12-15 23:26:31 CET
Hello and thank you for your answers

---> "Are you able to do some tiny compilation of modules locally for testing?"
me : I'm just a basic user, who also doesn't speak English. I have consulted the web pages on https://github.com concerning the compilation of the module and it is beyond my competence.

---> "When you delete the kernel module 8812au.ko , does your wi-fi work too ? "
me : wifi has always used the rtl8821ae module without any problems

---> "As a last resort, you can create a file named
/etc/modprobe.d/blacklist-bad-module.conf containing : blacklist 8812au, and
you won't have to delete the module after each kernel update."
me : I tested this solution and it seems to work
Comment 39 Giuseppe Ghibò 2023-12-15 23:46:34 CET
(In reply to jean-jacques skap from comment #38)
> Hello and thank you for your answers
> 
> ---> "Are you able to do some tiny compilation of modules locally for
> testing?"
> me : I'm just a basic user, who also doesn't speak English. I have consulted
> the web pages on https://github.com concerning the compilation of the module
> and it is beyond my competence.

Ok, let's skip self compilation, can you try just with kernel-desktop-6.5.13-3.mga9 from backports_testing?
Comment 40 jean-jacques skap 2023-12-16 19:36:39 CET
(In reply to Giuseppe Ghibò from comment #39)
> can you try just with
> kernel-desktop-6.5.13-3.mga9 from backports_testing?

I just tested, the result is still negative.
my device is still associated with module rtl8812au
if I delete the 8812au.ko.xz file or blacklist it, when I reboot, the btusb module is used and works.

I think we need to break the association between this module rtl8812au and the device :
Manufacturer identifier: 0x0bda
Device identifier: 0x0821

This way, devices compatible with this module will still be able to use it.
Comment 41 Guillaume Bedot 2023-12-16 19:45:01 CET
(In reply to Morgan Leijström from comment #37)
> @Guillaume please test and we might put this kernel in normal updates.

I do not own the same bluetooth chipset as the reporter either. I aws just trying to give him some tips to handle these kind of situations.
Comment 42 Morgan Leijström 2023-12-16 20:28:51 CET
Ah sorry

@jean-jacques

Can you try kernel 6.5.13-3.mga9 in backports_testing ?
Comment 43 Giuseppe Ghibò 2023-12-16 20:30:03 CET
(In reply to Morgan Leijström from comment #42)
> Ah sorry
> 
> @jean-jacques
> 
> Can you try kernel 6.5.13-3.mga9 in backports_testing ?

He did, didn't worked.
Comment 44 Giuseppe Ghibò 2023-12-16 21:15:12 CET
(In reply to jean-jacques skap from comment #40)
> (In reply to Giuseppe Ghibò from comment #39)
> > can you try just with
> > kernel-desktop-6.5.13-3.mga9 from backports_testing?
> 
> I just tested, the result is still negative.
> my device is still associated with module rtl8812au
> if I delete the 8812au.ko.xz file or blacklist it, when I reboot, the btusb
> module is used and works.
> 
> I think we need to break the association between this module rtl8812au and
> the device :
> Manufacturer identifier: 0x0bda
> Device identifier: 0x0821
> 
> This way, devices compatible with this module will still be able to use it.

What is weird is that with with rtl8812au.ko it doesn't work as apparently the code that uses device ID 0x0821 is this one (which I suppose initially comes from Realtek):

https://github.com/aircrack-ng/rtl8812au/blob/v5.6.4.2/os_dep/linux/usb_intf.c#L188

Anyway, it lists those ID as "=== Realtek demoboard ===". An hyphotesis is that those IDs were recycled for other devices later. So I wonder if other devices with IDs 0x0811, 0x0820, 0x0823, 0x8821, 0xA811 requires the same treatment.

Please test with kernel-desktop-6.5.13-5.mga9 (from mga9's core/backports_testing) tomorrow.
Comment 45 jean-jacques skap 2023-12-17 18:55:37 CET
GOOD !

With kernel-desktop-6.5.13-5.mga9 it's OK !

The module loaded is now btusb (as with Mageia 8 and kernels 5.x.x.x.).

A successful transfer from PC to smartphone and vice versa.

Thanks for all your hard work.

Good luck and long live Mageia

I'll set the status to resolute when the kernel is no longer in testing but definitive.
Comment 46 Morgan Leijström 2023-12-17 19:33:00 CET
Nice!

I will now test this kernel on my system.

Giuseppe, i think we should retarget
  Bug 32623 - Update request: kernel-6.5.13-1.mga9
to this -5 kernel?

And then set this bug to depend on that.
Comment 47 Giuseppe Ghibò 2023-12-17 21:55:56 CET
(In reply to Morgan Leijström from comment #46)
> Nice!
> 
> I will now test this kernel on my system.
> 
> Giuseppe, i think we should retarget
>   Bug 32623 - Update request: kernel-6.5.13-1.mga9
> to this -5 kernel?
> 
> And then set this bug to depend on that.

kernel-6.5.13-6.mga9 (so even) to updates_testing. Should be the final one for 32623.
Morgan Leijström 2023-12-17 22:04:29 CET

Depends on: (none) => 32623

Comment 48 Giuseppe Ghibò 2023-12-17 22:08:25 CET
(In reply to jean-jacques skap from comment #45)
> GOOD !
> 
> With kernel-desktop-6.5.13-5.mga9 it's OK !
> 
> The module loaded is now btusb (as with Mageia 8 and kernels 5.x.x.x.).
> 
> A successful transfer from PC to smartphone and vice versa.
> 
> Thanks for all your hard work.
> 
> Good luck and long live Mageia
> 
> I'll set the status to resolute when the kernel is no longer in testing but
> definitive.

I think it should be reported also to the github page maintaining the extra modules https://github.com/aircrack-ng/rtl8812au/issues so that next releases won't require extra patches anymore.
Comment 49 Guillaume Bedot 2023-12-18 07:47:10 CET
It seems already reverted, maybe it took some time to get to staging ?

https://github.com/aircrack-ng/rtl8812au/commit/ca797e114b0640d6aa1a573ea68526a9263df81e
Comment 50 Giuseppe Ghibò 2023-12-18 11:22:32 CET
(In reply to Guillaume Bedot from comment #49)

> It seems already reverted, maybe it took some time to get to staging ?
> 
> https://github.com/aircrack-ng/rtl8812au/commit/
> ca797e114b0640d6aa1a573ea68526a9263df81e

Indeed that's another, and was already included in 6.5.13-3-mga9 fixes which hadn't effects. See:

https://svnweb.mageia.org/packages/backports/9/kernel/current/SOURCES/staging-rtl8812au-20231117.patch?revision=2018037&view=markup
Comment 51 Guillaume Bedot 2023-12-18 13:16:09 CET
oh, but, there are 13 patches in there ! The first seems identical, maybe one of the others caused issues too.
Comment 53 Giuseppe Ghibò 2023-12-18 13:32:40 CET
(In reply to Guillaume Bedot from comment #52)

> sorry, apparently the final fix is :
> https://svnweb.mageia.org/packages/backports/9/kernel/current/SOURCES/0001-
> Pull-device-0x0821-from-list-of-device-supported-by-.
> patch?revision=2018471&view=markup&sortby=date

Yes, that's the "fix" we used.
Comment 54 Guillaume Bedot 2023-12-18 14:00:03 CET
(In reply to Giuseppe Ghibò from comment #53)
> Yes, that's the "fix" we used.

Clear uniline fix, and patch that removes code at the same time, be proud of it.
Comment 55 Guillaume Bedot 2023-12-18 15:59:09 CET
There was a time when, in harddrake, one could choose between the modules that handles sound (oss or alsa), or graphics (fglrx or radeon), could this be an more general way to fix this kind of bug ? IMHO, working out of the box is better, but...
Comment 56 Morgan Leijström 2023-12-26 00:29:39 CET
Fix is in now released mga9 kernel.
Cauldron is supposed to have the same.
MGA8 is to be EOL

Status: NEW => RESOLVED
Resolution: (none) => FIXED
Whiteboard: MGA8TOO => MGA9TOO


Note You need to log in before you can comment on or make changes to this bug.