dmesg output:

has compilation issues with timers..

Looks like this version is not working with rtl8811au.
Does the driver support that chip?

Using git clone and dkms I've got two modules
/lib/modules/4.13.0-17-lowlatency/kernel/drivers/net/wireless/rtl8812au.ko
/lib/modules/4.13.0-17-lowlatency/updates/dkms/8812au.ko

dkms status gives
§§:~/rtl8812au-driver-5.2.9$ dkms status
rtl8812au, 5.2.9, 4.13.0-17-lowlatency, x86_64: installed

modinfo gives two different results for rtl8812au and 8812au.ko

filename: /lib/modules/4.13.0-17-lowlatency/kernel/drivers/net/wireless/rtl8812au.ko
version: v4.3.14_13455.20150212_BTCOEX20150128-5

modinfo 8812au
filename: /lib/modules/4.13.0-17-lowlatency/updates/dkms/8812au.ko
version: v5.2.9.3_24903.20171101

in gord-rtl8812au-dkms_1.0-13_amd64.deb and in the normal dkms.conf there is
BUILT_MODULE_NAME[0]="8812au"
What did I wrong by installing . My USB-Wifi stick works . But when I used newer Kernel 4.0.13-19 things got wrong after some upgrade and dkms complained . Had to go back to my older 4.0.13-17 and used sudo dpkg -i gord-rtl8812au-dkms_1.0-13_amd64.deb to make it work again. I confess, that I tried different github modules in between, so maybe tis mixed up things.
Nevertheless I like to unterstand the two modules ( above) rtl8812au.ko and 8812au.ko and how to handle this correct.
Yours sincerely hhgoten
10829 Berlin , Germany

Arch Linux, Kernel 4.14.r1108.g87df261-1

It seems to be related with this openzfs/spl#656, which can be solved by sarangnemo/anbox@f633d64

Error when compiling with DKMS on Fedora 27
Error is as follows from dkms log

/var/lib/dkms/rtl8812au/5.2.9/build/include/osdep_service_linux.h: In function ‘_init_timer’:
/var/lib/dkms/rtl8812au/5.2.9/build/include/osdep_service_linux.h:301:8: error: ‘_timer {aka struct timer_list}’ has no member named ‘data’
ptimer->data = (unsigned long)cntx;
^~
/var/lib/dkms/rtl8812au/5.2.9/build/include/osdep_service_linux.h:302:2: error: implicit declaration of function ‘init_timer’; did you mean ‘_init_timer’? [-Werror=implicit-function-declaration]
init_timer(ptimer);

Attached is make.log

make.log

Yesterday I updated my KDE neon system and went to bed. Today, when I got home, I booted up my computer and found that the driver stopped working. I went from Kernel 4.10 (I believe) to 4.13. I've used a few drivers I found off of GitHub over a year ago for this dongle, as it's given me headaches in the past. Most didn't work except for one, and I'd just run make, make install and modprobe -a 8812au to get it to install since I can't get dkms to install anything for me. However today when I tried that it didn't work. I'm using an 802.11g dongle right now which is not ideal.

This driver, unlike the others I tried, actually compiles. No errors with sudo make, no errors with make install; everything is like it was when I installed the driver that used to work. However my system still doesn't recognize my rtl8812au chipset dongle. A few posts I've read from around the Internet all sent me to this driver.

I could always get another dongle, but I'd like to see if I can get this one working. Any suggestions on what to try?

In my environment the following message is always ongoing.
I can not connect to WiFi either.

Dec 22 11:30:13 rdbox-master kernel: [  107.178321] RTW: rtw_update_ramask => mac_id:0, networkType:0x44, mask:0x00000000ffff0000
Dec 22 11:30:13 rdbox-master kernel: [  107.178321]      ==> rssi_level:6, rate_bitmap:0x0000000000000000, shortGIrate=1
Dec 22 11:30:13 rdbox-master kernel: [  107.178321]      ==> bw:2, ignore_bw:0x1
Dec 22 11:30:15 rdbox-master kernel: [  109.248294] RTW: vht_mcs_map=fa ff, nss=2 => bitmap=00000000000fffff
Dec 22 11:30:15 rdbox-master kernel: [  109.255001] RTW: add_RATid=> mac_id:0 , raid:9, tx_ra_bitmap:0x00000000fffffff0, networkType:0x44
Dec 22 11:30:15 rdbox-master kernel: [  109.264302] RTW: rtw_update_ramask => mac_id:0, networkType:0x44, mask:0x00000000ffff0000
Dec 22 11:30:15 rdbox-master kernel: [  109.264302]      ==> rssi_level:6, rate_bitmap:0x0000000000000000, shortGIrate=1
Dec 22 11:30:15 rdbox-master kernel: [  109.264302]      ==> bw:2, ignore_bw:0x1

Hi there

I've got the new 5.2.9.3 source from a couple of days ago, and I am testing it just now, with the usual fixes for kernel 4.11, 4.11.9 and 4.12 and also the expanded device list and a small compile warning which only really came to light with a new compiler.

They have added some new features and fixes which look quite promising, but their kernel base is a few versions old, which is an annoyance. They are making changes to their old 5.2.9 rather than rebasing with a more modern kernel.

Anyways, the code, as fixed, compiles with no errors or warnings like before. If it passes muster, then I will push the changes to my repo. One thing I have noticed is the ping anomalies are a bit less, and I don't lose the signal if I go too close to the router (a weirdness with older drivers that has baffled me for some months) ... here's hoping we have a winner.

I am getting the following error(s).

root@System-name-Kali:/home/User/rtl8812au-driver-5.2.9# make
/bin/sh: 1: bc: not found
make ARCH=x86_64 CROSS_COMPILE= -C /lib/modules/5.2.0-kali2-amd64/build M=/home/User/rtl8812au-driver-5.2.9 modules
make[1]: Entering directory '/usr/src/linux-headers-5.2.0-kali2-amd64'
/bin/sh: 1: bc: not found
CC [M] /home/User/rtl8812au-driver-5.2.9/core/rtw_cmd.o
In file included from /home/User/rtl8812au-driver-5.2.9/include/osdep_service.h:47,
from /home/User/rtl8812au-driver-5.2.9/include/drv_types.h:32,
from /home/User/rtl8812au-driver-5.2.9/core/rtw_cmd.c:22:
/home/User/rtl8812au-driver-5.2.9/include/osdep_service_linux.h: In function ‘_init_timer’:
/home/User/rtl8812au-driver-5.2.9/include/osdep_service_linux.h:301:8: error: ‘_timer’ {aka ‘struct timer_list’} has no member named ‘data’
ptimer->data = (unsigned long)cntx;
^~
/home/User/rtl8812au-driver-5.2.9/include/osdep_service_linux.h:302:2: error: implicit declaration of function ‘init_timer’; did you mean ‘_init_timer’? [-Werror=implicit-function-declaration]
init_timer(ptimer);
^~~~~~~~~~
_init_timer
In file included from /home/User/rtl8812au-driver-5.2.9/include/drv_types.h:35,
from /home/User/rtl8812au-driver-5.2.9/core/rtw_cmd.c:22:
/home/User/rtl8812au-driver-5.2.9/include/wifi.h: At top level:
/home/User/rtl8812au-driver-5.2.9/include/wifi.h:1012: warning: "IEEE80211_MAX_AMPDU_BUF" redefined
#define IEEE80211_MAX_AMPDU_BUF 0x40

In file included from /home/User/rtl8812au-driver-5.2.9/include/osdep_service_linux.h:92,
from /home/User/rtl8812au-driver-5.2.9/include/osdep_service.h:47,
from /home/User/rtl8812au-driver-5.2.9/include/drv_types.h:32,
from /home/User/rtl8812au-driver-5.2.9/core/rtw_cmd.c:22:
/usr/src/linux-headers-5.2.0-kali2-common/include/linux/ieee80211.h:1441: note: this is the location of the previous definition
#define IEEE80211_MAX_AMPDU_BUF 0x100

cc1: some warnings being treated as errors
make[4]: *** [/usr/src/linux-headers-5.2.0-kali2-common/scripts/Makefile.build:284: /home/User/rtl8812au-driver-5.2.9/core/rtw_cmd.o] Error 1
make[3]: *** [/usr/src/linux-headers-5.2.0-kali2-common/Makefile:1610: module/home/User/rtl8812au-driver-5.2.9] Error 2
make[2]: *** [Makefile:179: sub-make] Error 2
make[1]: *** [Makefile:8: all] Error 2
make[1]: Leaving directory '/usr/src/linux-headers-5.2.0-kali2-amd64'
make: *** [Makefile:1799: modules] Error 2
root@System-name-Kali:/home/User/rtl8812au-driver-5.2.9# uname
Linux
root@System-name-Kali:/home/User/rtl8812au-driver-5.2.9# uname -a
Linux System-name-Kali 5.2.0-kali2-amd64 #1 SMP Debian 5.2.9-2kali1 (2019-08-22) x86_64 GNU/Linux
root@System-name-Kali:/home/User/rtl8812au-driver-5.2.9# ^C
root@System-name-Kali:/home/User/rtl8812au-driver-5.2.9#

Maybe I did something wrong but the only thing before this was installing and running wicd-gtk.
Of course, it did not find the Device.

It is running in a VM VirtualBox.
Also, the eth0 and the network cable are on the same DHCP server.

greetz Chastic Fy.

Obviously you can do what you want in your own branch.

However, there is an issue in your 'fix' for dkms.conf.

The original line

PROCS_NUM=nproc

(which cannot be easily quoted here due to the backticks with no spaces around them) is correct, in that

PROCS_NUM=$(nproc)

is a Bashism, that does not work in other shells. Both are functionally equivalent, but only one works across different shells.