[Bug report] - RTL8821CE 802.11ac PCIe Wireless Network Adapter - Not handled properly by the rtw_8821ce dirver on Kernel 5.11.11

[Bug report] - RTL8821CE 802.11ac PCIe Wireless Network Adapter - Not handled properly by the rtw_8821ce dirver on Kernel 5.11.11

[Michal TOMA]

The RTL8821CE driver doesn't work on several recent kernels (probably since 
5.9).


journalctl -b | grep rtw
avril 10 14:18:41 localhost kernel: rtw_8821ce 0000:07:00.0: Firmware version 24.8.0, H2C version 12
avril 10 14:18:41 localhost NetworkManager[682]: <info>  [1618057121.6199] rfkill2: found Wi-Fi radio killswitch (at /sys/devices/pci0000:00/0000:00:1d.4/0000:07:00.0/ieee80211/phy0/rfkill2) (driver rtw_8821ce)
avril 10 14:18:41 localhost kernel: rtw_8821ce 0000:07:00.0 wlp7s0: renamed from wlan0
avril 10 14:18:42 localhost kernel: rtw_8821ce 0000:07:00.0: start vif 28:39:26:ab:b4:5d on port 0
avril 10 14:18:42 localhost kernel: rtw_8821ce 0000:07:00.0: stop vif 28:39:26:ab:b4:5d on port 0
avril 10 14:18:42 localhost kernel: rtw_8821ce 0000:07:00.0: start vif 5a:f0:11:63:85:65 on port 0
avril 10 14:25:35 Host-001 kernel: rtw_8821ce 0000:07:00.0: stop vif 5a:f0:11:63:85:65 on port 0
avril 10 14:25:35 Host-001 kernel: rtw_8821ce 0000:07:00.0: start vif 32:45:38:02:34:09 on port 0
avril 10 14:32:28 Host-001 kernel: rtw_8821ce 0000:07:00.0: stop vif 32:45:38:02:34:09 on port 0
avril 10 14:32:28 Host-001 kernel: rtw_8821ce 0000:07:00.0: start vif c2:5c:03:32:bc:19 on port 0
avril 10 14:39:21 Host-001 kernel: rtw_8821ce 0000:07:00.0: stop vif c2:5c:03:32:bc:19 on port 0
avril 10 14:39:21 Host-001 kernel: rtw_8821ce 0000:07:00.0: start vif de:9c:ac:b2:58:ae on port 0
avril 10 14:46:14 Host-001 kernel: rtw_8821ce 0000:07:00.0: stop vif de:9c:ac:b2:58:ae on port 0
avril 10 14:46:14 Host-001 kernel: rtw_8821ce 0000:07:00.0: start vif 46:5c:c2:91:f8:7f on port 0
...
And continues like this indefinitely

uname -a
Linux Host-001 5.11.11-1-default #1 SMP Tue Mar 30 17:57:52 UTC 2021 (dbc4a02) x86_64 x86_64 x86_64 GNU/Linux


rfkill list 
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

Re: [Bug report] - RTL8821CE 802.11ac PCIe Wireless Network Adapter - Not handled properly by the rtw_8821ce dirver on Kernel 5.11.11

[Larry Finger]

On 4/14/21 4:12 AM, Michal TOMA wrote:
> The RTL8821CE driver doesn't work on several recent kernels (probably since
> 5.9).
> 
> 
> journalctl -b | grep rtw
> avril 10 14:18:41 localhost kernel: rtw_8821ce 0000:07:00.0: Firmware version 24.8.0, H2C version 12
> avril 10 14:18:41 localhost NetworkManager[682]: <info>  [1618057121.6199] rfkill2: found Wi-Fi radio killswitch (at /sys/devices/pci0000:00/0000:00:1d.4/0000:07:00.0/ieee80211/phy0/rfkill2) (driver rtw_8821ce)
> avril 10 14:18:41 localhost kernel: rtw_8821ce 0000:07:00.0 wlp7s0: renamed from wlan0
> avril 10 14:18:42 localhost kernel: rtw_8821ce 0000:07:00.0: start vif 28:39:26:ab:b4:5d on port 0
> avril 10 14:18:42 localhost kernel: rtw_8821ce 0000:07:00.0: stop vif 28:39:26:ab:b4:5d on port 0
> avril 10 14:18:42 localhost kernel: rtw_8821ce 0000:07:00.0: start vif 5a:f0:11:63:85:65 on port 0
> avril 10 14:25:35 Host-001 kernel: rtw_8821ce 0000:07:00.0: stop vif 5a:f0:11:63:85:65 on port 0
> avril 10 14:25:35 Host-001 kernel: rtw_8821ce 0000:07:00.0: start vif 32:45:38:02:34:09 on port 0
> avril 10 14:32:28 Host-001 kernel: rtw_8821ce 0000:07:00.0: stop vif 32:45:38:02:34:09 on port 0
> avril 10 14:32:28 Host-001 kernel: rtw_8821ce 0000:07:00.0: start vif c2:5c:03:32:bc:19 on port 0
> avril 10 14:39:21 Host-001 kernel: rtw_8821ce 0000:07:00.0: stop vif c2:5c:03:32:bc:19 on port 0
> avril 10 14:39:21 Host-001 kernel: rtw_8821ce 0000:07:00.0: start vif de:9c:ac:b2:58:ae on port 0
> avril 10 14:46:14 Host-001 kernel: rtw_8821ce 0000:07:00.0: stop vif de:9c:ac:b2:58:ae on port 0
> avril 10 14:46:14 Host-001 kernel: rtw_8821ce 0000:07:00.0: start vif 46:5c:c2:91:f8:7f on port 0
> ...
> And continues like this indefinitely
> 
> uname -a
> Linux Host-001 5.11.11-1-default #1 SMP Tue Mar 30 17:57:52 UTC 2021 (dbc4a02) x86_64 x86_64 x86_64 GNU/Linux
> 
> 
> rfkill list
> 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
> 

Please copy file dmesg.txt from 'dmesg > dmesg.txt' to some pastebin that I can 
access. The sending of only lines with rtw in them may miss some valid information.

Larry

Re: [Bug report] - RTL8821CE 802.11ac PCIe Wireless Network Adapter - Not handled properly by the rtw_8821ce dirver on Kernel 5.11.11

[Larry Finger]

On 4/15/21 5:09 AM, michaltoma wrote:
> Hello Larry,
> 
> 
> Here is the full dmesg:
> 
> https://pastebin.com/ZUXQeKAt <https://pastebin.com/ZUXQeKAt>

On this list, please do not top post.

Thanks for the upload. Those 80 BIOS Errors do not lend much confidence in the 
quality of your BIOS! Have you checked for updates?

Have you tried when setting the options disable_msi=y and disable_aspm=y for 
module rtw88_pci if you use the kernel driver, or module rtw_pci if you use my 
GitHub repo. That fixes a lot of strange errors.

Larry

Re: [Bug report] - RTL8821CE 802.11ac PCIe Wireless Network Adapter - Not handled properly by the rtw_8821ce dirver on Kernel 5.11.11

[Larry Finger]

On 4/19/21 4:56 AM, michaltoma wrote:
> 
>  >*De :*Larry Finger <larry.finger@gmail.com> de la part de Larry Finger 
> <Larry.Finger@lwfinger.net>
>  >*Envoyé :*jeudi 15 avril 2021 19:57
>  >*À :*michaltoma; linux-wireless@vger.kernel.org
>  >*Objet :*Re: [Bug report] - RTL8821CE 802.11ac PCIe Wireless Network Adapter - 
> Not handled properly by the rtw_8821ce dirver on Kernel 5.11.11
>  >
>> > On 4/15/21 5:09 AM, michaltoma wrote:
>> > Hello Larry,
>> > 
>> > 
>> > Here is the full dmesg:
>> > 
>> > https://pastebin.com/ZUXQeKAt <https://pastebin.com/ZUXQeKAt> 
> <https://pastebin.com/ZUXQeKAt <https://pastebin.com/ZUXQeKAt>>
>> 
>> On this list, please do not top post.
>> 
>> Thanks for the upload. Those 80 BIOS Errors do not lend much confidence in the 
>> quality of your BIOS! Have you checked for updates?
>> 
>> Have you tried when setting the options disable_msi=y and disable_aspm=y for 
>> module rtw88_pci if you use the kernel driver, or module rtw_pci if you use my 
>> GitHub repo. That fixes a lot of strange errors.
>> 
>> Larry
> 
> Hello Larry,
> 
> I tried to update the BIOS (without success so far as Lenovo doesn't provide 
> anything else than a windows 32 bit exe file).
> But this led me to boot a windows 10 on my laptop. Surprisingly this somehow 
> "unlocked" the Wifi.
> Seems the issue is somehow related to the chip deep sleep mode (I remember 
> reading something about this a while back), from which the driver is unable to 
> wake it up but the windows driver manages to do so.
> Not sure if this might help in diagnosing the problem but just reporting back.

If you boot Windows and Linux, it is absolutely imperative that you disable 
"quick boot" in Windows. That mode sometimes leaves devices disables for other 
OS's. I think this is what happened here.

Larry

Re: [Bug report] - RTL8821CE 802.11ac PCIe Wireless Network Adapter - Not handled properly by the rtw_8821ce dirver on Kernel 5.11.11

[Larry Finger]

On 4/19/21 2:50 PM, michaltoma wrote:
> 
> *> De :* Larry Finger <larry.finger@gmail.com> de la part de Larry Finger 
> <Larry.Finger@lwfinger.net>
> *> Envoyé :* lundi 19 avril 2021 19:38
> *> À :* michaltoma; linux-wireless@vger.kernel.org
> *> Objet :* Re: [Bug report] - RTL8821CE 802.11ac PCIe Wireless Network Adapter 
> - Not handled properly by the rtw_8821ce dirver on Kernel 5.11.11
>  >
>  > On 4/19/21 4:56 AM, michaltoma wrote:
>  > >
>  > >  >*De :*Larry Finger <larry.finger@gmail.com> de la part de Larry Finger
>  > > <Larry.Finger@lwfinger.net>
>  > >  >*Envoyé :*jeudi 15 avril 2021 19:57
>  > >  >*À :*michaltoma; linux-wireless@vger.kernel.org
>  > >  >*Objet :*Re: [Bug report] - RTL8821CE 802.11ac PCIe Wireless Network 
> Adapter -
>  > > Not handled properly by the rtw_8821ce dirver on Kernel 5.11.11
>  > >  >
>  > >> > On 4/15/21 5:09 AM, michaltoma wrote:
>  > >> > Hello Larry,
>  > >> >
>  > >> >
>  > >> > Here is the full dmesg:
>  > >> >
>  > >> > https://pastebin.com/ZUXQeKAt <https://pastebin.com/ZUXQeKAt> 
> <https://pastebin.com/ZUXQeKAt <https://pastebin.com/ZUXQeKAt>>
>  > > <https://pastebin.com/ZUXQeKAt <https://pastebin.com/ZUXQeKAt 
> <https://pastebin.com/ZUXQeKAt>>>
>  > >>
>  > >> On this list, please do not top post.
>  > >>
>  > >> Thanks for the upload. Those 80 BIOS Errors do not lend much confidence in 
> the
>  >>> quality of your BIOS! Have you checked for updates?
>  >>>
>  >>> Have you tried when setting the options disable_msi=y and disable_aspm=y for
>  >>> module rtw88_pci if you use the kernel driver, or module rtw_pci if you use my
>  >>> GitHub repo. That fixes a lot of strange errors.
>  >>>
>  >>> Larry
>  >>
>  >> Hello Larry,
>  >>
>  >> I tried to update the BIOS (without success so far as Lenovo doesn't provide
>  >> anything else than a windows 32 bit exe file).
>  >> But this led me to boot a windows 10 on my laptop. Surprisingly this somehow
>  >> "unlocked" the Wifi.
>  >> Seems the issue is somehow related to the chip deep sleep mode (I remember
>  >> reading something about this a while back), from which the driver is unable to
>  >> wake it up but the windows driver manages to do so.
>  >> Not sure if this might help in diagnosing the problem but just reporting back.
>  >
>  > If you boot Windows and Linux, it is absolutely imperative that you disable
>  > "quick boot" in Windows. That mode sometimes leaves devices disables for other
>  > OS's. I think this is what happened here.
>  >
>  > Larry
> 
> Unfortunately no. I'm only booting Linux since day 0 of the laptop. This was 
> only required to try to flash the bios so I installed a copy of windows on an 
> external drive. __

Someone (the Factory) may have had Windows on it and did a quick-boot shutdown. 
I do not know any other way the iface could have been left in the wrong state.

Larry