And I had a typo in my commands, āimagesā instead of āimageā, so without correcting that, the download (and installation) would have failed. Did you fix that to have the kernel package downloaded correctly?
W: Possible missing firmware /lib/firmware/rtl_nic/rtl8126a-2.fw for module r8169
W: Possible missing firmware /lib/firmware/rockchip/dptx.bin for module rockchipdrm
Okay, still need one with R5C to test. On GitHub however, we have a case where the kernel with this patch fails to boot, just no idea yet why, or how it can even be (as the patch is pretty uncritical, even if it would not work).
You can ignore this. The armbian-firmware package misses a few firmware binaries for certain network drivers, while all network drivers are added to the initramfs, regardless which ones are really used by the device. So it warns if any firmware binary for any network driver is missing.
uname -a
dpkg -l | grep linux-image-current-rockchip64
Works well for me:
root@NanoPiR5S:~# uname -a
Linux NanoPiR5S 6.12.34-current-rockchip64 #1 SMP PREEMPT Thu Jun 19 13:32:38 UTC 2025 aarch64 GNU/Linux
root@NanoPiR5S:~# dpkg -l | grep linux-image-current-rockchip64
ii linux-image-current-rockchip64 25.08.0-trunk-dietpi2 arm64 Armbian Linux current kernel image 6.12.34-current-rockchip64
root@NanoPiR5S:~# ip l
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT group default qlen 1000
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
2: eth0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000
link/ether 7a:b8:44:18:54:c4 brd ff:ff:ff:ff:ff:ff
3: eth1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000
link/ether 7a:b8:44:18:54:c5 brd ff:ff:ff:ff:ff:ff
4: eth2: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000
link/ether 7a:b8:44:18:54:c7 brd ff:ff:ff:ff:ff:ff
That the MAC addresses differ in the last digit only is the indicator. U-Boot generates these, the first from some hardware identifier, the second by inverting the last bit of the last hex digit, and with our patch, also the 3rd by inverting the second last hex digit. The kernel also needed a patch to contain device tree aliases for the two 2.5 Gbit ports. Mainline Linux assigns only one for the 1 Gbit port, hence this was the only one with a static MAC.
Just in case, you did recognize that eth1 and eth2 switched place, and either accepted to have configs adjusted automatically, or done by yourself?
root@pihole:~# uname -a
Linux pihole 6.12.34-current-rockchip64 #1 SMP PREEMPT Thu Jun 19 13:32:38 UTC 2025 aarch64 GNU/Linux
root@pihole:~# dpkg -l | grep linux-image-current-rockchip64
ii linux-image-current-rockchip64 25.08.0-trunk-dietpi2 arm64 Armbian Linux current kernel image 6.12.34-current-rockchip64
root@pihole:~# ip l
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT group default qlen 1000
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP mode DEFAULT group default qlen 1000
link/ether 76:e4:b5:41:b9:7f brd ff:ff:ff:ff:ff:ff
3: eth1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000
link/ether 66:e3:ba:ae:b2:d8 brd ff:ff:ff:ff:ff:ff
4: wlan0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue state DOWN mode DEFAULT group default qlen 1000
link/ether e8:fb:1c:35:5c:39 brd ff:ff:ff:ff:ff:ff
5: tun0: <POINTOPOINT,MULTICAST,NOARP,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UNKNOWN mode DEFAULT group default qlen 500
link/none
6: wg0: <POINTOPOINT,NOARP,UP,LOWER_UP> mtu 1420 qdisc noqueue state UNKNOWN mode DEFAULT group default qlen 1000
link/none
7: docker0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP mode DEFAULT group default
link/ether 36:98:24:ea:bc:57 brd ff:ff:ff:ff:ff:ff
8: veth67e84e9@if2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue master docker0 state UP mode DEFAULT group default
link/ether 56:28:d6:44:f2:8e brd ff:ff:ff:ff:ff:ff link-netnsid 0
Ohh, it is an R5C in your case, isnāt it? Because you have 2 Ethernet interfaces only. Or is one missing unexpectedly?
cat /proc/device-tree/model
If it is an R5C, and you had static MAC addresses when using the testing kernel build, that would be pretty good news. One other tester had boot issues with that kernel, and it is the one of the R5S/R5C/R6S/R6C that I do not have here for testing yet. So I did not release the static MAC address kernel patch for the R5C yet.
As long as you do not need the board features of the R5S all good . Okay that explains it. Let me trigger a kernel build with the patch for R5C included again. Looks like in the other case, the boot issue was not related to it then. I was already wondering, because I could not find the issue, and even if it was not functional, that kind of patch should have never caused boot issues, but worst case the Ethernet ports would not have been functional or so.