driver v5.2.20.2 released

[kimocoder]

Driver v5.2.20.2 is released. for 8812au at the moment. More information found here

A complete changelog may be found here

I've started the work on this! Currently, frame injection (packet injection) has been added and is reported working, but there is lots to do on it still. Progress will be posted below.

Feel free to tag along, all contributions are welcome to make this a good driver.

[RJVB]

One thing I noticed with the current driver in this repo as compared to the one from umdlife/rtl8812au_rtl8821au is that reported signal strengths are much lower, sometimes almost half. I now which value I find less surprising when I'm only getting about 75% signal strength when sitting next to the wifi modem with one driver while the other shows me a value close to 99%...

It'd be interesting to see if that new driver is any different in that department.

[RJVB]

Any progress here? I'm about to file a ticket for another issue I'm having with the current driver (5Ghz networks no longer being detected after a few sleep/wake cycles) ...

[kimocoder]

It's only for the 8812au. We use the rtl8821au version, which we added the 8814au HAL files too. This issue was just for informational purposes.

[RJVB]

8812au is exactly the chipset that interests me.

[kimocoder]

A solution for is may be to break them up in two, and let 8814au stay on v5.1.5 and update 8812au to v5.2.20 and add injection to it but it would take some time and I'd don't know whats the best right now

[RJVB]

If the 2 drivers are in separate subtrees (or otherwise independent enough) you could also include the latest version of both, under whatever version number you deem best.

I don't know what changed (in the air?) but since a few days it's become increasingly difficult to get a connection to the 5Ghz network; yesterday the connection I had simply stopped working without giving me as much as an error.

[kimocoder]

will order myself a 8812au adapter and start looking into it (adding injection to it for a starter) then maybe actually divide them up. seems like the smartest thing, if not they both stuck on v5.1.5 with all the issues it actually got (according to the v5.2.20 changelog).

[RJVB]

Sounds like a plan.

Actually I'm beginning to think my adapter has an issue itself. Still have to try it on a 2nd Win10 machine but the old one I tried it on showed the same behaviour as I've been starting to see on my Linux notebook. The WiFi stack shows a good, strong connection, but at some point nothing is actually being sent or received any longer. That happens with the 4.2 and 5.1 driver on Linux and the native one in Win10, so I think we can exclude any issues on the host and blame the adapter.

Fortunately these things cost almost nothing and this one might still be under warranty but getting it repaired might in fact turn out just as expensive as getting a new one. In which case I think I'll try to find a model with 2.4 and 5Ghz capabilities that doesn't require an external driver...

[kimocoder]

@RJVB well, here you go! @kcdtv you're a good one, you should also check this out. the v5.2.20 driver for 8812au with packet injection added.

I surely will need some testers over here, so I tagged you along..

https://github.com/aircrack-ng/rtl8812au/tree/v5.2.20

cloned gordboy's repo, enabled monitor mode & added radiotap headers and packet injection. It may be a little bit dirty still, but I just ordered myself a 8812au chipset (the AWUS036ACH) to play with, but it may at least be tested now, so shortly when I got mine I may continue the work.

damn it's nice to play fast for once :D

[RJVB]

Well, with the tip of that new branch built and loaded I see this in my syslog when I insert the USB adapter

Jun 22 22:09:53 bola kernel: [84115.645161] usb 1-2: new high-speed USB device number 7 using xhci_hcd
Jun 22 22:09:53 bola kernel: [84115.816851] usb 1-2: New USB device found, idVendor=0bda, idProduct=a811
Jun 22 22:09:53 bola kernel: [84115.816857] usb 1-2: New USB device strings: Mfr=1, Product=2, SerialNumber=3
Jun 22 22:09:53 bola kernel: [84115.816860] usb 1-2: Product: 802.11ac WLAN Adapter 
Jun 22 22:09:53 bola kernel: [84115.816865] usb 1-2: Manufacturer: Realtek 
Jun 22 22:09:53 bola kernel: [84115.816867] usb 1-2: SerialNumber: 00e04c000001
Jun 22 22:09:54 bola mtp-probe: checking bus 1, device 7: "/sys/devices/pci0000:00/0000:00:14.0/usb1/1-2"
Jun 22 22:09:54 bola mtp-probe: bus: 1, device: 7 was not an MTP device

and lsusb prints what looks like the expected output (the same as with the 5.1.5 driver), but I'm not seeing a wlan device appear (and the led on the dongle remains off).

[kimocoder]

Added your VID/PID "0bda:a811". Do a "git pull" to update and build/install it again and try. also, after the "make && make install" you may do the "modprobe -r 8812au" and then "modprobe 8812au" again.

Which adapter is this btw? :)

[RJVB]

I'll probably give you the verdict tomorrow.

The adapter:

https://www.amazon.fr/gp/product/B06XF118RS

[kimocoder]

Great. Added it to the driver, so it should be supported now. Also did some other nifty changes. The most important ones will have to wait until I get my own 8812 chipset, but it should work by now.

Will be back in a week for more. Thanks

[RJVB]

Not so fast ...

That latest version is progress in the sense that the device is recognised, but now I get a sort of boot loop when I insert it, and each time I remove and put it back the wlan device counter increases by 1. Below is an extract. That loop may exit at some point and then the wlanN device shows up in ifconfig but uninitialised, and it's not visible in the network manager.

Reloading the 5.1.5 or 4.3.20 driver gives me back the wlan3 device I'm used to.

Jun 23 00:02:02 bola NetworkManager[5874]: <error> [1529704922.840607] [nm-system.c:768] nm_system_iface_get_flags(): (unknown): failed to get interface link object
Jun 23 00:02:02 bola wpa_supplicant[6682]: Could not read interface wlan7 flags: No such device
Jun 23 00:02:02 bola wpa_supplicant[6682]: Could not read interface wlan7 flags: No such device
Jun 23 00:02:02 bola wpa_supplicant[6682]: WEXT: Could not set interface 'wlan7' UP
Jun 23 00:02:02 bola wpa_supplicant[6682]: wlan7: Failed to initialize driver interface
Jun 23 00:02:02 bola NetworkManager[5874]: <error> [1529704922.870647] [nm-supplicant-interface.c:997] interface_add_cb(): (wlan7): error adding interface: wpa_supplicant couldn't grab this interface.
Jun 23 00:02:02 bola NetworkManager[5874]: dbus_g_proxy_cancel_call: assertion 'pending != NULL' failed
Jun 23 00:02:02 bola NetworkManager[5874]: <warn> Trying to remove a non-existant call id.
Jun 23 00:02:03 bola kernel: [90844.985239] usb 1-2: reset high-speed USB device number 14 using xhci_hcd
Jun 23 00:02:15 bola kernel: [90857.494759] usbcore: registered new interface driver 8814au
Jun 23 00:02:18 bola kernel: [90860.861931] usb 1-2: USB disconnect, device number 14
Jun 23 00:02:18 bola laptop-mode: Laptop mode 
Jun 23 00:02:18 bola laptop-mode: enabled, not active
Jun 23 00:02:19 bola laptop-mode: Laptop mode 
Jun 23 00:02:19 bola laptop-mode: enabled, not active
Jun 23 00:02:21 bola kernel: [90863.504981] usb 1-2: new high-speed USB device number 15 using xhci_hcd
Jun 23 00:02:21 bola kernel: [90863.680577] usb 1-2: New USB device found, idVendor=0bda, idProduct=a811
Jun 23 00:02:21 bola kernel: [90863.680587] usb 1-2: New USB device strings: Mfr=1, Product=2, SerialNumber=3
Jun 23 00:02:21 bola kernel: [90863.680595] usb 1-2: Product: 802.11ac WLAN Adapter 
Jun 23 00:02:21 bola kernel: [90863.680602] usb 1-2: Manufacturer: Realtek 
Jun 23 00:02:21 bola kernel: [90863.680607] usb 1-2: SerialNumber: 00e04c000001
Jun 23 00:02:21 bola mtp-probe: checking bus 1, device 15: "/sys/devices/pci0000:00/0000:00:14.0/usb1/1-2"
Jun 23 00:02:21 bola mtp-probe: bus: 1, device: 15 was not an MTP device
Jun 23 00:02:21 bola laptop-mode: Laptop mode 
Jun 23 00:02:21 bola laptop-mode: enabled, not active
Jun 23 00:02:21 bola kernel: [90863.841468] usbcore: registered new interface driver rtl8812au
Jun 23 00:02:21 bola NetworkManager[5874]: <info> rfkill15: found WiFi radio killswitch (at /sys/devices/pci0000:00/0000:00:14.0/usb1/1-2/1-2:1.0/ieee80211/phy11/rfkill15) (driver rtl8812au)
Jun 23 00:02:21 bola laptop-mode: Laptop mode 
Jun 23 00:02:21 bola laptop-mode: enabled, not active
Jun 23 00:02:21 bola kernel: [90863.910732] rtl8812au 1-2:1.0 wlan8: renamed from wlan0
Jun 23 00:02:22 bola NetworkManager[5874]:    SCPlugin-Ifupdown: devices added (path: /sys/devices/pci0000:00/0000:00:14.0/usb1/1-2/1-2:1.0/net/wlan8, iface: wlan8)
Jun 23 00:02:22 bola NetworkManager[5874]:    SCPlugin-Ifupdown: device added (path: /sys/devices/pci0000:00/0000:00:14.0/usb1/1-2/1-2:1.0/net/wlan8, iface: wlan8): no ifupdown configuration found.
Jun 23 00:02:22 bola kernel: [90863.955581] systemd-udevd[31996]: renamed network interface wlan0 to wlan8
Jun 23 00:02:22 bola NetworkManager[5874]: <info> (wlan8): using nl80211 for WiFi device control
Jun 23 00:02:22 bola NetworkManager[5874]: <info> (wlan8): driver supports Access Point (AP) mode
Jun 23 00:02:22 bola NetworkManager[5874]: <info> (wlan8): new 802.11 WiFi device (driver: 'rtl8812au' ifindex: 14)
Jun 23 00:02:22 bola NetworkManager[5874]: <info> (wlan8): exported as /org/freedesktop/NetworkManager/Devices/16
Jun 23 00:02:22 bola NetworkManager[5874]: <info> (wlan8): device state change: unmanaged -> unavailable (reason 'managed') [10 20 2]
Jun 23 00:02:22 bola NetworkManager[5874]: <info> (wlan8): bringing up device.
UdevQt: unhandled device action "move"
Jun 23 00:02:22 bola NetworkManager[5874]: <info> (wlan8): deactivating device (reason 'managed') [2]
Jun 23 00:02:22 bola wpa_supplicant[6682]: Could not set interface wlan8 flags (UP): Operation not permitted
Jun 23 00:02:22 bola wpa_supplicant[6682]: nl80211: Could not set interface 'wlan8' UP
Jun 23 00:02:22 bola wpa_supplicant[6682]: Could not set interface wlan8 flags (UP): Operation not permitted
Jun 23 00:02:22 bola wpa_supplicant[6682]: WEXT: Could not set interface 'wlan8' UP
Jun 23 00:02:22 bola wpa_supplicant[6682]: wlan8: Failed to initialize driver interface
Jun 23 00:02:22 bola NetworkManager[5874]: <error> [1529704942.875948] [nm-supplicant-interface.c:997] interface_add_cb(): (wlan8): error adding interface: wpa_supplicant couldn't grab this interface.
Jun 23 00:02:22 bola NetworkManager[5874]: dbus_g_proxy_cancel_call: assertion 'pending != NULL' failed
Jun 23 00:02:22 bola NetworkManager[5874]: <info> (wlan8): supplicant interface state: starting -> down
Jun 23 00:02:22 bola NetworkManager[5874]: <warn> Trying to remove a non-existant call id.
Jun 23 00:02:23 bola wpa_supplicant[6682]: Could not set interface wlan8 flags (UP): Operation not permitted
Jun 23 00:02:23 bola wpa_supplicant[6682]: nl80211: Could not set interface 'wlan8' UP
Jun 23 00:02:23 bola wpa_supplicant[6682]: Could not set interface wlan8 flags (UP): Operation not permitted
Jun 23 00:02:23 bola wpa_supplicant[6682]: WEXT: Could not set interface 'wlan8' UP
Jun 23 00:02:23 bola wpa_supplicant[6682]: wlan8: Failed to initialize driver interface
Jun 23 00:02:23 bola NetworkManager[5874]: <error> [1529704943.438531] [nm-supplicant-interface.c:997] interface_add_cb(): (wlan8): error adding interface: wpa_supplicant couldn't grab this interface.
Jun 23 00:02:23 bola NetworkManager[5874]: dbus_g_proxy_cancel_call: assertion 'pending != NULL' failed
Jun 23 00:02:23 bola NetworkManager[5874]: <info> (wlan8): supplicant interface state: starting -> down
Jun 23 00:02:23 bola NetworkManager[5874]: <warn> Trying to remove a non-existant call id.
Jun 23 00:02:23 bola wpa_supplicant[6682]: Could not set interface wlan8 flags (UP): Operation not permitted
Jun 23 00:02:23 bola wpa_supplicant[6682]: nl80211: Could not set interface 'wlan8' UP
Jun 23 00:02:24 bola wpa_supplicant[6682]: Could not set interface wlan8 flags (UP): Operation not permitted
Jun 23 00:02:24 bola wpa_supplicant[6682]: WEXT: Could not set interface 'wlan8' UP
Jun 23 00:02:24 bola wpa_supplicant[6682]: wlan8: Failed to initialize driver interface
Jun 23 00:02:24 bola NetworkManager[5874]: <error> [1529704944.5889] [nm-supplicant-interface.c:997] interface_add_cb(): (wlan8): error adding interface: wpa_supplicant couldn't grab this interface.
Jun 23 00:02:24 bola NetworkManager[5874]: dbus_g_proxy_cancel_call: assertion 'pending != NULL' failed
Jun 23 00:02:24 bola NetworkManager[5874]: <info> (wlan8): supplicant interface state: starting -> down
Jun 23 00:02:24 bola NetworkManager[5874]: <warn> Trying to remove a non-existant call id.
Jun 23 00:02:24 bola wpa_supplicant[6682]: Could not set interface wlan8 flags (UP): Operation not permitted
Jun 23 00:02:24 bola wpa_supplicant[6682]: nl80211: Could not set interface 'wlan8' UP
Jun 23 00:02:24 bola ModemManager[5343]: <warn>  Couldn't find support for device at '/sys/devices/pci0000:00/0000:00:14.0/usb1/1-2': not supported by any plugin
Jun 23 00:02:24 bola wpa_supplicant[6682]: Could not set interface wlan8 flags (UP): Operation not permitted
Jun 23 00:02:24 bola wpa_supplicant[6682]: WEXT: Could not set interface 'wlan8' UP
Jun 23 00:02:24 bola wpa_supplicant[6682]: wlan8: Failed to initialize driver interface
Jun 23 00:02:24 bola NetworkManager[5874]: <error> [1529704944.567621] [nm-supplicant-interface.c:997] interface_add_cb(): (wlan8): error adding interface: wpa_supplicant couldn't grab this interface.
Jun 23 00:02:24 bola NetworkManager[5874]: dbus_g_proxy_cancel_call: assertion 'pending != NULL' failed
Jun 23 00:02:24 bola NetworkManager[5874]: <info> (wlan8): supplicant interface state: starting -> down
Jun 23 00:02:24 bola NetworkManager[5874]: <warn> Trying to remove a non-existant call id.
Jun 23 00:02:24 bola wpa_supplicant[6682]: Could not set interface wlan8 flags (UP): Operation not permitted
Jun 23 00:02:24 bola wpa_supplicant[6682]: nl80211: Could not set interface 'wlan8' UP
Jun 23 00:02:25 bola wpa_supplicant[6682]: Could not set interface wlan8 flags (UP): Operation not permitted
Jun 23 00:02:25 bola wpa_supplicant[6682]: WEXT: Could not set interface 'wlan8' UP
Jun 23 00:02:25 bola wpa_supplicant[6682]: wlan8: Failed to initialize driver interface
Jun 23 00:02:25 bola NetworkManager[5874]: <error> [1529704945.139873] [nm-supplicant-interface.c:997] interface_add_cb(): (wlan8): error adding interface: wpa_supplicant couldn't grab this interface.
Jun 23 00:02:25 bola NetworkManager[5874]: dbus_g_proxy_cancel_call: assertion 'pending != NULL' failed
Jun 23 00:02:25 bola NetworkManager[5874]: <info> (wlan8): supplicant interface state: starting -> down
Jun 23 00:02:25 bola NetworkManager[5874]: <warn> Trying to remove a non-existant call id.
Jun 23 00:02:25 bola wpa_supplicant[6682]: Could not set interface wlan8 flags (UP): Operation not permitted
Jun 23 00:02:25 bola wpa_supplicant[6682]: nl80211: Could not set interface 'wlan8' UP
Jun 23 00:02:25 bola wpa_supplicant[6682]: Could not set interface wlan8 flags (UP): Operation not permitted
Jun 23 00:02:25 bola wpa_supplicant[6682]: WEXT: Could not set interface 'wlan8' UP
Jun 23 00:02:25 bola wpa_supplicant[6682]: wlan8: Failed to initialize driver interface
Jun 23 00:02:25 bola NetworkManager[5874]: <error> [1529704945.703094] [nm-supplicant-interface.c:997] interface_add_cb(): (wlan8): error adding interface: wpa_supplicant couldn't grab this interface.
Jun 23 00:02:25 bola NetworkManager[5874]: dbus_g_proxy_cancel_call: assertion 'pending != NULL' failed
Jun 23 00:02:25 bola NetworkManager[5874]: <info> (wlan8): supplicant interface state: starting -> down
Jun 23 00:02:25 bola NetworkManager[5874]: <warn> Trying to remove a non-existant call id.
Jun 23 00:02:25 bola wpa_supplicant[6682]: Could not set interface wlan8 flags (UP): Operation not permitted
Jun 23 00:02:25 bola wpa_supplicant[6682]: nl80211: Could not set interface 'wlan8' UP
Jun 23 00:02:26 bola wpa_supplicant[6682]: Could not set interface wlan8 flags (UP): Operation not permitted
Jun 23 00:02:26 bola wpa_supplicant[6682]: WEXT: Could not set interface 'wlan8' UP
Jun 23 00:02:26 bola wpa_supplicant[6682]: wlan8: Failed to initialize driver interface
Jun 23 00:02:26 bola NetworkManager[5874]: <error> [1529704946.261850] [nm-supplicant-interface.c:997] interface_add_cb(): (wlan8): error adding interface: wpa_supplicant couldn't grab this interface.
Jun 23 00:02:26 bola NetworkManager[5874]: dbus_g_proxy_cancel_call: assertion 'pending != NULL' failed
Jun 23 00:02:26 bola NetworkManager[5874]: <info> (wlan8): supplicant interface state: starting -> down
Jun 23 00:02:26 bola NetworkManager[5874]: <warn> Trying to remove a non-existant call id.
Jun 23 00:02:26 bola wpa_supplicant[6682]: Could not set interface wlan8 flags (UP): Operation not permitted
Jun 23 00:02:26 bola wpa_supplicant[6682]: nl80211: Could not set interface 'wlan8' UP
Jun 23 00:02:26 bola wpa_supplicant[6682]: Could not set interface wlan8 flags (UP): Operation not permitted
Jun 23 00:02:26 bola wpa_supplicant[6682]: WEXT: Could not set interface 'wlan8' UP
Jun 23 00:02:26 bola wpa_supplicant[6682]: wlan8: Failed to initialize driver interface
Jun 23 00:02:26 bola NetworkManager[5874]: <error> [1529704946.829346] [nm-supplicant-interface.c:997] interface_add_cb(): (wlan8): error adding interface: wpa_supplicant couldn't grab this interface.
Jun 23 00:02:26 bola NetworkManager[5874]: dbus_g_proxy_cancel_call: assertion 'pending != NULL' failed
Jun 23 00:02:26 bola NetworkManager[5874]: <info> (wlan8): supplicant interface state: starting -> down
Jun 23 00:02:26 bola NetworkManager[5874]: <warn> Trying to remove a non-existant call id.
Jun 23 00:02:27 bola wpa_supplicant[6682]: Could not set interface wlan8 flags (UP): Operation not permitted
Jun 23 00:02:27 bola wpa_supplicant[6682]: nl80211: Could not set interface 'wlan8' UP
Jun 23 00:02:27 bola wpa_supplicant[6682]: Could not set interface wlan8 flags (UP): Operation not permitted
Jun 23 00:02:27 bola wpa_supplicant[6682]: WEXT: Could not set interface 'wlan8' UP
Jun 23 00:02:27 bola wpa_supplicant[6682]: wlan8: Failed to initialize driver interface
Jun 23 00:02:27 bola NetworkManager[5874]: <error> [1529704947.385816] [nm-supplicant-interface.c:997] interface_add_cb(): (wlan8): error adding interface: wpa_supplicant couldn't grab this interface.
Jun 23 00:02:27 bola NetworkManager[5874]: dbus_g_proxy_cancel_call: assertion 'pending != NULL' failed
Jun 23 00:02:27 bola NetworkManager[5874]: <info> (wlan8): supplicant interface state: starting -> down
Jun 23 00:02:27 bola NetworkManager[5874]: <warn> Trying to remove a non-existant call id.
Jun 23 00:02:27 bola wpa_supplicant[6682]: Could not set interface wlan8 flags (UP): Operation not permitted
Jun 23 00:02:27 bola wpa_supplicant[6682]: nl80211: Could not set interface 'wlan8' UP
Jun 23 00:02:27 bola wpa_supplicant[6682]: Could not set interface wlan8 flags (UP): Operation not permitted
Jun 23 00:02:27 bola wpa_supplicant[6682]: WEXT: Could not set interface 'wlan8' UP
Jun 23 00:02:27 bola wpa_supplicant[6682]: wlan8: Failed to initialize driver interface
Jun 23 00:02:27 bola NetworkManager[5874]: <error> [1529704947.942060] [nm-supplicant-interface.c:997] interface_add_cb(): (wlan8): error adding interface: wpa_supplicant couldn't grab this interface.
Jun 23 00:02:27 bola NetworkManager[5874]: dbus_g_proxy_cancel_call: assertion 'pending != NULL' failed
Jun 23 00:02:27 bola NetworkManager[5874]: <info> (wlan8): supplicant interface state: starting -> down
Jun 23 00:02:27 bola NetworkManager[5874]: <warn> Trying to remove a non-existant call id.
Jun 23 00:02:28 bola wpa_supplicant[6682]: Could not set interface wlan8 flags (UP): Operation not permitted
Jun 23 00:02:28 bola wpa_supplicant[6682]: nl80211: Could not set interface 'wlan8' UP
Jun 23 00:02:28 bola wpa_supplicant[6682]: Could not set interface wlan8 flags (UP): Operation not permitted
Jun 23 00:02:28 bola wpa_supplicant[6682]: WEXT: Could not set interface 'wlan8' UP
Jun 23 00:02:28 bola wpa_supplicant[6682]: wlan8: Failed to initialize driver interface
Jun 23 00:02:28 bola NetworkManager[5874]: <error> [1529704948.486354] [nm-supplicant-interface.c:997] interface_add_cb(): (wlan8): error adding interface: wpa_supplicant couldn't grab this interface.
Jun 23 00:02:28 bola NetworkManager[5874]: dbus_g_proxy_cancel_call: assertion 'pending != NULL' failed
Jun 23 00:02:28 bola NetworkManager[5874]: <info> (wlan8): supplicant interface state: starting -> down
Jun 23 00:02:28 bola NetworkManager[5874]: <warn> Trying to remove a non-existant call id.
Jun 23 00:02:28 bola wpa_supplicant[6682]: Could not set interface wlan8 flags (UP): Operation not permitted
Jun 23 00:02:28 bola wpa_supplicant[6682]: nl80211: Could not set interface 'wlan8' UP
Jun 23 00:02:29 bola wpa_supplicant[6682]: Could not set interface wlan8 flags (UP): Operation not permitted
Jun 23 00:02:29 bola wpa_supplicant[6682]: WEXT: Could not set interface 'wlan8' UP
Jun 23 00:02:29 bola wpa_supplicant[6682]: wlan8: Failed to initialize driver interface
Jun 23 00:02:29 bola NetworkManager[5874]: <error> [1529704949.33786] [nm-supplicant-interface.c:997] interface_add_cb(): (wlan8): error adding interface: wpa_supplicant couldn't grab this interface.
Jun 23 00:02:29 bola NetworkManager[5874]: dbus_g_proxy_cancel_call: assertion 'pending != NULL' failed
Jun 23 00:02:29 bola NetworkManager[5874]: <info> (wlan8): supplicant interface state: starting -> down
Jun 23 00:02:29 bola NetworkManager[5874]: <warn> Trying to remove a non-existant call id.
Jun 23 00:02:29 bola wpa_supplicant[6682]: Could not set interface wlan8 flags (UP): Operation not permitted
Jun 23 00:02:29 bola wpa_supplicant[6682]: nl80211: Could not set interface 'wlan8' UP
Jun 23 00:02:29 bola wpa_supplicant[6682]: Could not set interface wlan8 flags (UP): Operation not permitted
Jun 23 00:02:29 bola wpa_supplicant[6682]: WEXT: Could not set interface 'wlan8' UP
Jun 23 00:02:29 bola wpa_supplicant[6682]: wlan8: Failed to initialize driver interface
Jun 23 00:02:29 bola NetworkManager[5874]: <error> [1529704949.587329] [nm-supplicant-interface.c:997] interface_add_cb(): (wlan8): error adding interface: wpa_supplicant couldn't grab this interface.
Jun 23 00:02:29 bola NetworkManager[5874]: dbus_g_proxy_cancel_call: assertion 'pending != NULL' failed
Jun 23 00:02:29 bola NetworkManager[5874]: <info> (wlan8): supplicant interface state: starting -> down
Jun 23 00:02:29 bola NetworkManager[5874]: <warn> Trying to remove a non-existant call id.
Jun 23 00:02:29 bola wpa_supplicant[6682]: Could not set interface wlan8 flags (UP): Operation not permitted
Jun 23 00:02:29 bola wpa_supplicant[6682]: nl80211: Could not set interface 'wlan8' UP
Jun 23 00:02:30 bola wpa_supplicant[6682]: Could not set interface wlan8 flags (UP): Operation not permitted
Jun 23 00:02:30 bola wpa_supplicant[6682]: WEXT: Could not set interface 'wlan8' UP
Jun 23 00:02:30 bola wpa_supplicant[6682]: wlan8: Failed to initialize driver interface
Jun 23 00:02:30 bola NetworkManager[5874]: <error> [1529704950.132693] [nm-supplicant-interface.c:997] interface_add_cb(): (wlan8): error adding interface: wpa_supplicant couldn't grab this interface.
Jun 23 00:02:30 bola NetworkManager[5874]: dbus_g_proxy_cancel_call: assertion 'pending != NULL' failed
Jun 23 00:02:30 bola NetworkManager[5874]: <info> (wlan8): supplicant interface state: starting -> down
Jun 23 00:02:30 bola NetworkManager[5874]: <warn> Trying to remove a non-existant call id.
Jun 23 00:02:30 bola wpa_supplicant[6682]: Could not set interface wlan8 flags (UP): Operation not permitted
Jun 23 00:02:30 bola wpa_supplicant[6682]: nl80211: Could not set interface 'wlan8' UP

[kimocoder]

It says "[90857.494759] usbcore: registered new interface driver 8814au" .. so this is a 8814au chipset, not the 8812au? Not doing more until mine has come, but this seems like the 8814au chipset?

[kimocoder]

Yepp, this is the Realtek RTL8811 chipset you got there. It runs under RTL8821AU drivers, not the RTL8812AU unfortunately..

[RJVB]

Indeed, Win10 shows it as 8811au, but googling led (and still leads) me to the 8812au driver. I can confirm the 8814au driver doesn't work, but the 8812au driver does (until it doesn't any longer for some weird timeout reason).

I'm pretty certain the vendor also claims it has a 8812 chipset in the comments on that Amazon page.

[kimocoder]

It points to the 8812au, but the one marked 8812a_8821a (which almost is the same) and also is a part of the Makefile, but we don't have the HAL files for it to add it to the driver. This driver only supports the 8812au specific, while v5.1.5 supports 8812au, 8814au and 8821au .. it's a bit messy this one but this is how it is. Take a look St the screenshot from a "working" version from abperiasany

[RJVB]

In short, I'm stuck on 5.1.5 or older unless we can somehow get those HAL files? (I thought HAL was obsolete and no longer used?)

[kimocoder]

I bet so, could always take the HAL from v5.1.5, but that won't bring the changes from v5.2.20 to it anyway. I really think the HAL files is needed, you may try to disable 8812 compilation (turn it to "n" in Makefile) and enabling the 8821A instead and try compiling and see if that works

[RJVB]

I should be able to test it quickly, yes. (From the moment I actually see the notification ;) ).

[kimocoder]

It won't work. Also, tried to add the HAL for 8821a from v5.1.5 but that was not to easy as the though was, it needs quite some work

[RJVB]

Well, I guess I'll continue to juggle between the 5.1.5 driver and my own 4.3.20 version (https://github.com/RJVB/rtl8812au_rtl8821au) then...

[kcdtv]

Hi there! I played a little around with this new branch and the chip rtl8812au (awuso36ach) in kali linux (4.16.0-kali2-amd64)
Compilation was clean with no warnings and mode monitor and packet injections seems to work correctly in both bands ex: DoS in 5Ghz to collect the handshake: I checked also the handshake with wireshark and everything was fine, no "corrupted packet" I can't say if the mode monitor behavior as improved or not regarding the other branchs. At least it doesn't look worse at first sight. Unfortunately i couldn't check with the latest revisions of reaver and wash as i have a generic error pointing to pcap But it worked as intended with versions up to 1.6.4... We will check that

[kimocoder]

Aaaah this is great! Will continue to work when I get my own 8812au chipset shortly. The differences between the branches is found in the docs/ folder, the Realtek_changelog, there are quite some fixes since the v5.1.5 there. My additions is in the other changelog, there's also an TODO/upcoming notes. txpower setting (unleashing of power) hasn't been added yet too, but it's tba.

Great! I think the packet injection solution here is a future solution for Realtek drivers / other drivers from Realtek which for me is just awesome to know / learn.

Be back later, thanks for the feedback @kcdtv -- this looks promising

[kcdtv]

I tried in Xubuntu 18.06 the branch 5.2.20 and i have an error compiling:

make ARCH=x86_64 CROSS_COMPILE= -C /lib/modules/4.15.0-23-generic/build M=/home/kcdtv/realtekac/aircrack-ng/rtl8812au  modules
make[1] : on entre dans le répertoire « /usr/src/linux-headers-4.15.0-23-generic »
Makefile:976: "Cannot use CONFIG_STACK_VALIDATION=y, please install libelf-dev, libelf-devel or elfutils-libelf-devel"
  CC [M]  /home/kcdtv/realtekac/aircrack-ng/rtl8812au/core/rtw_cmd.o
In file included from /home/kcdtv/realtekac/aircrack-ng/rtl8812au/include/osdep_service.h:41:0,
                 from /home/kcdtv/realtekac/aircrack-ng/rtl8812au/include/drv_types.h:32,
                 from /home/kcdtv/realtekac/aircrack-ng/rtl8812au/core/rtw_cmd.c:22:
/home/kcdtv/realtekac/aircrack-ng/rtl8812au/include/osdep_service_linux.h: In function ‘_init_timer’:
/home/kcdtv/realtekac/aircrack-ng/rtl8812au/include/osdep_service_linux.h:280:8: error: ‘_timer {aka struct timer_list}’ has no member named ‘data’
  ptimer->data = (unsigned long)cntx;
        ^~
/home/kcdtv/realtekac/aircrack-ng/rtl8812au/include/osdep_service_linux.h:281:2: error: implicit declaration of function ‘init_timer’; did you mean ‘_init_timer’? [-Werror=implicit-function-declaration]
  init_timer(ptimer);
  ^~~~~~~~~~
  _init_timer
cc1: some warnings being treated as errors
scripts/Makefile.build:332: recipe for target '/home/kcdtv/realtekac/aircrack-ng/rtl8812au/core/rtw_cmd.o' failed
make[2]: *** [/home/kcdtv/realtekac/aircrack-ng/rtl8812au/core/rtw_cmd.o] Error 1
Makefile:1552: recipe for target '_module_/home/kcdtv/realtekac/aircrack-ng/rtl8812au' failed
make[1]: *** [_module_/home/kcdtv/realtekac/aircrack-ng/rtl8812au] Error 2
make[1] : on quitte le répertoire « /usr/src/linux-headers-4.15.0-23-generic »
Makefile:1264: recipe for target 'modules' failed
make: *** [modules] Error 2

It first points out some missing dependencies which doesn't really make sense as i can compile the other branches. Anyway, i installed the missing packages "to be sure" and the same error happened (without any reference to the missing dependencies this time.) Gordboy v5.2.20 version compiles correctly (not a surprise as he develops on Ubuntu) Linux Mint users might face the same error as it is Ubuntu based. Cheers

[kimocoder]

Yes, this is the _init_timer which was introduced with kernel v4.15 and above. May fix it when I'm back to my laptop. Thanks for the heads up.

If your in a hurry, you might find a solution to it here https://github.com/aircrack-ng/rtl8812au/commit/f221a169f281dab9756a176ec2abd91e0eba7d19

[kimocoder]

@kcdtv it builds fine under Ubuntu Bionic using kernel 4.15.0-23 ? I installed the libelf-dev too and it stopped nagging about it straight away.. compiled without warnings at all.

The error you reported are the init timer changes as I said above, but fix for that was allready added in the gordboy branch before I started. But this is great anyway, it means I may continue my work without the hazzle :D

[kekkokk]

are these suitable for 8814au?

[kimocoder]

As the README says, no. But will add the HAL files and support it om the future, if it fits to it

[kimocoder]

• Fix injected frames drop
• Fixed compilation on some ARM platforms
• Added support for PPC platform
• Minor cleanups

Just merged! More tba in a week or so

[kimocoder]

• Enable TX queue in the monitor mode
• Added support for namespaces (CONFIG_NET_NS)
• Added more Makefile debug
• Added more VID/PID's for future use (8814/8821)
• Updated to a more informational README

Just merged. More tba

[kimocoder]

Ok.. used some hours today looking into 8814 and 8821 issue today, and what I've learned is that the v5.2.20 got some additions that makes is hard to add those chipset support to it, the only way is that we get the files needed from other releases, which doesn't exist at the moment.

But I've learned something else too, if we use the v5.2.9.3 we could add HAL support for both 8814 and 8821 to it. That would anyway bring some updates (as seen on the Realtek changelog) and that would anyway be better than the v5.1.5 I think.

As for Kali, the drivers should be divided into 2 (3) I think, when finished anyway. The 8812 on v5.2.20 and 8814 (and 8821) on v5.2.9.3 possibly.

Will take some time to complete all of this, but I'm on the case (on the v5.2.20 currently)

[kimocoder]

Driver v5.2.20.2 was just released, fixing:

• Android 8+ support
• Fixing 4-second delay/lag issue

But I will put this on hold for some weeks/a month and fix the v5.2.9.3, because the priority should be to have proper frame injection support and most up-to-date drivers in Kali and we wan't all chipsets, not only the 8812, but also 8811, 8814 and 8821 support.

[RJVB]

Is there any way to get a statement from Realtek why they're excluding still very common (and for sale) chipsets from their latest drivers?

[kimocoder]

Nah, probably not. But these changes may be ported anyway, so it's great with a new release. I'm happy, starting on v5.2.9.3 today, easy to add 8821 and 8814 to it, so I'll add frame injection and some monitor patches for a starter, that should bring you (8811) and me (8814) to a next level thing at least

[kimocoder]

Ok. I've uploaded the v5.2.9.3 to the "v5.2.9" branch today, with 8821 support, but here again.. I can't manage it to support (adding 8814 HAL) 8814 again. These driver make me go nuts.

Will get the 8821 up to v5.2.9.3 anyway, but adding all these patches to both branches will take some time. It's missing frame injection atm, but I'll add in in couple of days along with some monitor/channel patches.

[kcdtv]

Just to confirm that the compiling worked in ubuntu this time with the updates you made. I personally do not notice any advantages in mode monitor with 5.2.9 and 5.2.20 over 5.1.5 I guess the real plus is more in managed mode with this vulnerabilities and bugs patched. And the code optimizations. A shame that adaptation for rtl8814au isn't more easy. But for me branch 5.1.5 delivers it too in mode managed so the situation is OK. Cheers and thanks for all this great job :smiley_cat:

[kimocoder]

I'll start on the v5.2.20.2 asap, as the phydm has been improved reducing the 4 second glitch bug. The 8814 sadly must stay om v5.1.5 until I get a working HAL for it, or if a update to it is released. Info tba

[kimocoder]

@kcdtv i pushed the v5.2.20.2 driver in the repo, upon the v5.2.20. It needs some testing, just see how it works.. all the above patches have been added to it manually (by hand) but it still needs to be checked. Still haven't got my 8812 adapter, but I can't wait to continue :dancer:

[kimocoder]

• Fix the CONFIG_NET_NS, support for network namespaces
• Count injected frames in interface stats
• Be quiet about power save
• Unmask all available channels
• Enable the channel 14 and 80 MHz width channels in the regdom
• Fix error switching to 80MHz channels under certain conditions
• Stop exporting unnecessary symbols
• Fix channel switch in monitor mode if both CONFIG_WIRELESS_EXT and CONFIG_CFG80211_WEXT are enabled

Just got merged. Enjoy!

[kimocoder]

• Add CONFIG_USB2_EXTERNAL_POWER supported
• unlock all channels in REALTEK_DEFINE (default) channel plan
• enable DBG_HIGHEST_RATE_BMP_BW_CHANGE with CONFIG_DBG_TXPOWER
• fix txpwr-by-rate logging and value when txpower-override is enabled
• use BIT(n) macros instead of pre-calculated masks in bunch of places
• Add override module parameter for tx-power-index
• ioctl_cfg80211.init_vht_capab: print highest rates that are set there
• Enable rx_stbc bit for 5G in HT mode
• Updated README.md

Just merged. Enjoy! I think this is where I have to stop until å get the 8812au chipset, cause adding txpower switching with "iwconfig txpower" would needs testing. Also, my USB ammeter is on it's way from China, which would also be needed for the tests. It's on the TODO, so it's coming, one of the most important thing in this driver is that we fix the range (power / txpower) issue which you may find in another issue.

What's next?

• txpower control needs to be added
• USB2/3 switching/function needs to be tested
• iw infomation needs to get more output

[RJVB]

I'm lost a bit, what branch was that committed to?

Anyway, I fetched the 5.2.9.3 branch and built it. Seems to work with my adapter.

FWIW, I tried a few build tricks to see if those had any incidence on the disconnects and connection timeouts I've been seeing.

For now, it looks like building with "-O3 -ftracer -march=native" gives me the most stable connections (using GCC 7.3 on an Intel N3150).

[kimocoder]

To the v5.2.20 which now actually is v5.2.20.2. I've just added the v5.2.9.3 to the v5.2.9 branch, but haven't added frame injection to it, but at least you should have a bit more working driver then the v5.1.5

[kimocoder]

Thanks to another repo, a fork of Realtek which are based on another "version" of this driver, I've managed to pull some ideas on howto add the 8821 chipset.. therefor, we've added the 8821au support! It needs to be tested still, but it looks good!

• Added support for 8821au chipset Just flip the switches in Makefile, both "CONFIG_RTL8812A" and "CONFIG_RTL8821A" should be turned to "y" to build the 8821au driver.

@RJVB this is for you :) you got the v5.2.20.2 support

[RJVB]

@RJVB this is for you :)

Why thanks ... with this version the interface no longer keeps increasing each time I reconnect the device, but I'm not seeing any networks through it ;)

With yesterday's 5.2.9.3 rebuild I could use a 5Ghz network for the day (hadn't happened in a long time), but this morning I was again asked to enter the WPA key after waking the machine from sleep. That's a sure sign of "don't bother, won't connect anyway" with this device.

Why oh why is there no IWL-based USB WiFi dongle ...

[etem]

Hi guys, I've been following this and I can also confirm that 5Ghz networks aren't working with v5.2.20.2 and 8811au chipset. Haven't tried with 5.2.9.3 though.

[RJVB]

To be clear: I'm not seeing any networks with this version of the driver.

[kimocoder]

Ok. Could you try this version and see if that works? It's the v5.2.20 with 8821au support, see if the problems exists over there? Then I may have a pinpoint on in the issue, as the support patch is the same on that repo

[RJVB]

No, this one seems to work. I had to enter the 5ghz WPA key (of course, even though it was registed in the corresponding network manager profile). The first connection attempt got stuck on "configuring interface" so I disconnected, removed the device and stuck it back in - then it connected just fine.

We'll see how it holds up.