Testers: Please check in here to let me know you are here...

[morrownr]

I would appreciate it if you would post a message here so that I know you successfully made it to the repo. This issue is for you to check in and for discussion. If you find a bug that needs to be fixed, please post a new issue.

Welcome. I hope we can complete and release this driver to the public by the end of the year. We are in no hurry but I do need regular reports on what you are finding. Thanks and enjoy.

@morrownr

[pesa1234]

Thanks for the invitation link.
In the next days I'll try to compile it for ubuntu for my tplink TXE70UH.

Then I'll try to check on openwrt.

Thanks again

[morrownr]

Then I'll try to check on openwrt.

Ouch. Have you ever tried to compile a Realtek driver for OpenWRT?

[pablo-cabrera]

Thanks for the invitation link and for your work in general. Much appreciated!
I can test it on a NanoPi R4S and a NanoPi NEO LTS

[morrownr]

Testers: I created a new issue today for AP mode. It is issue 8. I think we need to keep AP stuff consolidated.

[Digital-Daz]

Thank you, I have just installed the driver.

[fabiano-cancela]

Hi!
I've just installed the driver. No changes at all to the settings.
Using a cheap AX300 antenna from AliExpress in Tumbleweed, 6.11.5-1-default.
Device is ID 0bda:c832 Realtek Semiconductor Corp. 802.11ax WLAN Adapter.

Thank you!

[morrownr]

@fabiano-cancela @PBXForums

Thanks for letting me know.

It is easy to upgrade this driver and I will be merging things on a regular basis. Here is all it takes to upgrade:

From a terminal:

$ git pull
$ sudo sh install-driver.sh

The install-driver.sh script automatically uninstalls previous installations for you and then goes about compiling the new source and installing the new driver. Recommend you upgrade once per week until we decide it is time to take the repo public. So far things look much better than I expected but I still think that the end of the year is a good time to shoot for as far as going public is concerned.

I need everyone to mention the following when you make reports:

CPU (arm64, amd64, x86_64, etc.)

Distro and kernel version

Mode that you are testing (Managed (client), AP, P2P...)

Remember to report anything that needs to be improved in the docs: README.md, FAQ.md and 8852cu.conf.

Thanks for all you are doing.

@morrownr

[lilkidsuave]

Im here

[wxlfsangel]

Thank you very much for having me here.

[morrownr]

Does anyone have a system with kernel 6.12 that they can test with temporarily? I just need to see if there are compilation problems and I am not ready to test 6.12 here yet.

[lilkidsuave]

Does anyone have a system with kernel 6.12 that they can test with temporarily? I just need to see if there are compilation problems and I am not ready to test 6.12 here yet.

6.12 isnt in my arch repos rn. Newest I have is 6.11.6. imma looks for a 6.12 RC

[lilkidsuave]

update: compileing 6.12 rc6

[lilkidsuave]

update finally did it. compiling took forever and had to read up how to do it manually again

[lilkidsuave]

gcc matches 14.1.1 on both kernel and gcc version

[lilkidsuave]

Does anyone have a system with kernel 6.12 that they can test with temporarily? I just need to see if there are compilation problems and I am not ready to test 6.12 here yet.

@morrownir
6GHZ works great with the driver on 6.12rc6. 1100mbps down and 800mbps up using openspeedtest which is also hosted on the same nuc.
For context, my 5GHZ router does 1000mbps on both up and down on openspeedtest

[lilkidsuave]

My unit gets pretty toasty when under load, has anyone else's done the same

[morrownr]

@lilkidsuave

Your help is so valuable and saved me time to work on other things. My lab here is organized to maximize productivity as I do not have unlimited resources. How I normally test newer kernels is to use a system Ubuntu and the Ubuntu Mainline kernel repo. However, that repo has been broken since kernel 6.11 was released so me taking time from my schedule to do a manual compilation was not in my schedule right now. I will change the README to reflect kernel 6.12 support.

6GHZ works great with the driver on 6.12rc6. 1100mbps down and 800mbps...

I've been surprised at how solid AP mode is with this driver... on all bands. I finished my AP mode testing yesterday and am moving on to P2P testing.

My unit gets pretty toasty when under load, has anyone else's done the same...

I will watch for this with my adapter but my adapter that uses this driver is not an Edup adapter. I can test power usage so that we have an idea what kind of power this chip is pulling. My testing over the years has shown that Realtek chips are usually more power hungry than Mediatek. I'll post some results... probably in a new issue so we can keep other replies in one place,

I do have an Edup Ep-1672 that appears to have exactly the same case and antennas as the one you have so I will throw it into the power consumption test... it has the mt7921au chip.

[clyvari]

Thanks for the invite! I'll test a default install in Debian stable tonight or tomorrow in a VM, then i'll move into other testing as needed

[morrownr]

I made a mistake in the top message when I wrote it. It said I do not need regular reports... well, that was wrong. In fact I do want regular reports. If you find a bug, start a new issue. Same thing applies when you are seeing success, please report good results along with the details of your CPU, distro and what you are testing. It is important for testing that good results are known so that we know where to concentrate work.

Please concentrate in installation and removal of the driver with the instructions in the README and remember, regular patches are going into this repo so updating is necessary:

$ git pull
$ sudo sh install-driver.sh

Remember that install-driver.sh detects and removes currently installed versions of the driver so you do not need to remove the driver to update it.

@clyvari

Glad to have you here. Looking forward to your reports.

[clyvari]

@morrownr
Checking-in with my report

Test setup:

• Default Debian 12 install (Linux debian 6.1.0-26-amd64 #1 SMP PREEMPT_DYNAMIC Debian 6.1.112-1 (2024-09-30) x86_64 GNU/Linux) with Gnome desktop
• Machine: VMWare Workstation VM w/ 3d accelerated graphics, 1CPU w/ 8 cores, 16GB RAM, No secure boot. Windows Host.
• Device (TXE70UH) set to auto-connect to the VM

Device:

• TP-Link AXE5400 (Archer TXE70UH) - EU version.
• Windows drivers are for a rtl8852cu chipset.
• 2-state device: 1st connects as mass storage, then after eject reconnects as a WLAN device
• Mass storage VID:PID => 0BDA:1A2B
• 802.11ax WLAN Adapter VID:PID => 35BC:0102
• Even without the driver installed and without a custom usb_modeswitch rule, the mass storage device appears to be auto-ejected as soon as connected, so as to end up in the WLAN mode.

Compiling the driver:

After installing the required dependencies listed in the README, everything compiled flawlessly.
Successively running install-driver.sh worked great as well, modified code is correctly compiled and installed.

Making the driver work for the TXE70UH

• These specific changes made the device work:
  • As mentioned in Works with tp-link TXE70UH (8852cu) lwfinger/rtw8852cu#17, modified os_dep/linux/usb_intf.c to add the following:

    --- a/os_dep/linux/usb_intf.c
    +++ b/os_dep/linux/usb_intf.c
    @@ -170,6 +170,8 @@ static struct usb_device_id rtw_usb_id_tbl[] = {
            {USB_DEVICE_AND_INTERFACE_INFO(USB_VENDER_ID_REALTEK, 0xC832, 0xff, 0xff, 0xff), .driver_info = RTL8852C},
            {USB_DEVICE_AND_INTERFACE_INFO(USB_VENDER_ID_REALTEK, 0xC85A, 0xff, 0xff, 0xff), .driver_info = RTL8852C},
            {USB_DEVICE_AND_INTERFACE_INFO(USB_VENDER_ID_REALTEK, 0xC85D, 0xff, 0xff, 0xff), .driver_info = RTL8852C},
    +       {USB_DEVICE_AND_INTERFACE_INFO(0x35b2, 0x0502, 0xff, 0xff, 0xff), .driver_info = RTL8852C}, /* TP-Link AXE5400(RTW8832CU) */
    +       {USB_DEVICE_AND_INTERFACE_INFO(0x35bc, 0x0102, 0xff, 0xff, 0xff), .driver_info = RTL8852C}, /* TP-Link AXE5400(RTW8852CU) */
     #endif /* CONFIG_RTL8852C */
     
    {}      /* Terminating entry */
    

  • When prompted during install, changed the configuration to set the region to FR:

    - options 8852cu rtw_country_code=US rtw_switch_usb_mode=0 rtw_he_enable=1 rtw_vht_enable=1 rtw_dfs_region_domain=0
    + options 8852cu rtw_country_code=FR rtw_switch_usb_mode=0 rtw_he_enable=1 rtw_vht_enable=1 rtw_dfs_region_domain=0
    

• Unsure wich of these changes are strictly necessary for making the TXE70UH work
• After installing the driver with those changes and after a reboot, the driver correctly pick up the device and a wifi option is enabled in the desktop. iw dev lists the device as well.

Actually using the device

This is were things went not too well. To sum up my experience, all wifi operations are reeally slow. More specifically:

• System hang: when booting the machine with the device plugged-in or when plugin the device some operations seem to hang:
  • When booting, and logging into my user session, there's a ~10 sec hang before the desktop appears. The hang doesn't happen if the device isn't connected.
  • When pluggin the device and running dmesg for instance, it also hangs for ~10 sec
  • Unsure of the cause, but might be caused by slow wifi operations, see below
• "Network scanning" (listing and refreshing the list of available SSIDs) is really slow
• Connecting to an AP with a valid password is really slow as well (~10-30 secs). Initial connection fails 50% of the time, and after a bit of retrying, it does connect (and stays connected)
• Wifi download speed is also about 2-4x slower than connecting the device to my windows machine and sharing this connection with the VM.
  • My WiFi is spotty, and the particular AP i'm using for this test is only 802.11 n, the expected bandwith for this setup is ~35Mbps. However, in my tests i'm only able to get about 12Mbps
• There's a lot of dmesg spamming from the driver, but that's maybe intended since it's in development. Especially when scanning SSIDs / connecting, it calms down a lot after that.

On a more positive note:

• Cpu usage during operations seem non-existant, which is good
• Plugin / Unplugging, rebooting, ... all works fine

Next steps

I'm really curious about these slow operations. Unless somebody already as an idea of the cause, i'll try to run experiements on baremetal, to eliminate the possibility that it's caused by the VM setup. I'll also ditch the desktop environment to check if it might be caused by whatever connection manager Gnome is using.

I hope that a detailed post like that is what was expected and is approriate for this particular issue. I don't feel confident yet to open an issue on the slow operations I encountered, I'll experiment a bit more. Also let's see first if somebody can replicate my result (or have better results than me !) with a TXE70UH.

[morrownr]

@clyvari

Very good report.

•   {USB_DEVICE_AND_INTERFACE_INFO(0x35b2, 0x0502, 0xff, 0xff, 0xff), .driver_info = RTL8852C}, /* TP-Link AXE5400(RTW8832CU) */
  

•   {USB_DEVICE_AND_INTERFACE_INFO(0x35bc, 0x0102, 0xff, 0xff, 0xff), .driver_info = RTL8852C}, /* TP-Link AXE5400(RTW8852CU) */
  

Help me out. You have 2 adapters? Right? So I need to add both VID/PIDs to the driver?

Unsure wich of these changes are strictly necessary for making the TXE70UH work

I will look at clarifying the documentation in 8852cu.conf.

I'm really curious about these slow operations.

Can I get you to copy this report into a new Issue with a title like Slow operations so that we can look specifically at that issue

Even without the driver installed and without a custom usb_modeswitch rule, the mass storage device appears to be auto-ejected as soon as connected, so as to end up in the WLAN mode.

Are both of your adapter multi-state? The adapter I was sent by a maker is multi-state as well and I have seen problems on 3 machines so far. On one machine it works okay. Another tester had a multi-state adapter that we simply could not get to work right and he is sending it back to get another adapter. My limited opinion at this point is that there is something a little wrong with the storage setup in these rtl8852/32cu based adapters and I do not fully understand what is yet. I have a couple of workarounds but I do not know what the real problem is yet. Please test on all Linux machines that you have and I need everyone to test on all Linux machines they can so as to figure out the scope of this problem. This is not a driver problem.

It will be interesting to see the results of your bare metal test.

Thanks.

[morrownr]

General Broadcast to all testers:

For those of you with adapters that are multi-state (windows driver onboard), please test on all Linux machines that you have access to. I am seeing multi-state problems with my adapter on multiple systems. I have been able to work around the issue on 2 systems but one system is a mystery at this point. One of my systems shows no problem. One tester has now sent his adapter back to the seller in order to get a different adapter. We need to know the extent of this problem and hopefully we can figure out what is causing it.

I have two other adapters from the same maker that are multi-state and I have no problems with them. They are based on the rtl8852/32bu chip.

If you are unsure how to turn off usb_modeswitch to see if your adapter is multi-state, ask me and I will post the simple instructions.

[pablo-cabrera]

I noticed that on a cold boot, an already plugged adapter doesn't switch, even when I try to call usb_modeswitch manually (even forcing the usb reset).
When I unplug and replug it, it switches normally. When I reboot the machine with an already switched adapter, it stays switched after reboot.
I wasn't able to fix the cold boot switch problem so far.

[morrownr]

@pablo-cabrera

adapter doesn't switch

I just started a new issue (10) to capture this issue. You are third person here that is seeing problems similar to this. I have struggled with it. I have a workaround that helped 2 systems that I tested it on but 1 system is just such that I can't get it to work. Please post your follow up messages there and maybe by comparing notes we can figure something out.

[clyvari]

@morrownr

Help me out. You have 2 adapters? Right? So I need to add both VID/PIDs to the driver?

No actually I have only 1 device. It was just incorrect testing on my part. I only have one, with VID:PID = 35BC:0102 (the second line).
Somehow, I thought that I needed both lines to make it work, even tough one did not match my VID:PID.
I re-tested it today, and I can confirm I only need the line for 35BC:0102 to make it work. So this one can go in the driver.
I have no idea where the other VID:PID comes from, so maybe it should not be added right away.

Can I get you to copy this report into a new Issue with a title like Slow operations so that we can look specifically at that issue

I will

Are both of your adapter multi-state?

So only one adapter as indicated above, and yes it's multi-state. Debian (?) seems to eject it right await, and it ends up in the WLan mode.
The dmesg log is quite terse about it, but from a clean Debian 12 without the driver being ever installed, here is the log:

usb 1-1: new high-speed USB device number 2 using ehci-pci
usb 1-1: New USB device found, idVendor=0bda, idProduct=1a2b, bcdDevice= 0.00
usb 1-1: New USB device strings: Mfr=1, Product=2, SerialNumber=0
usb 1-1: Product: DISK
usb 1-1: Manufacturer: Realtek
usb-storage 1-1:1.0: USB Mass Storage device detected
scsi host33: usb-storage 1-1:1.0
usbcore: registered new interface driver usb-storage
usbcore: registered new interface driver uas
usb 1-1: USB disconnect, device number 2    <---------------------- this happens without any input on my part
usb 1-1: new high-speed USB device number 3 using ehci-pci
usb 1-1: New USB device found, idVendor=35bc, idProduct=0102, bcdDevice= 0.00
usb 1-1: New USB device strings: Mfr=1, Product=2, SerialNumber=3
usb 1-1: Product: 802.11ax WLAN Adapter
usb 1-1: Manufacturer: Realtek
usb 1-1: SerialNumber: 00e04c000001

I have another device, a laptop I can test it with.
Maybe it's a distro issue ? I can test with other distro as well.

It will be interesting to see the results of your bare metal test.

I will conduct some baremetal test this weekend.

[morrownr]

@clyvari

No actually I have only 1 device. It was just incorrect testing on my part. I only have one, with VID:PID = 35BC:0102 (the second line).

Then that is what I will add. Can I get you to post the results of:

$ lsusb

Also, can I get you to give me the real name of the adapter? AXE5400 is just the class of adapter.

[clyvari]

@morrownr

$ lsusb

Bus 001 Device 003: ID 35bc:0102 Realtek 802.11ax WLAN Adapter

Also, can I get you to give me the real name of the adapter? AXE5400 is just the class of adapter.

Complete Model name as indicated on device is: Archer TXE70UH(EU) Ver:1.0

[morrownr]

@clyvari

Patch to include the VID/PID for your Archer TXE70UH(EU) Ver:1.0 went in this morning. Please test.

[MassonY]

Hi !
Thank you for your work and for the access to the repo.

Test hardware
I installed the driver on my home computer: Nobara 6.11.7-201.fsync.fc40.x8 gcc 14.2.1. (gcc version is the exact same)
I am testing the tp-link AX3000 - Archer TX50UH
lsusb output :
Bus 001 Device 006: ID 35bc:0101 Realtek 802.11ax WLAN Adapter

Making the driver 'work'

The driver as is doesnt detect the ID and nothing works so i applied the fix found here: lwfinger/rtw8852cu#20 .

Result
I recompiled it using the same script and it does detect the usb dongle as a wifi card but it cannot connect to any network due to timeout. It does apply to networks with and without password protection.

journalctl -S today | grep wlp3s0f0u1 | grep ERROR :

nov. 13 10:29:09 nobara-pc kernel: RTW: ERROR _netdev_open(wlp3s0f0u1) Failed!! (netif_up=0)
nov. 13 10:44:22 nobara-pc kernel: RTW: ERROR _connect_msg_hdlr(wlp3s0f0u1): assoc_AP_vendor gone!
nov. 13 10:44:37 nobara-pc kernel: RTW: ERROR _connect_msg_hdlr(wlp3s0f0u1): assoc_AP_vendor gone!

I do reproduce the usb switching problem mentionned by @pablo-cabrera

I also reproduce the System hang problem of @clyvari

[morrownr]

Hi @MassonY

Thank you for your work and for the access to the repo.

Glad to have you here. There are about 10 of us right now. My flexible goal is to have the driver ready to go public by the end of the year. That is flexible.

The driver as is doesn't detect the ID and nothing works...

I understand and have now added the VID/PID for your adapter. Since you have edited your copy of the driver, it might be easier for you to delete the driver folder and clone it again. In the future, if there are no edits, you can update to keep up with code changes by running:

$ git pull
$ sudo sh install-driver.sh

Please do test my patch for your adapter and let me know the results.

I recompiled it using the same script and it does detect the usb dongle as a wifi card but it cannot connect to any network due to timeout. It does apply to networks with and without password protection.

We need to work this problem. Can you open a new issue for this problem?

I do reproduce the usb switching problem mentionned by @pablo-cabrera

That makes 4 or 5 of us out of 10 with multi-state (storage switching problems) Issue #10 is about this problem so if you could go to that issue and provide details about the problem, that would be great. This issue seems to happen with adapters from various makers. I have experienced this problem with 2 of 3 systems I have tested my adapter on. On one system I used a usb quirks workaround but on one system, nothing would fix it and the adapter will simply not work on the system. This is bad and it is not a driver problem.

I also reproduce the System hang problem of @clyvari

I looked back over the message from @clyvari but am not sure what you are talking about. Can you be specific?

Cheers

[MassonY]

I also reproduce the System hang problem of @clyvari

I looked back over the message from @clyvari but am not sure what you are talking about. Can you be specific?

I am refering to this issue mentionned by @clyvari

System hang: when booting the machine with the device plugged-in or when plugin the device some operations seem to hang:

• When booting, and logging into my user session, there's a ~10 sec hang before the desktop appears. The hang doesn't happen if the device isn't connected.
• When pluggin the device and running dmesg for instance, it also hangs for ~10 sec

But this is a weird issue, it used to hang the first time I installed the driver but it seems to be fine now, I cannot reproduce.

[morrownr]

@MassonY

I added a new issue, #13 , to work the 10 second system hang issue.

We seem to have 3 issue that may require a lot of work to determine what the cause is:

#13
#10
#4

We just need to stay at it. We can do this.

[morrownr]

I updated the Development Progress Log ( #1 ) today.

I think we are making good progress but I am particularly concerned about the multi-state (Windows driver onboard) issue that some of us, including me, have experienced. I have experienced the problem on multiple computers and given that others here have experienced the problem as well with adapters from various adapter makers, I think I can say confidently that the problem is not specific to any particular adapter maker but is a general problem.

Is it a problem with usb-modeswitch? Or is it a problem with something in eeprom? I do not know yet. Any help that any of you can provide is welcome. If this issue is not solved in a suitable manner, then this chipset and driver is of very limited value to Linux users.

I am very aware that many users around the world buy TP-Link adapters because that brand is all that is readily available to them via limited sources. In that light, I posted information about a new adapter from TP-Link recently:

morrownr/USB-WiFi#534

The TXE50UH is a new adapter that appears to be single-state (no hassles with this storage mode crap) and it uses a Mediatek driver, mt7921u, so it is plug and play. Think about it... no more chasing around after Realtek out-of-kernel drivers.

That is all for this update. I'll be waiting on your reports,

@morrownr

[smoke]

I am having EDUP WiFi 6E AX5400 Wireless USB Adapter EP-AX1671 (NOT EP-AX1671S version) with Realtek RTL8832CU chipset,
bought from https://www.aliexpress.com/item/1005005708499877.html?spm=a2g0o.order_list.order_list_main.17.2bbf1802bUEm30#nav-specification

some product pictures

  

I have just pulled, compiled and loaded the module and everything works more than fine for start. I have only changed rtw_switch_usb_mode=1 to use USB 3.0 (not tested other mode) as this is what the device is intented to use. I have connected on 5Ghz and 2.4Ghz just fine, I don't have WiFi 6.

No packets dropped or anything noticable at a glance

smoke@rkirilov-work-pc ~/projects/open-source/morrownr $ mtr -rwn -c 50 gentoo.org
Start: 2024-11-21T11:44:09+0200


HOST: rkirilov-work-pc Loss%   Snt   Last   Avg  Best  Wrst StDev
  1.|-- 192.168.21.1      0.0%    50    0.8   0.8   0.7   1.8   0.2
  2.|-- 78.130.196.1      0.0%    50   21.1  19.4   4.1 114.9  14.7
  3.|-- 212.116.148.65    0.0%    50   17.5   1.6   0.9  17.5   2.4
  4.|-- 212.116.147.41    0.0%    50    3.6   4.2   3.3  17.6   2.2
  5.|-- 212.116.147.42    0.0%    50    3.6   4.2   3.5  13.6   1.8
  6.|-- 212.116.133.129   0.0%    50    3.8   4.4   3.7  16.6   2.1
  7.|-- 87.245.231.176    0.0%    50    4.3   5.8   3.7  26.0   5.1
  8.|-- 87.245.232.139    8.0%    50   28.0  28.1  27.8  29.6   0.3
  9.|-- 87.245.236.129    0.0%    50   28.1  28.1  27.8  29.3   0.3
 10.|-- 151.101.129.91    0.0%    50   28.0  27.9  27.7  29.1   0.2

Speedtest by Ookla

I will use it through the days to come and will report accordingly.

Thank you @morrownr !!!

Here some more details of my setup:

smoke@rkirilov-work-pc ~/projects/open-source/morrownr/rtl8852cu-20240510 $ cat /etc/modprobe.d/8852cu.conf | grep '^options'
options 8852cu rtw_country_code=BG rtw_switch_usb_mode=1

smoke@rkirilov-work-pc ~/projects/open-source/morrownr/rtl8852cu-20240510 $ sudo modinfo 8852cu
filename:       /lib/modules/6.6.58-gentoo-dist/updates/dkms/8852cu.ko
version:        v1.19.2.1-0-gfc5aab822.20240510
author:         Realtek Semiconductor Corp.
description:    Realtek Wireless Lan Driver
license:        GPL
srcversion:     E3C8E21E79453A1ED268AEB
alias:          usb:v35BCp0102d*dc*dsc*dp*icFFiscFFipFFin*
alias:          usb:v35BCp0101d*dc*dsc*dp*icFFiscFFipFFin*
alias:          usb:v0BDApC85Dd*dc*dsc*dp*icFFiscFFipFFin*
alias:          usb:v0BDApC85Ad*dc*dsc*dp*icFFiscFFipFFin*
alias:          usb:v0BDApC832d*dc*dsc*dp*icFFiscFFipFFin*
depends:        cfg80211
retpoline:      Y
name:           8852cu
vermagic:       6.6.58-gentoo-dist SMP preempt mod_unload 
sig_id:         PKCS#7
signer:         
sig_key:        
sig_hashalgo:   unknown
signature:      
parm:           rtw_wireless_mode:int
parm:           rtw_band_type:int
parm:           rtw_max_bss_cnt:int
parm:           rtw_usb_rxagg_mode:int
parm:           rtw_dynamic_agg_enable:int
parm:           rtw_drv_log_level:set log level when insert driver module, default log level is _DRV_INFO_ = 4 (uint)
parm:           rtw_tx_bw_mode:The max tx bw for 2G/5G/6G. format is the same as rtw_bw_mode (uint)
parm:           rtw_max_amsdu_len:uint
parm:           rtw_tx_ampdu_num:int
parm:           rtw_rx_ampdu_sz_limit_1ss:RX AMPDU size limit for 1SS link of each BW, 0xFF: no limitation (array of uint)
parm:           rtw_rx_ampdu_sz_limit_2ss:RX AMPDU size limit for 2SS link of each BW, 0xFF: no limitation (array of uint)
parm:           rtw_rx_ampdu_sz_limit_3ss:RX AMPDU size limit for 3SS link of each BW, 0xFF: no limitation (array of uint)
parm:           rtw_rx_ampdu_sz_limit_4ss:RX AMPDU size limit for 4SS link of each BW, 0xFF: no limitation (array of uint)
parm:           rtw_vht_enable:int
parm:           rtw_vht_24g_enable:int
parm:           rtw_vht_rx_mcs_map:VHT RX MCS map (uint)
parm:           rtw_he_enable:int
parm:           rtw_rf_path:int
parm:           rtw_tx_nss:int
parm:           rtw_rx_nss:int
parm:           rtw_init_regd_always_apply:Whether INIT regd request is always applied (being included when taking intersection together with higher priority requests) when regd source is RTK_PRIV (uint)
parm:           rtw_user_regd_always_apply:Whether USER regd request is always applied (being included when taking intersection together with higher priority requests) when regd source is RTK_PRIV (uint)
parm:           rtw_country_code:The default country code (in alpha2) (charp)
parm:           rtw_channel_plan:The default chplan ID when rtw_alpha2 is not specified or valid (uint)
parm:           rtw_excl_chs:Exclusive channel list of 2G and 5G band (array of uint)
parm:           rtw_channel_plan_6g:The default chplan_6g ID when rtw_alpha2 is not specified or valid (uint)
parm:           rtw_excl_chs_6g:Exclusive channel list of 6G band (array of uint)
parm:           rtw_dis_ch_flags:The flags with which channel is to be disabled (charp)
parm:           rtw_bcn_hint_valid_ms:The length of time beacon hint continue (uint)
parm:           rtw_country_ie_slave_en_mode:802.11d country IE slave enable mode: 0: disable, 1: enable, 2: enable when INIT/USER set world wide mode (uint)
parm:           rtw_country_ie_slave_flags:802.11d country IE slave flags: BIT0: take intersection when having multiple received IEs, otherwise choose effected one from received IEs, BIT1: consider all environment BSSs, otherwise associated BSSs only (uint)
parm:           rtw_country_ie_slave_en_role:802.11d country IE slave enable role: BIT0:pure STA mode, BIT1:P2P group client (uint)
parm:           rtw_country_ie_slave_en_ifbmp:802.11d country IE slave enable iface bitmap (uint)
parm:           rtw_country_ie_slave_scan_int_ms:802.11d country IE slave auto scan interval in ms to find environment BSSs. 0: no environment BSS auto scan triggered by driver self (uint)
parm:           rtw_btcoex_enable:BT co-existence on/off, 0:off, 1:on, 2:by efuse (int)
parm:           rtw_ant_num:Antenna number setting, 0:by efuse (int)
parm:           rtw_pci_dynamic_aspm_linkctrl:int
parm:           rtw_qos_opt_enable:int
parm:           ifname:The default name to allocate for first interface (charp)
parm:           if2name:The default name to allocate for second interface (charp)
parm:           rtw_vo_edca:uint
parm:           rtw_ap_src_b2u_flags:int
parm:           rtw_ap_fwd_b2u_flags:int
parm:           rtw_wowlan_sta_mix_mode:int
parm:           rtw_pwrtrim_enable:int
parm:           rtw_initmac:charp
parm:           rtw_chip_version:int
parm:           rtw_rfintfs:int
parm:           rtw_lbkmode:int
parm:           rtw_network_mode:int
parm:           rtw_band:int
parm:           rtw_channel:int
parm:           rtw_mp_mode:int
parm:           rtw_wmm_enable:int
parm:           rtw_vrtl_carrier_sense:int
parm:           rtw_vcs_type:int
parm:           rtw_hw_rts_en:int
parm:           rtw_busy_thresh:int
parm:           rtw_ht_enable:int
parm:           rtw_bw_mode:int
parm:           rtw_ampdu_enable:int
parm:           rtw_stbc_cap:int
parm:           rtw_rx_ampdu_amsdu:int
parm:           rtw_tx_ampdu_amsdu:int
parm:           rtw_quick_addba_req:int
parm:           rtw_beamform_cap:int
parm:           rtw_sw_proto_bf_cap_phy0:int
parm:           rtw_sw_proto_bf_cap_phy1:int
parm:           rtw_dyn_txbf:int
parm:           rtw_lowrate_two_xmit:int
parm:           rtw_low_power:int
parm:           rtw_wifi_spec:int
parm:           rtw_full_ch_in_p2p_handshake:int
parm:           rtw_antdiv_cfg:int
parm:           rtw_antdiv_type:int
parm:           rtw_drv_ant_band_switch:int
parm:           rtw_single_ant_path:int
parm:           rtw_switch_usb_mode:int
parm:           rtw_hwpwrp_detect:int
parm:           rtw_hw_wps_pbc:int
parm:           rtw_check_hw_status:int
parm:           rtw_max_roaming_times:The max roaming times to try (uint)
parm:           rtw_hiq_filter:0:allow all, 1:allow special, 2:deny all (uint)
parm:           rtw_adaptivity_en:0:disable, 1:enable, 2:auto (uint)
parm:           rtw_adaptivity_mode:0:normal, 1:carrier sense (uint)
parm:           rtw_adaptivity_th_l2h_ini:th_l2h_ini for Adaptivity (int)
parm:           rtw_adaptivity_th_edcca_hl_diff:th_edcca_hl_diff for Adaptivity (int)
parm:           rtw_adaptivity_idle_probability:rtw_adaptivity_idle_probability (int)
parm:           rtw_dfs_region_domain:0:NONE, 1:FCC, 2:MKK, 3:ETSI, 4:KCC (uint)
parm:           rtw_amsdu_mode:0:non-spp, 1:spp, 2:all drop (uint)
parm:           rtw_amplifier_type_2g:BIT3:2G ext-PA, BIT4:2G ext-LNA (uint)
parm:           rtw_amplifier_type_5g:BIT6:5G ext-PA, BIT7:5G ext-LNA (uint)
parm:           rtw_rfe_type:default init value:64 (uint)
parm:           rtw_powertracking_type:default init value:64 (uint)
parm:           rtw_GLNA_type:default init value:0 (uint)
parm:           rtw_TxBBSwing_2G:default init value:0xFF (uint)
parm:           rtw_TxBBSwing_5G:default init value:0xFF (uint)
parm:           rtw_OffEfuseMask:default open Efuse Mask value:0 (uint)
parm:           rtw_FileMaskEfuse:default drv Mask Efuse value:0 (uint)
parm:           rtw_rxgain_offset_2g:default RF Gain 2G Offset value:0 (uint)
parm:           rtw_rxgain_offset_5gl:default RF Gain 5GL Offset value:0 (uint)
parm:           rtw_rxgain_offset_5gh:uint
parm:           rtw_rxgain_offset_5gm:default RF Gain 5GM Offset value:0 (uint)
parm:           rtw_pll_ref_clk_sel:force pll_ref_clk_sel, 0xF:use autoload value (uint)
parm:           rtw_tx_pwr_by_rate:0:Disable, 1:Enable, 2: Depend on efuse (int)
parm:           rtw_tx_pwr_lmt_enable:0:Disable, 1:Enable, 2: Depend on efuse (int)
parm:           rtw_target_tx_pwr_2g_a:2.4G target tx power (unit:dBm) of RF path A for each rate section, should match the real calibrate power, -1: undefined (array of int)
parm:           rtw_target_tx_pwr_2g_b:2.4G target tx power (unit:dBm) of RF path B for each rate section, should match the real calibrate power, -1: undefined (array of int)
parm:           rtw_target_tx_pwr_2g_c:2.4G target tx power (unit:dBm) of RF path C for each rate section, should match the real calibrate power, -1: undefined (array of int)
parm:           rtw_target_tx_pwr_2g_d:2.4G target tx power (unit:dBm) of RF path D for each rate section, should match the real calibrate power, -1: undefined (array of int)
parm:           rtw_target_tx_pwr_5g_a:5G target tx power (unit:dBm) of RF path A for each rate section, should match the real calibrate power, -1: undefined (array of int)
parm:           rtw_target_tx_pwr_5g_b:5G target tx power (unit:dBm) of RF path B for each rate section, should match the real calibrate power, -1: undefined (array of int)
parm:           rtw_target_tx_pwr_5g_c:5G target tx power (unit:dBm) of RF path C for each rate section, should match the real calibrate power, -1: undefined (array of int)
parm:           rtw_target_tx_pwr_5g_d:5G target tx power (unit:dBm) of RF path D for each rate section, should match the real calibrate power, -1: undefined (array of int)
parm:           rtw_phy_file_path:The path of phy parameter (charp)
parm:           rtw_load_phy_file:PHY File Bit Map (int)
parm:           rtw_decrypt_phy_file:Enable Decrypt PHY File (int)
parm:           rtw_phydm_ability:uint
parm:           rtw_halrf_ability:uint
parm:           rtw_rfk_ability:uint
parm:           rtw_edcca_th_2g:uint
parm:           rtw_edcca_th_5g:uint
parm:           rtw_edcca_cs_th:uint
parm:           rtw_en_napi:int
parm:           rtw_en_gro:int
parm:           rtw_iqk_fw_offload:int
parm:           rtw_ch_switch_offload:int
parm:           rtw_en_dyn_rrsr:int
parm:           rtw_rrsr_value:int
parm:           rtw_scan_interval_thr:Threshold used to judge if scan request comes from scan UI, unit is ms. (uint)
parm:           rtw_roch_min_home_dur:uint
parm:           rtw_roch_max_away_dur:uint
parm:           rtw_roch_extend_dur:uint
parm:           rtw_en_ecsa:uint
parm:           rtw_quota_turbo_en:int
parm:           rtw_scan_pch_ex:uint

smoke@rkirilov-work-pc ~/projects/open-source/morrownr/rtl8852cu-20240510 $ sudo lsusb -v -d 0bda:c832

Bus 002 Device 005: ID 0bda:c832 Realtek Semiconductor Corp. 802.11ax WLAN Adapter
Device Descriptor:
  bLength                18
  bDescriptorType         1
  bcdUSB               3.00
  bDeviceClass            0 [unknown]
  bDeviceSubClass         0 [unknown]
  bDeviceProtocol         0 
  bMaxPacketSize0         9
  idVendor           0x0bda Realtek Semiconductor Corp.
  idProduct          0xc832 802.11ax WLAN Adapter
  bcdDevice            0.00
  iManufacturer           1 Realtek
  iProduct                2 802.11ax WLAN Adapter
  iSerial                 3 00e04c000001
  bNumConfigurations      1
  Configuration Descriptor:
    bLength                 9
    bDescriptorType         2
    wTotalLength       0x007a
    bNumInterfaces          1
    bConfigurationValue     1
    iConfiguration          0 
    bmAttributes         0xe0
      Self Powered
      Remote Wakeup
    MaxPower              896mA
    Interface Descriptor:
      bLength                 9
      bDescriptorType         4
      bInterfaceNumber        0
      bAlternateSetting       0
      bNumEndpoints           8
      bInterfaceClass       255 Vendor Specific Class
      bInterfaceSubClass    255 Vendor Specific Subclass
      bInterfaceProtocol    255 Vendor Specific Protocol
      iInterface              2 802.11ax WLAN Adapter
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x84  EP 4 IN
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0400  1x 1024 bytes
        bInterval               0
        bMaxBurst               3
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x05  EP 5 OUT
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0400  1x 1024 bytes
        bInterval               0
        bMaxBurst               3
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x06  EP 6 OUT
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0400  1x 1024 bytes
        bInterval               0
        bMaxBurst               3
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x07  EP 7 OUT
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0400  1x 1024 bytes
        bInterval               0
        bMaxBurst               3
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x09  EP 9 OUT
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0400  1x 1024 bytes
        bInterval               0
        bMaxBurst               3
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x0a  EP 10 OUT
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0400  1x 1024 bytes
        bInterval               0
        bMaxBurst               3
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x0b  EP 11 OUT
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0400  1x 1024 bytes
        bInterval               0
        bMaxBurst               3
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x0c  EP 12 OUT
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0400  1x 1024 bytes
        bInterval               0
        bMaxBurst               3
Binary Object Store Descriptor:
  bLength                 5
  bDescriptorType        15
  wTotalLength       0x0016
  bNumDeviceCaps          2
  USB 2.0 Extension Device Capability:
    bLength                 7
    bDescriptorType        16
    bDevCapabilityType      2
    bmAttributes   0x00000002
      HIRD Link Power Management (LPM) Supported
  SuperSpeed USB Device Capability:
    bLength                10
    bDescriptorType        16
    bDevCapabilityType      3
    bmAttributes         0x00
    wSpeedsSupported   0x000e
      Device can operate at Full Speed (12Mbps)
      Device can operate at High Speed (480Mbps)
      Device can operate at SuperSpeed (5Gbps)
    bFunctionalitySupport   1
      Lowest fully-functional device speed is Full Speed (12Mbps)
    bU1DevExitLat          10 micro seconds
    bU2DevExitLat         287 micro seconds
Device Status:     0x000f
  Self Powered
  Remote Wakeup Enabled
  U1 Enabled
  U2 Enabled

smoke@rkirilov-work-pc ~/projects/open-source/morrownr/rtl8852cu-20240510 $ uname -a
Linux rkirilov-work-pc 6.6.58-gentoo-dist #1 SMP PREEMPT_DYNAMIC Sat Oct 26 06:21:22 -00 2024 x86_64 13th Gen Intel(R) Core(TM) i7-13700H GenuineIntel GNU/Linux

[morrownr]

@smoke

Thanks for checking in. Let me know if you have questions. You will have an advantage over the rest of us as I think you are the only tester with an adapter that is single-state. We have been running into really bad problems related to storage mode and something seems to be amiss with adapters with the rtl8852/32cu chip that are multi-state so I am attempting to do a deep dive in udev and systemd to see if I can see what the problem is.

Let us know when you see any problems. This seems to be an anomaly for a Realtek out-of-kernel driver in the WiFi 6 catagory as many things appear to be working. Some bugs have been fixed already but I am sure things can be made better. Please pay attention to the documentation files and report anything that would make the docs better.

[morrownr]

Greetings all.

Issue 1 has been updated again. Please review the Progress Log and To-Do List. It seems we may have solved the issue of the driver not finding the kernel header files on some aarch64 distros so that is is closed. We still have some open issues that need attention and I am doing my best. There is a new problem shown in the To-Do List and I wrote an explanation as follows:

Explanation for the following To-Do item:

• discover why chplan and country code do not seem to be working as expected

In the current state of the driver, it seems to not follow wireless regulatory guidelines/laws. Example: In the US, it is necessary to have the ability to scan for interfering signals on 5 GHz DFS channels. In my opinion, it is unlikely that USB WiFi adapters based on the rtl8852/32cu chips have this capability and they do not seem to delay for the 60 seconds while this scan is taking place. This is a violation of US law. I also see other anomalies regarding regulatory guidelines/laws. I intend to investigate this. I do not feel comfortable releasing this driver to the public until such time as this issue is resolved.

Looking forward to your reports over the next week.

@morrownr

[morrownr]

FYI for all: I have run into some challenges trying to sort out a couple of the bugs we have run into so my plan is to freeze the code in this repo for a couple of weeks or so while I bring up another private repo based on much more recent code. It is code for the rtl8852bu. It may provide clues regarding possible fixes for this driver. If any of you have adapters based on the rtl8852/32bu chip, let me know if you want to test that driver

[Kiwimannen]

Lenovo Yoga-Slim-7

OS: Ubuntu 24.10 aarch64
Host: 83ED Yoga Slim 7 14Q8X9
Kernel: 6.12.0-17-qcom-x1e
Uptime: 30 mins
Packages: 1835 (dpkg), 13 (snap)
Shell: bash 5.2.32
Resolution: 2944x1840
DE: GNOME 47.0
WM: Mutter
WM Theme: Adwaita
Theme: Yaru-sage-dark [GTK2/3]
Icons: Yaru-sage-dark [GTK2/3]
Terminal: gnome-terminal
CPU: (12) @ 3.417GHz
Memory: 4707MiB / 14535MiB

Device: FENVI AX5400 WiFi 6E USB Adapter
Bus 006 Device 003: ID 0bda:c832 Realtek Semiconductor Corp. 802.11ax WLAN Adapter

Connected at 6GHz to my 6E router with:

2161.3 Mbit/s, 160 MHz, HE-MCS 10, HE-NSS 2
1921.5 Mbit/s, 160 MHz, HE-MCS 9, HE-NSS 2

Have a second (noname) device with a different antenna that has the same chip, and I get similar results with that one.

[morrownr]

@Kiwimannen

Any problems installing? Was the README clear?

[Kiwimannen]

I had some initial problems signing in, but that was not related to the driver ;)

Since I have installed Larry:s files with dkms before, this was pretty much straight forward - even easier with the settings, good work :)

I do have the same problems as others have pointed out when you have to detach and reinsert the usb-adapter sometimes for it to be recognized properly after install.

The readme is clear, but there is plenty of information (which is good) - maybe there could be a short "fast install" in the beginning ?

"git clone https://github.com/morrownr/rtl8852cu-20240510.git
cd rtl8852cu-20240510
sudo ./install-driver.sh

Read on for detailed information how to install and remove the drivers for your system." ?

[morrownr]

@Kiwimannen

I do have the same problems as others have pointed out when you have to detach and reinsert the usb-adapter sometimes for it to be recognized properly after install.

This is worrisome. This appears related to the adapters being multi-state. It also appears to have nothing to do with the driver because we can test it on a system without the driver installed. It usually shows up with the standard Realtek storage vid-pid when you run lsusb. Then if you unplug and replug, in some systems, it will switch modes and we see the vid-pid for wifi mode.

As I get time, I am going to do a deep dive and see if I can figure out what the problem is. I have adapters by the same maker but they use a rtl8832bu chip and I do not see this problem. The problem just appeared with the rtl8832cu chip and it appears widespread as in several adapters by different makers are showing the same or similar problems. I have tested on 4 systems here and have seen 4 different results that vary from works fine to can't get it work even with workarounds.

This is bordering on being a show stopper for me in that I hesitate to release this driver to the public as there will be some people that will look at the driver as an invitation to buy an adapter with the this chip. This and the ~10 second system freeze issue must be resolved in my mind before releasing this driver. As a last resort, I can try to get an updated version of this driver in the secord quarter of next year and hopefully we see some improvements on the system freeze issue but it is not clear how the multi-state problem is going to be resolved.

The readme is clear, but there is plenty of information (which is good) - maybe there could be a short "fast install" in the beginning ?

This is a good idea. I'll need to give this some thought. I've been maintaining some of these Realtek drivers for several years so I have seen a lot of things that can confuse users. I'll add this to the to-do list.

[morrownr]

@Kiwimannen

I had some initial problems signing in, but that was not related to the driver ;)

I have another user that is trying to get in and is having problems. What kind of problem did you see and how did you fix it?

[Kiwimannen]

I had to setup two-factor authentication on my account and sign in from console using (I think it was) "gh auth", or "gh auth login"

[morrownr]

All:

I merged 3 patches today. I have tested kernel 6.13 and we should be good to go. Please pull and test.

I am still looking at the problematic multi-state bug. I know more than I did but not everything I need to know.

[vincelawdev]

I am here.

I just installed rtl8852cu-20240510.

dkms status
nvidia/535.183.01, 6.8.0-49-generic, x86_64: installed
rtl8852cu/1.19.2.1, 6.8.0-49-generic, x86_64: installed

The driver is installed but I cannot see 6GHz WiFi and I cannot connect to any WiFi SSID. This may be an error caused by my previous attempts to install other drivers for this.

[vincelawdev]

lspci | grep -i network
06:00.0 Network controller: Intel Corporation Wi-Fi 6 AX200 (rev 1a)

Linux Mint sees my mainboard's built-in WiFi adaptor instead.

I can see USB WiFi in Network Settings and a list of SSIDs but all of them fail to connect.

[morrownr]

Hi @vincelawdev

Glad you made it.

Can I get you to start a new issue for this problem. In the new issue, post the results of:

$ lsusb

[morrownr]

Hello all,

I pushed a fairly large patch today. The biggest change was to rename remove-driver.sh to uninstall-driver.sh. I did this to have more consistency in naming.

That caused me to have to edit the docs and I also fixed a couple of things in the Makefile.

Request that everyone run:

$ git pull
$ sudo sh install-driver.sh

Let me know how things are going.

I am still working on various things at this point, including the multi-state problem that some of us have been seeing on some systems.

Nick

[Kiwimannen]

I updated and made a new install.

First time I choose "options 8852cu rtw_switch_usb_mode=1 rtw_country_code=SE"

after install a "lsusb" showed the usb-wifi in cd-rom mode, I removed and reinserted the usb-wifi, but after that the usb-wifi did not show up at all.

A second install I choose "options 8852cu rtw_switch_usb_mode=0 rtw_country_code=SE"

same thing happens with cd-rom mode, but after I reinsert the usb-wifi it shows up correct..

lsusb -v shows that the usb-wifi is in usb2 mode but I can use it and connect in 6GHz.

I tried a new install with mode=1, but the same thing happend.

[morrownr]

@Kiwimannen

Issue #10 is on this topic so I copied this message to issue #10 and will answer there.

[morrownr]

Hello everyone,

I merged a big patch today that contained mostly cosmetic issues. You can run git pull to update your code.

When this project started, I said that we would shoot for a public release on or about 2025-01-01. I think it is time to make the repo public today as we need to gather more information from a wider audience. While there are some issues that are not fixed, such as the mode switching issue, that is not a problem with this driver anyway so I think the driver is in good enough shape to open up to the public. Please do continue testing and I thank you for the help.

@morrownr