-
Notifications
You must be signed in to change notification settings - Fork 179
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Need Testers: New in-kernel (mac80211) driver for rtl8821au, rtl8811au and rtl8812au chipsets. #430
Comments
My TP Link T2Uv3 (USB ID 0bda:c811) should be qualified for testing? And which kernel version I need to use to test? |
Hi @fakemanhk
We can certainly use your help testing but I need to give you a different link than the one above. You adapter uses the rtl8821cu chipset which uses a different driver than the one shown in the subject but it just so happens we already have an updated version of the driver for your chipset that needs testing also. The driver is located at: https://github.com/lwfinger/rtw88 Follow the instructions in the README. Below is the location of the issue where you should report bugs or suggestions:
5.4 or later. If using 6.2 or later, you will need to blacklist the driver that is already in the kernel. |
I found that I also have a few other Realtek based USB wifi at home (some are unknown brand), not sure if any testing needed for them?? I am listing the USB ID below:
|
Yes, indeed. ID 2357:011e /* TP Link */ That is a rtl8811au chipset. We could use help testing the other one when the time is right but for now we need to get the new driver in the kernel. Go back to the first message and follow the link to get info on the location of the driver and to see what is going right now. |
How is the testing on the adapters going? |
Are the 3 devices I list above using same rtw driver? I have installed it and when I plug one of the above 3 dongles it just works. |
8821au = wifi plus bluetooth and uses the *8821au driver |
Sorry forgot to reply this, previous laptop got some stupid problem which I wasn't able to test properly, just now I had an old Mac Mini which installed a fresh Peppermint Linux (with Debian backport 6.9.7) The system:
After installation and modprobe command, it didn't load until I unplug/plug back the dongle, below are info of
dmesg log after plugging in:
Testing with public WiFi, the speed isn't bad, I can get ~100Mbps on average Below are the iw output:
|
Another dongle: Also TP-Link, this time USB information:
After plug in, dmesg output:
Speedtest using same WiFi network, maybe due to a little bit larger form factor, I can see slight speed increase.
|
Here is the unknown 2x2 AC1200 USB WiFi I bought from China long time ago, also working with this driver??
dmesg output, though it says USB3.0 but I plugged into other Linux machine with USB3 port and found that it only binds to USB 2.0 ports:
The speed is a lot faster for this one, using the same WiFi network (I noticed that the WiFi network I am testing has only 40MHz channel width), I can get ~30% faster than previous dongle, occasionally getting close to 200Mbps speed test, I believe if 80MHz being used it should be significantly faster.
|
A couple of adapters have temp code to make the LEDs work. The plan is to do permanent code that works for all rtw88 supported adapters. It is coming. Thanks for the report on this.
Does this 8822bu adapter support bluetooth? If so, it will be locked to USB2 speeds. You can check the USB speed with the following command: $ lsusb -t |
Mine doesn't have bluetooth, but it's still running in USB2 mode:
|
I happened to take another look at the info you provided:
Note that the log is showing lines related to two drivers: rtw88_8822bu and rtw_8822bu rtw88_8822bu is the in-kernel driver so that seems to indicate that you did not blacklist it. Those are conflicting drivers. You can't have both going. |
Looks like it's compiled in, not module base so I wasn't able to blacklist it. |
No distro for public consumption should be doing that. What distro and kernel is this? If they really are compiling in rtw88_8822bu, then you need to have a serious conversation with the leaders of the distro.
The USB drivers in the rtw88 in the mainline kernel were not great. The worse was rtw88_8821cu. We started on it and by early this year we send 5 patches to mainline. This helped greatly with that driver. That was in kernel 6.7. Work has continued. Numerous patches have been and continue to go in. They need to be sent in in a specific order. Once all of the existing patches are in and the new rtw88_8821au and rtw88_8812au are in, you should see reasonable performance. We can't fix your distro so maybe switching distros is something to look at. |
Mine is a Peppermint Linux with Debian Back port 6.9.7 kernel.
Compiled in is just a wild guess, however the module name you mentioned
doesn't show up when I check with lsmod. Even my blacklist command can't
stop it.
…On Tue, Aug 6, 2024, 12:07 PM morrownr ***@***.***> wrote:
@fakemanhk <https://github.com/fakemanhk>
Looks like it's compiled in, not module base so I wasn't able to blacklist
it.
No distro for public consumption should be doing that. What distro and
kernel is this? If they really are compiling in rtw88_8822bu, then you need
to have a serious conversation with the leaders of the distro.
which implies the in kernel one is actually not functional?
The USB drivers in the rtw88 in the mainline kernel were not great. The
worse was rtw88_8821cu. We started on it and by early this year we send 5
patches to mainline. This helped greatly with that driver. That was in
kernel 6.7. Work has continued. Numerous patches have been and continue to
go in. They need to be sent in in a specific order. Once all of the
existing patches are in and the new rtw88_8821au and rtw88_8812au are in,
you should see reasonable performance.
We can't fix your distro so maybe switching distros is something to look
at.
—
Reply to this email directly, view it on GitHub
<#430 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AFCR7WPNR6EG2TKEECPG5D3ZQA4Y7AVCNFSM6AAAAABHJXSOO2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDENZQGI4TOMZVGY>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
How are you doing the blacklist? |
mini PC with RTL8821AElspci -d 10ec:8821 -vv Card work extremely slow. |
I made a mistake in the title of this issue. I corrected it today. rtl8821ae should have read rtl8812au. While there is an in-kernel driver for the rtl8821ae, it is not part of this project that is designed to go into rtw88. Maybe someday the rtl8821ae support can be moved to rtw88 and improved. |
Thanks. |
New in-kernel (mac80211) driver for rtl8821au, rtl8811au and rtl8821ae chipsets.
Information about testing is located at:
morrownr/8821au-20210708#133
That is Issue 133 in the 8821au repo here at this site. That issue will be used for discussion and reporting. The first message shows the location of the driver to be tested. The subject chipsets have been popular with Linux users for several years.
The intent is to upstream the driver to the Linux kernel once it is ready. There is nothing like Plug and Play adapters that work well. Please help if you can. I have been testing managed and monitor mode today and the results are good. We need to test rigorously.
HOT: All devs currently only have adapters with rtl8811au chips. We need Linux users with rtl8821au and rtl8821ce chips.
@morrownr
The text was updated successfully, but these errors were encountered: