denni_isl

Had to look up NXP, newer heard of it before and it is huge Multi National company originated out of Holland from Philips to be more precise https://en.wikipedia.org/wiki/NXP_Semiconductors employing 31.000 people (11.2000 engineers). Remember when I did read about The Chinese graduating around 300.000 engineers a year, about 20 years ago, which made you came to the conclusion that the world was changing rapidly.

Why are just the kernel headers under /usr/src/ but not the whole kernel source? And another question, does one need to freeze the kernel upgrade in armbian-config to keep compiled drivers?

SUCCESS AT LAST! Installing a Realtech wireless wifi dongle driver on rockpro64. Okay being a stubborn person, is sometimes beneficial. I was at last able to install driver for my speedy Chinese USB dongle with Realtech ship. lsusb gives vendor 0bda and product b812 witch is dependent upon driver rtl8812bu. Now I want to contribute to the community on how. Did fallow the instructions on https://docs.armbian.com/Developer-Guide_Build-Preparation/ compiled a new kernel and U-boot, It did produce 5 *.deb files; linux-source-default-rockchip64_5.67_all.deb, linux-headers-rockchip64_5.67_arm64.deb, linux-u-boot-rockpro64_5.67_arm64.deb, linux-image-rockchip64_5.67_arm64.deb and installed them on the rockpro64 board. It did also include /usr/src/linux-source-4.4.167-rockchip64.tar.xz which I extracted into /usr/src/linux-headers-4.4.167-rockchip64 directory and did make clean (because of previous compile error of lacking files) and then make scripts (and viola it did work!!!). Then it was github next and this driver did look promising https://github.com/cilynx/rtl88x2BU_WiFi_linux_v5.3.1_27678.20180430_COEX20180427-5959 Fallowed the instructions of dkms install there And it did work! Almost all the headache was because of the driver needing kernel source not just the headers under /usr/src/linux-headers-$(uname -r)

Did not see this threat before. Interesting. If only I had programming experience I would be willing to try to help out with migrating all those rtl--- modules on github to arm64. What i can see is that coordinated effort must be the best way to get the compiling environment right and then being able to adjust those modules. Btw. you at armbian are doing a excellent job.

Both firefly rk3399 and rockpro64 are amazing products, and working like a charm with armbian though still lacking some things. Firefly is more working out of the box when rockpro64 does not yet support for it's components like the audio jack, wifi module and isa(okay in cli) slot. Then it is the usual thing that the hardware that is ahead of the software. But what did take me with a surprise is the lack of comparability with usb wifi dongles now almost as common as coffee .

Thank you for the respond. It is actually a good point, and understood. And this https://www.armbian.com/get-involved/#communication and the Armbian documentations are actually the best source I have seen on the net for the everyday hobbyist wanting to get a better understanding of computers and now those days embedded systems. Regarding this issues with being unable to compile drivers for for arm64 boards, I have came to the conclusions that it is actually because aarch64 is just not there yet. And I am not a programmer so I have to remind my self ... hands off. arm64 is like a new kid in the block in the linux world. Android is just another story. Is there any migrating from Android to linux in the arm and arm64 support? Or is the Android source closed from the linux world? Now I am wondering how far arm64 is in it's development. Did see this of non supported software from debian site. https://wiki.debian.org/Arm64PortANFUList and it is quite a lot.

It is almost2019 and this is still preventing one to compile like rtl wifi modules. Get this error when ... sudo make scripts

Being a stubborn person, I am still trying to get a rtl8811cu for usb wifi dongle vendor 0bda and product c811 up and running. The main hurdle have been a this error messages when doing make scripts from /usr/src/linux-headers-4.4.166-rockchip64/

Fair enough. Here are some good explanations about the difference between GBT and MBR https://unix.stackexchange.com/questions/120221/gpt-or-mbr-how-do-i-know and https://www.linkedin.com/pulse/whats-difference-between-gpt-mbr-when-partitioning-drive-tiwari This might also be of interest from firefly rk3399 manual http://wiki.t-firefly.com/en/Firefly-RK3399/type-description.html

When you do gdisk -l /dev/mmcblk1 or fdisk -l you can se that the partition table that armbian is using is MBR (dos). Is any particular reason for this? When you go to the new and improved firefly rk3399 manual page you can see it is making problem when flashing through USB-C cable and upgrade_tool v. 1.24 and v. 1.34 (GBT friendly) http://wiki.t-firefly.com/en/Firefly-RK3399/flashing-notes.html .

Now when development kernel 4.19.0-rk3399 is running on both firefly rk3399 and rockpro64, On the first day I have notices 3 issues: #1. Serial debugging through minicom is not possible, there is no /dev/ttyUSB0 (ls /dev/ttyUSB*). tail -f /var/log/syslog #2. It is not possible to install rtl8812au driver for my usb wifi dongle on rockpro64. But this is a work in progress. #3. Bluetooth missing. "This probably means that there were no Bluetooth adapters detected or Bluetooth daemon was not started.

Did armbian-config through ssh and did system and nightly development build. Now the display is back with kernel 4.19.0 .

Ok, this is getting somewhere. But there is no display connection. So just connecting from firefly rk3399 to rockpro64 through minicom /dev/ttsyUSB0 and serial debug device, and ssh and trying to figure out why there is no display connection through hdmi with kernel 4.18.0. Do you have any ideas? This is what is in /etc/X11/xorg.conf.d/01-armbian-defaults.conf

Thank you for that, will try that out. It seems to be okay both on firefly rk3399 and rockpro64.

It is not so simple. It seems to me that the kernel headers and kernel source are not available for kernel 4.4.166.